Deliverable D7 Outline of a New Empirical Road Damage Experiment

Deliverable D7 Outline of a New Empirical Road Damage Experiment

CATRIN - Deliverable D7 – January 2009

SIXTH FRAMEWORK PROGRAMME

PRIORITY 1.6.2

Sustainable Surface Transport

CATRIN

Cost Allocation of TRansport INfrastructure cost

Deliverable D7

Outline of a New Empirical Road Damage Experiment

1

Version 1.0

January 2009

Authors:

Bernhard Hofko, Ronald Blab (TUV-ISTU), Robert Karlsson (VTI) with contribution from partners

Contract no.: 038422

Project Co-ordinator: VTI

Funded by the European Commission

Sixth Framework Programme

CATRIN Partner Organisations

VTI; University of Gdansk, ITS Leeds, DIW, Ecoplan, Manchester Metropolitan University, TUV

Vienna University of Technology, EIT University of Las Palmas; Swedish Maritime Administration,

University of Turku/Centre for Maritime Studies

CATRIN - Deliverable D7 – January 2009

2

CATRIN

FP6-038422

Cost Allocation of TRansport INfrastructure cost

This document should be referenced as:

Hofko B., Blab R. (TUV-ISTU), Karlsson R. (VTI), CATRIN (Cost Allocation of TRansport

INfrastructure cost), Deliverable D7 Outline of a New Empirical Road Damage Experiment.

Funded by Sixth Framework Programme. VTI, Stockholm, December 2008

Date: January 2009

Version No: 1.0 above.

PROJECT INFORMATION

Contract no: FP6 - 038422

Cost Allocation of TRansport INfrastructure cost

Website: www.catrin-eu.org

Commissioned by: Sixth Framework Programme Priority [Sustainable surface transport]

Call identifier: FP6-2005-TREN-4

Lead Partner: Statens Väg- och Transportforskningsinstitut (VTI)

Partners:

VTI; University of Gdansk, ITS Leeds, DIW, Ecoplan, Manchester Metropolitan University,

TUV Vienna University of Technology, EIT University of Las Palmas; Swedish Maritime

Administration, University of Turku/Centre for Maritime Studies

DOCUMENT CONTROL INFORMATION

Status:

Distribution:

Availability:

Filename:

Quality assurance:

Co-ordinator’s review:

Signed:

Draft/Final submitted

European Commission and Consortium Partners

Public on acceptance by EC

CATRIN D7 090115.doc

Chris Nash

Gunnar Lindberg

Date:

CATRIN - Deliverable D7 – January 2009

3

Table of Contents

Abbreviations ............................................................................................................................. 7

0 Executive Summary ........................................................................................................... 8

1 Introduction ...................................................................................................................... 12

2 Inventory of European full-scale pavement test facilities................................................ 14

2.1 Definitions.................................................................................................... 14

2.2

2.3

History of full-scale Pavement Testing........................................................ 15

European full-scale ALT Facilities.............................................................. 18

2.3.1

2.3.2

2.3.3

2.3.4

2.3.5

2.3.6

Circular Test Facilities................................................................................. 18

Linear Test Facilities.................................................................................... 18

Pulse Loading Devices................................................................................. 20

Test Tracks................................................................................................... 20

Pavement Instrumentation ........................................................................... 25

Pavement Condition Evaluation .................................................................. 25

2.3.7 Strengths and Weaknesses of European ALT facilities............................... 26

3 Factors for Pavement Deterioration, Distress and Performance ...................................... 31

3.1

3.2

Pavement Deterioration Factors................................................................... 31

Pavement Distress........................................................................................ 32

3.3 Pavement Performance ................................................................................ 34

4 Data Guidelines ................................................................................................................ 35

4.1 Data Element Types..................................................................................... 35

4.1.1

4.1.2

4.1.3

Administrative Data..................................................................................... 36

Load Application Data................................................................................. 36

Pavement Description Data.......................................................................... 38

4.1.4

4.1.5

4.1.6

4.1.7

4.2

4.3

4.3.1

Material Characterization Data .................................................................... 39

Environmental Conditions Data................................................................... 41

Pavement Response Data............................................................................. 42

Pavement Performance Data........................................................................ 43

Sampling Frequency of Data Measurements ............................................... 44

Data Storage and Retrieval (EURODEX-Database).................................... 45

Hardware...................................................................................................... 46

4.3.2 Software ....................................................................................................... 46

5 Strategic Plan for EURODEX.......................................................................................... 48

5.1 Foundations of EURODEX ......................................................................... 49

5.2

5.3

5.4

First Level - Basics ...................................................................................... 51

Second Level - Framework.......................................................................... 51

Third Level – Preparation ............................................................................ 52

5.5 Fourth Level – Launching EURODEX........................................................ 53

6 Benefits from EURODEX................................................................................................ 54

6.1

6.1.1

Politics.......................................................................................................... 54

European Commission ................................................................................. 54

6.1.2

6.1.3

6.2

6.2.1

6.2.2

6.2.3

6.3

6.4

6.5

Member States ............................................................................................. 54

Taxpayers..................................................................................................... 55

Road Infrastructure ...................................................................................... 55

Road Managers ............................................................................................ 55

Road Industry............................................................................................... 56

Road Users ................................................................................................... 56

Research Community................................................................................... 57

Benefits from improved performance model............................................... 57

Benefits from research database .................................................................. 58

CATRIN - Deliverable D7 – January 2009

4

7 Conclusions ...................................................................................................................... 60

8 References ........................................................................................................................ 63

Appendix A: Relationships between traffic and pavement maintenance costs........................ 65

Appendix B: Datasheets of European ALT facilities............................................................. 104

CATRIN - Deliverable D7 – January 2009

5

List of Tables

Table 1: European ALT facilities with circular test tracks, according to (Dawson 2002)....... 22

Table 2: European ALT facilities with linear test tracks, according to (Dawson 2002).......... 23

Table 3: European ALT facilities with pulse loading devices, according to (Dawson 2002).. 24

Table 4: Test termination criteria, according to (Dawson 2002) ............................................. 26

Table 5: Factors to be considered in the S-W analysis ............................................................ 27

Table 6: Importance of S-W for specific tests, according to (Dawson 2002).......................... 28

Table 7: Evaluation of strengths for European ALT facilities, according to (Dawson 2002) . 29

Table 8: Evaluation of weaknesses for European ALT facilities, according to (Dawson 2002)

.......................................................................................................................................... 30

Table 9: Facility administration data elements, according to (Saeed 2003) ............................ 36

Table 10: Project administration data elements, according to (Saeed 2003) ........................... 37

Table 11: Load application data elements, according to (Saeed 2003).................................... 37

Table 12: Pavement physical description data elements, according to (Saeed 2003).............. 38

Table 13: HMA characterization data elements, according to (Saeed 2003)........................... 39

Table 14: PCC characterization data elements, according to (Saeed 2003)............................. 40

Table 15: Reinforcement and load transfer device characterization data elements, according to

(Saeed 2003)..................................................................................................................... 40

Table 16: Bituminous stabilized base/subbase characterization data elements, according to

(Saeed 2003)..................................................................................................................... 40

Table 17: Cement, lime and fly ash stabilized base/subgrade characterization data elements, according to (Saeed 2003)................................................................................................ 40

Table 18: Unbound aggregate materials characterization data elements, according to (Saeed

2003)................................................................................................................................. 41

Table 19: Subgrade characterization data elements, according to (Saeed 2003) ..................... 41

Table 20: Stabilized subgrade characterization data elements, according to (Saeed 2003)..... 41

Table 21: Environmental and climate data elements, according to (Saeed 2003) ................... 42

Table 22: Pavement response data elements, according to (Saeed 2003)................................ 43

Table 23: Pavement performance data elements, according to (Saeed 2003).......................... 44

Table 24: Sampling frequency of data measurements, according to (Saeed 2003) ................. 45

Table 25: Data storage media characteristics, according to (Saeed 2003)............................... 46

List of Figures

Figure 1: Strategic plan for EURODEX – EUropean ROad Damage EXperiment ................. 10

Figure 2: Interrelationship between pavement engineering facets that collectively and individually contribute to knowledge (Hugo 1991) ......................................................... 12

Figure 3: Engineering tools to correlate simulation to reality (Golkowski 2004) ................... 15

Figure 4: The first European full-scale pavement testing device at the British NPL in 1911

(NPL 2007)....................................................................................................................... 15

Figure 5: Growth of European full-scale ALT facilities (Dawson 2002) ................................ 16

Figure 6: ALT research topics versus time (Hildebrand 2004)................................................ 16

Figure 7: Circular ALT facilities – LIRA in Romania (left) and LCPC in France (right) (pavetest.org 2008).................................................................................................................... 18

Figure 8: Linear ALT facilities – RTM in Denmark (left) and PTF at TRL in UK (right)

(pave-test.org 2008) ......................................................................................................... 19

Figure 9: CEDEX test facility in Spain – overall view (left), loading device (right) (pavetest.org 2008).................................................................................................................... 20

CATRIN - Deliverable D7 – January 2009

6

Figure 10: Vehicles generating costs for maintaining road network. Arrows indicate direction of consequence ................................................................................................................. 31

Figure 11: Data categories (Saeed 2003) ................................................................................. 36

Figure 12: Strategic plan for EURODEX ................................................................................ 50

Figure 13: Benefits from EURODEX findings (performance model) ..................................... 57

Figure 14: Benefits from EURODEX findings (research database) ........................................ 58

CATRIN - Deliverable D7 – January 2009

Abbreviations

AASH(T)O American Association of State Highway and Transportation Officials

ASCII

BASt

American Standard Code for Information Interchange

Bundesanstalt für Straßenwesen

CATRIN Cost Allocation of Transport Infrastructure Cost

COST

DBMS

Coopération européenne dans le domaine de la recherche scientifique et technique

Database Management System

EURODEX European Road Damage Experiment

FWD Falling Weight Deflectometer

GB

GPS

Gigabyte = 1024 Megabyte

Global Positioning System

LEF Load Equivalency Factor

LVDT

LTPP

Linear Variable Differential Transformer

Long Term Pavement Performance Program

NCHRP National Cooperative Highway Research Program

HGV Heavy Goods Vehicle

HMA Hot Mix Asphalt

MB Megabyte

NPL

PCC

National Physical Laboratory

Portland Cement Concrete

QC/QA

RLT

Quality Control and Quality Assurance

Real-time Load Test

SQL Structured Query Language

S-W Strength-Weakness

TB Terabyte = 1024 Gigabyte

7

CATRIN - Deliverable D7 – January 2009

8

0 Executive Summary

The EU-project CATRIN (Cost Allocation of Transport Infrastructure Cost) supports the

European Transport Policy, specifically to assist in the implementation of transport pricing for all modes of transport. Since CATRIN recognizes that cost allocation or pricing principle recommendations need to be given in a short-term and a long-term perspective, the project gives suggestions of short-term policy relevance on the one hand, meaning that CATRIN will deliver recommendations suitable for pricing implementation. On the other hand, in certain fields of research, more effort will be necessary to assess sustainable and fair transport pricing in the future. For the road sector, the 50-year-old “4 th

power rule” is still the basis for pavement design guides around the world. Although it was clear from the beginning of the US

AASHO Road Test (1958-1960) that the rule obtained from this comprehensive pavement performance test are only valid under specific conditions of the test with regard to time, place, environment and material properties, it is still used as a pavement deterioration model around the world. Factors like temperature, moisture content, vehicle speed and vehicle configuration are not taken into account by applying the 4 th

power rule, although they have an essential effect on pavement deterioration.

An efficient transport infrastructure is the lifeline for a sustainable and prospering European economy. Efficiency as well as sustainability regarding road infrastructure can only be achieved when construction and maintenance also work efficiently and sustainably. Therefore a deeper knowledge of the complex material-vehicle-environment interaction is needed to develop an improved relationship between pavement deterioration and material, environment and vehicle parameters. In order to create an enhanced rule for this interaction, it is necessary to combine comprehensive material testing with statistical, analytical and numerical methods that have been developed in the last 50 years. This can only be achieved if the research is done on a European level by EURODEX, a EUuropean ROad Damage EXperiment. Results from this experiment will lead to an improved pavement performance and deterioration model that can be used for various means. On the one hand it will be a strong tool for a fair and sustainable transport pricing on European roads, on the other hand it can be used to make pavement design and construction as well as maintenance more efficient and economic.

This report of deliverable D7, Outline of a new empirical Road Damage Experiment, gives a preliminary design and layout for EURODEX. Therefore an inventory of European ALT facilities is provided, the most important factors for pavement deterioration, distress and performance are described and guidelines for common data acquisition, data storage and retrieval are given. Furthermore a strategic plan for EURODEX has been developed within this WP. Basic information on pavement engineering and a detailed analysis of pavement deterioration factors, pavement distress, as well as current modelling of pavement performance and methods of maintenance and reconstruction are given in Appendix A.

In order to overcome the 4 th

power rule and find improved relationships between material, environment and vehicle parameters, it is necessary to gather the already existing knowledge in this field of research systematically, find out where gaps in knowledge are located and fill these gaps by a comprehensive research programme. This programme will consist of smallscale laboratory testing whenever adequate and necessary, full-scale ALT and RLT as the central point of EURODEX combined with numerical simulation and statistical methods.

Since accelerated load testing will be the main part of EURODEX, the report focuses on this test method.

CATRIN - Deliverable D7 – January 2009

9

Therefore Chapter 2 gives a detailed inventory of European ALT facilities and a Strength-

Weakness-Analysis of European test sites. The latter analysis provides a comprehensive summary, which parts of necessary research work can be covered by European facilities at this point of time and where gaps in equipment and instrumentation can be identified.

Since the 12 active European ALT facilities up to this day mainly served national research topics, there are few co-operations between different test sites to bundle their strengths and insufficient co-ordination of the various research programmes carried out by the institutes. To start a successful EURODEX it will be necessary to strengthen the co-operation between

European ALT-facilities and co-ordinate the research efficiently. Only if comparable and reliable data are produced by the test sites, improved pavement performance and deterioration models may be obtained.

Chapter 3 gives a short summary of the most important pavement deterioration, distress and

performance factors. Appendix A provides this analysis in greater detail. The chapter summarizes a review of current knowledge on the vehicle and pavement characteristics that are important to take into account when studying pavement deterioration. The factors discussed are related to traffic, climate, pavement materials, geometry and construction and show which parameters have to be taken care of by EURODEX.

Chapter 4 provides guidelines and definitions. These guidelines will ensure proper

interpretation of data and facilitate their use in different, participating laboratories.

Duplications of research efforts can be reduced and benefits from EURODEX will be enhanced. The guidelines deal with all kinds of data elements from project description data to pavement response and performance data. There are also recommendations for the collection, storage and retrieval of data in a European pavement research database. This database will be the core of EURODEX and should serve as counterpart and addendum to the US-American

Long Term Pavement Program (LTPP). It should contain all relevant data, results and findings from literature review, small-scale lab tests, full-scale pavement tests and numerical simulation.

The strategic plan for EURODEX is presented in Chapter 5. As shown in Figure 1, the way to

EURODEX consists of a foundation with basic and most important elements and four levels of planning, each relying on the level below, producing more pertinence and serving as input for the next level.

For EURODEX to be carried out in the forthcoming FP it is of grave importance to plan and prepare this comprehensive project thoroughly. An outline is drawn within this report of deliverable D7 of CATRIN; on this basis further actions have to be taken. The foundation of

EURODEX is to involve people who will participate and work for this project from an early stage on by organizing workshops and meetings. Furthermore a European committee concerning the co-operation and co-ordination of EURODEX has to be installed. The first level contains basic elements to be planned for the damage experiment. Inventories of

European test sites have to be set up – for ALT-facilities this task has already been done in

this EU-project (Chapter 2). Data from previous pavement research projects have to be

collected and performance and distress indicators isolated. The last-mentioned point has been completed by Task 4.3 and can be found in Appendix A. A short summary on pavement

deterioration, distress and performance factors is provided in Chapter 3. On the basis of the

tasks mentioned above the European database for pavement research can be developed.

Respective recommendations are given in Chapter 4. Common standards for all kind of tests

to be carried out within EURODEX shall be defined to produce comparable data and results.

A decision about pavement designs and materials to be tested within EURODEX has to be taken. Finally, the collected data and results have to be evaluated and analyzed, needs for additional equipment for European test facilities should be identified as well as gaps in knowledge concerning pavement performance and deterioration models. On this basis a

CATRIN - Deliverable D7 – January 2009

10 mission statement and detailed objectives for EURODEX may finally be defined to launch this pioneering European road infrastructure experiment.

Chapter 7 summarizes the findings of this report.

Evaluate and analyse existing data

Create European database for pavement research data & results

Identify need for additional equipment and instrumentation

Define standards for

EURODEX including

QC/QA

Decide about most relevant pavement designs and materials

Inventory of European pavement test facilities

EURODEX

Detailed planning

Define mission statement and objectives

Inventory of RLT* on a

European level

Collect data and results from previous pavement research

Isolate performance and distress indicators of pavements

Install European committee concerning the co-operation and co-ordination of EURODEX

Involve economists, researchers, owners and operators of pavement test facilities, EC and national representatives, private and state road managers by organizing meetings and workshops

*test sections on public road network

Figure 1: Strategic plan for EURODEX – EUropean ROad Damage EXperiment

From research and literature review the following insights were gained:

-

With 12 active full-scale ALT facilities combined with small-scale pavement research

- laboratories and an unknown number of test sections on the public road network (RLT),

Europe is excellently equipped for a new empirical European road damage experiment

(EURODEX).

When it comes to RLT, there are no statistics or even numbers about test sections on the

-

European level and in hardly any of the member states. It is of crucial importance to inventory these test sections with as many pieces of information about location, materials, objectives, etc. as possible to find out about the state-of-the-art in RLT. This is necessary to carry out EURODEX efficiently.

The full-scale ALT facilities mainly work for national research purposes. However, there

-

- are only few co-operations and a lack of co-ordination on a European level. An exception is the HVS-Nordic that is jointly owned and operated by Sweden and Finland. The research done in each of the facilities has high quality, but for EURODEX it is important to strengthen the co-operation and install a co-ordinating committee for full-scale pavement testing as the core of EURODEX.

Since ALT facilities work in particular for national research, every test site uses different standards when it comes to constructing pavements, pavement material testing and the collection of pavement performance and deterioration data. For example, there are 12 different test termination criteria in the 12 European ALT-facilities. For EURODEX, the participants have to agree on which standards to use for each step of the project.

Respective data guidelines are presented in this report (Chapter 4).

A Strength-Weakness-Analysis carried out in COST 347 found that the European ALT facilities when working together and bundling their strengths are excellently equipped.

CATRIN - Deliverable D7 – January 2009

11

-

-

-

There is no need to build any new facilities in the EU but to adapt one facility or the other and install additional equipment according to a detailed plan.

A lot of research effort has been made in the last 20 years in European pavement research.

The problem today is that data and results have not been collected in a common European pavement research database. To develop such a database is crucial for EURODEX to be carried out successfully. Relevant data from previous research programmes have to be collected, evaluated and analyzed. The data guidelines provided in this report will help to evaluate existing data in a uniform manner and store and analyze data from EURODEX and other pavement research projects in a common way. Therefore a QC/QA system for

EURODEX should be implemented.

The objective of EURODEX is to find improved models for pavement performance and deterioration by means of laboratory and full-scale pavement testing, as well as by means of numerical simulation. The derived models will be a reliable and improved basis for a sustainable and fair transport pricing on European roads, they will contribute to make road construction and especially maintenance more efficient on an innovative life-cycle analysis approach. It will provide important insights for the pavement research community. Furthermore a database for pavement research shall be developed. To make

EURODEX a strong tool, it is necessary to find participants from many different fields, like economists, researchers, owners and operators of testing facilities, EC representatives, private and state road managers, etc.

It is important for EURODEX to concentrate on most important performance indicators and distress functions as well as on a particular number of pavement materials and designs commonly used by the member states. This is necessary to stay within acceptable limits regarding time and financial efforts. Therefore the participants of EURODEX must find an agreement especially when it comes to pavement materials and designs. Thus one objective of EURODEX is also to find correlations and relationships between the tested

-

- materials/designs and those used by member states.

For a comprehensive research project like EURODEX it is essential to carry out the planning systematically and thoroughly. Therefore a strategic plan is presented in this

report (Chapter 5) to plan each step of the experiment.

To link laboratory small-scale testing, ALT and RLT (instrumented roads, road service measurements, etc.) in an adequate manner is of crucial importance to have data for a wide field of analysis, such as maintenance strategies, marginal cost analysis, optimization of vehicles, etc.

All stakeholders in European road infrastructure will substantially profit from EURODEX, its findings and the implementation of results. Politics, the research community as well as players in road infrastructure will benefit from the project:

EURODEX will provide a new solid basis for a European Transport Policy and a sourcerelated cost allocation on European roads. With the database, the EC will possess a strong tool for the review of future project proposals and EU research funds can be spent even more efficiently. In the same way, member states and taxpayers will be on the winning side. Results can be used by road managers to make maintenance works more efficient and economic, the road industry can improve its competitiveness in international bidding. Road users will find a fair pricing principle on roads and travel safer if the findings of EURODEX are realized.

Results and findings from EURODEX will form a strong tool to improve not only materials and pavement design, it will also set the directives in which manner tyres, suspension systems and other parts of vehicles may be optimized to reduce pavement and vehicle deterioration and therefore improve the life expectancy of roads. A detailed analysis of benefits from

EURODEX is given in Chapter 6.

CATRIN - Deliverable D7 – January 2009

12

1 Introduction

The objective of CATRIN is to support policy makers in implementing efficient pricing strategies in all modes of transport. Regarding the road sector, the most important cost factor is rehabilitation works due to pavement deterioration. The damage is notably caused by vehicle loading and axle/tyre configuration but also by climatic conditions.

To design pavements, as well as to quantify pavement deterioration the 50-year-old 4 th

power rule is still used by the majority of EU member states. This prominent rule states that pavement damage caused by vehicles is related to the 4 th

power of their axle weight. It was derived from the US AASHO Road Test in the late 1950s. Although it was clear from the beginning that the obtained rule is only valid under the specific conditions of the test with regard to time, place, environment and material properties, it has been used around the world regardless of actual conditions. Ever since the AASHO Road Test has been completed, researchers proved that a constant pavement deterioration exponent of 4 which depends only on the axle weight does not meet the requirements of the process of pavement damage.

Factors like temperature, moisture content, vehicle speed, etc. have an essential effect on deterioration. Later full-scale pavement tests resulted in deterioration exponents between 1.7 and 10.

A source-related, fair pricing principle for European roads can only be implemented, if we actually know how much each factor (e.g. temperature, axle weight, axle configuration, etc.) contributes to the damage of a certain pavement design. Therefore the most important performance indicators and distress factors of pavements have to be studied systematically on a European level. This will lead us to a deeper knowledge of the complex material-vehicleenvironment interaction and an improved model for pavement performance and deterioration can be developed.

To overcome the 4 th

power rule and to provide an improved model is one major

objective of a new empirical European road damage experiment (EURODEX). We will make use of literature study to find out about the existing knowledge, as well as small-scale laboratory testing. Full-scale pavement testing on special test sites (ALT facilities) will be the core of the experimental part and therefore the emphasis of this report is laid upon these matters. Test section on the public road network (RLT) loaded by real traffic will validate the

model. Figure 2 shows the different tools employed in EURODEX, their costs and the

achievable knowledge.

Figure 2: Interrelationship between pavement engineering facets that collectively and individually contribute to knowledge (Hugo 1991)

CATRIN - Deliverable D7 – January 2009

13

Since EURODEX will bring together European pavement research on a large scale for the first time, the project has to be planned carefully and thoroughly. Especially when it comes to full-scale pavement testing in European ALT facilities, it will be necessary to define common standards used by all participants to obtain comparable data from different test sites.

To make sure that the data produced by different participants of EURODEX are comparable,

the other major objective of EURODEX is to create a European pavement research

database to collect, evaluate and analyze research data in a common and thus comparable way. This database should be accessible to research teams, economists and the EC to gain maximum profit from the results economically, politically and scientifically.

The strategic plan for EURODEX which we provide in this report makes sure that the approach is co-ordinated and that research funds are used most efficiently.

CATRIN - Deliverable D7 – January 2009

14

2 Inventory of European full-scale pavement test facilities

COST 347 (“Improvements in Pavement Research with Accelerated Load Testing”) was initiated in 2000 to develop a European code of good practice in European ALT facilities. A detailed inventory is provided in the report of WP 1 (Dawson 2002). This chapter is built on this report and updated considering all chances in ALT facility since 2002.

2.1 Definitions

The WP1-report of COST 347 defines an accelerated load test (ALT) “as an installation

where a full or reduced-scale pavement section consisting of several layers can be tested by

means of rolling wheels or any other device that simulates traffic loading”.

Within the United States ALT is more often called accelerated pavement testing (APT).

(Metcalf 1996) states in his more detailed definition that “full-scale accelerated pavement

testing (APT) is defined as the controlled application of a prototype wheel loading, at or above the appropriate legal load limit to a prototype or actual, layered, structural pavement system to determine pavement response and performance under a controlled, accelerated

accumulation of damage in a compressed time period.

The acceleration of damage is achieved by means of increased repetitions, modified loading conditions, imposed climatic conditions (e.g. temperature and/or moisture), the use of thinner pavements with a decreased structural capacity and thus shorter design lives or a combination of these factors. Full-scale construction by conventional plant and processes is necessary so that real world conditions are modelled.”

ALT includes controllable loading and more or less controllable environmental conditions.

Data from ALT can be used to analyze the influence of single factors, e.g. increase of axle weight, on the pavement behaviour and performance. On the other hand it is more difficult to reproduce changes in pavement behaviour due to long-term material change, especially ageing and hardening of bitumen in an accelerated load test.

In general ALT devices can be divided into the following groups:

-

- linear vs. circular facilities indoor vs. outdoor facilities

- loaded by wheels vs. loaded by impulse actuators (pulse loading)

RLT is defined as the assessment of full-scale pavement structures built into the public road network and trafficked under real loading and environmental conditions. Performance related parameters, such as stresses and strains, are monitored. Data from RLT is difficult to analyze because of the more complex real-life situation and therefore the influence of single factors cannot be separated from the bulk of data. Still, RLT is the prime source to calibrate and validate performance models developed on the basis of laboratory tests and ALT.

As there is no index or European statistic about RLT, i.e. test sections on the public road network, no inventory or even numbers about these tests can be given here. It is of crucial importance to start a European inventory of RLT as soon as possible, to have numbers and objectives of at least present real-time load tests within Europe. RLT on different in-service pavements will be an important factor for EURODEX to validate results and findings from laboratory tests and ALT.

CATRIN - Deliverable D7 – January 2009

15

To determine a pavement’s response and performance the measurement of an extensive number of variables is necessary. These variables can be grouped into two main categories:

-

-

Variables related to structural response (vertical stress, horizontal strain, deflection, etc.) and climatic parameters (asphalt temperature, soil moisture, etc.)

Variables related to pavement deterioration (cracking, permanent deformation, loss of bearing capacity, etc.). More information about these variable can be found in the first report of this deliverable (“Relationship between traffic and pavement maintenance costs”)

As shown in Figure 3 ALT is the link between theoretical approaches coupled with laboratory

tests and the behaviour of pavements in practice (RLT). ALT is one of the most important factors in establishing pavement response models and in combination with RLT pavement performance models.

Figure 3: Engineering tools to correlate simulation to reality (Golkowski 2004)

2.2 History of full-scale Pavement Testing

One of the first, probably the very first full-scale pavement test was an installation in the

Public Works Department in Detroit. The circular facility called the “Paving Determinator” was constructed in 1909. The loading consisted of steel shod shoes on one end of a rotating arm to simulate a horse and an iron-rimmed wheel on the other end to simulate the cart.

Europe’s first test track was built at the British National Physical Laboratory only a few years later in 1911. This circular track was also loaded by a steel (and later rubber) wheel.

Figure 4: The first European full-scale pavement testing device at the British NPL in

1911 (NPL 2007)

The first British pavement testing programme was active until 1933 and reactivated at the

British Road Research Laboratory in 1963. The number of ALT facilities in Europe grew

steadily; many facilities were commissioned in the 1970s and 1980s as shown in Figure 5. At

that time the focus of pavement research lay on developing of new pavement structure designs and innovative materials. The tests were focused on the bearing capacity as an increasing

CATRIN - Deliverable D7 – January 2009

16 number of HGVs travelled on European highways. ALT could provide short term validation and speed up the implementation of new designs and materials. Today the attention shifts towards maintenance strategies and complex pavement performance models depending on vehicle and environmental parameters. Pavement deterioration requires more attention than bearing capacity. Sustainable road construction and maintenance strategies including

recycling gain importance. Figure 6 provides an overview of ALT research topics from 1978

to 2001.

Today 44 full-scale ALT facilities operate worldwide, 12 of them in Europe. 21 ALT facilities can be found in the United States, 3 in Japan, 2 in China, 2 in South Africa, and 1 in

Australia, Brazil, Korea and New Zealand. (NPL 2007)

Figure 5: Growth of European full-scale ALT facilities (Dawson 2002)

In 1999 the First International Conference on Accelerated Pavement Testing was held in

Reno, Nevada to improve communication between researchers working on the same field of research. 5 years later a second conference was held in Minneapolis, Minnesota. In October

2008 the third conference took its place in Spain.

Figure 6: ALT research topics versus time (Hildebrand 2004)

AASHO Road Test

Of great influence was the so-called AASHO Road Test from 1958 to 1960. It established a statistical relationship between pavement design parameters, axle load and configuration and the number of load repetitions. The road test consisted of six two-lane loops. Each lane was subjected to repeated loading by a specific vehicle type and weight. Also the pavement

CATRIN - Deliverable D7 – January 2009

17 structure within each loop was varied so that the interaction of vehicle loads and pavement structure could be investigated. The prominent “4 th

power rule” stating that pavement damage caused by vehicles is related to the 4 th

power of their axle weight is derived from the AASHO road test. The results were used to develop a pavement design guide for the United States. It was first issued in 1961 and – with updates in 1972 and 1993 – is still in use, based on results of 40-year-old test data. (Metcalf 1996)

Looking at the history of the 4 th

power rule and the conditions for the AASHO Road Test is necessary to establish a more refined model for pavement deterioration in a future

EURODEX. It is important to pin point valuable experiences and identify shortcomings in today and future applications. Some criticism against the application of the AASHO Road

Test to derive Load Equivalency Factors (LEF) today is (Mn/DOT, 1999):

-

-

In accelerated tests, environment, age and mixed traffic patterns are not considered.

Furthermore, a limited number of pavement designs were constructed on the same soil in one climate.

It did not consider vehicle characteristics (suspension, tyres, axle configuration etc.), which have also changed significantly since the test. Dynamic effects, loaded steering axles, and tridem axles other vehicle related topics not taken into account.

-

-

-

Lateral distribution was not considered and is important (for both flexible and: authors comment) rigid pavements.

Pavement design has significantly departed from the practice used at the time of the test.

The LEFs derived from AASHO Road Test have not been shown to be applicable to

- specific distress elements, such as rutting.

Pavement type and structure is needed information in a model for LEF. This is excluded in the simplified form (the 4th power law).

Although it was clear from the beginning that the 4 th

power rule is only valid under specific conditions of the test with regard to time, place, environment and material properties, it is still the basis for many pavement design guides around the world.

Ever since the AASHO Road Test had been completed, researchers proved that a constant pavement deterioration exponent of 4 which depends only on the axle weight does not meet the requirements of the process of road damage. Factors like temperature, moisture content, vehicle speed, etc. have an essential effect on pavement deterioration. Later full-scale pavement tests resulted in deterioration exponents between 1.7 and 10. (Hugo 2004)

For economic and sustainable road construction and maintenance meeting the requirements of the 21st century it is necessary to review the 4th power rule taking into account the complexity of material-vehicle-environment-interaction and develop an improved relationship between pavement deterioration, material and environment parameters. In order to create a new pavement performance and deterioration model, it is necessary to combine material testing within a European Road Damage Experiment (EURODEX) with statistical, analytical and numerical methods which have been developed in the last 40 years. The most important performance indicators and distress factors of pavements have to be studied systematically on a European level for the first time. Therefore the 12 active European full-scale ALT-facilities, once installed to serve national research topics, have to pool their resources and strengthen their co-operation. At the same time European standards for ALT from building test section to data acquisition and evaluation methods have to be defined to get comparable data.

CATRIN - Deliverable D7 – January 2009

2.3 European full-scale ALT Facilities

18

2.3.1 Circular Test Facilities

Circular test facilities use chassis with arms rotating around a central axis with loaded wheels.

Larger facilities usually operate outdoors while smaller facilities (diameter up to 16 m) can be housed in a building. Indoor facilities can be equipped with a system to control the climatic conditions. The wheels or axles on each arm are either driven each directly by an electric motor or the arms are driven by an electric or hydraulic motor situated in the central axis. The speed range is up to 100 kph with a maximum number of loads per month of 500,000. Most of these facilities follow the half-axle concept with load applied through single or dual tired wheels in a single axle arrangement. Nevertheless there is one circular test facility in France

(LCPC, Figure 7) with single, tandem or tridem axle configuration. One facility offers an

arrangement with complete axles. The loading can be applied using either a mass on each arm or by pneumatic forces. Each of the European facilities is equipped with a system for distributing the wheel path transversely over the pavement. Between four to eight different test sections with different materials are located on the same test track and can so be loaded simultaneously within one test. A direct comparison of the behaviour of different pavement types under the same environmental conditions is possible.

Table 1 describe the three European facilities with circular test tracks. The first two of them

are outdoor facilities with mean diameters between 32 to 35 m and are equipped with 3 and 4

arms respectively. The Romanian (LIRA, Figure 7) two-arm facility with a mean diameter of

15 m is installed indoors. The facilities in France and Romania are equipped with half axles; in Slovakia (KSD) the only facility with complete axle arrangement can be found. The axle loads are variable on two facilities over a wide range and are fixed at 115 kN at LIRA. The transverse distribution varies between ±300 mm and ±900 mm; the test speed is between 20 and 70 kph (normal speed) and 40 to 100 kph (maximum speed). Therefore the maximum test frequency and number of loads range from 800 to 3,400 passes per hour and 50,880 and

500,000 loads per month respectively.

The number of pavement test sections to be tested simultaneously is between 4 and 8 with the pavement width being between 2,500 and 6,000 mm.

Figure 7: Circular ALT facilities – LIRA in Romania (left) and LCPC in France (right)

(pave-test.org 2008)

2.3.2 Linear Test Facilities

Linear test facilities use constructions with a straight line arrangement. The test wheel runs back and forth over a linear test track. Parts of the test track have to be reserved for acceleration and deceleration of the load wheel and cannot be used for evaluation. Most linear

CATRIN - Deliverable D7 – January 2009

19 facilities are placed indoors due to their reduced dimensions compared with circular test facilities.

CEDEX (Spain, Figure 9) represents a hybrid type, a combination of a linear and circular test facility. The track consists of two straight stretches joined by two circular curves. Six pavement sections can be loaded and therefore compared simultaneously by two separated axles. CEDEX represents the only linear facility where more than one pavement can be tested at the same time.

The direction of th e loaded wheels is fixed at CEDEX. They can go only one-way, whereas

the RTM (Denmark, Figure 8) and LAVOC (Switzerland) facilities allow only two-way

loading. All other facilities can be loaded one- and two-way. The speed range for linear facilities is generally lower than for circular test facilities. The maximum speed is 25 kph with

CEDEX as an exception. The maximum speed for this combined facility is 60 kph. Therefore the practical number of loads per month varies between 50,000 and 600,000.

The loading device for most of the facilities consists of a half-axle with sing le or dual tired wheels in single axle arrangement. LAVOC represents a complete-axle concept. The loading is achieved by gravity or hydraulic and pneumatic forces respectively. All facilities allow a transverse distribution of the wheel paths.

Four of the seven linear ALT facilities are fixed indoors; the HVS-Nordic is a mobile facility which can be moved as a semi-trailer over long distances and parts of the public road network can be tested directly.

Figure 8: Linear ALT facilities – RTM in Denmark (left) and PTF at TRL in UK (right)

(pave-test.org 2008)

The length of the test tracks ranges between 4 and 27 m, the effective length with constant speed within the section varies between 2 and 12 m. Again CEDEX is an exception with

288 m testing length. The transverse distribution is variable on all test facilities ranging from

±200 to ±600 mm, as is the test speed (normal speed 8 to 20 kph, maximum speed 12 to

25 kph). CEDEX has a speed limit of 60 kph.

Air temperature is monitored on three and c ontrolled on four ALT facilities whereas the pavement temperature can be fully controlled on three and indirectly controlled by air temperature also on three test facilities. A freeze-thaw cycle control is possible at LAVOC and at the HVS-Nordic test site in Finland. RTM allows monitoring of freeze-thaw cycles.

CEDEX has an equipment to control rainfall.

Table 2 gives detailed information about Euro pean full-scale linear ALT facilities.

CATRIN - Deliverable D7 – January 2009

2.3.3 Pulse Loading Devices

20

Pulse loading devices operate with a hydraulic pulsed load equipment. The sinusoidal load generated by a hydraulic jack is applied by a circular plate to the pavement surface. The loading device can be equipped with a longitudinal displacement device to simulate moving traffic. The great advantage of pulse loading devices is the acceleration effect. 1 to 12 million load repetitions per month are possible. Also facilities of this type require very little space, can be placed indoors and the initial costs as well as the annual fixed costs are relatively small. The problem is that the degree of simulation of real traffic conditions can never be as high as for facilities using rolling wheels.

Table 3 shows the two European facilities, both of them located in Germany. At the BASt test

track with a length of 38 m and a width of 7.5 m up to three hydraulic pulse loading devices can be used at the same time. The load is applied by a circular plate with a diameter of

300 mm, the load range is between 0 and 90 kN. There is an automatic transverse distribution of ±75 mm. A haversine shaped load pulse (25 ms) and a frequency of 145 pulses per minute allow about 1 million load repetitions per month. The temperature can be controlled by air temperature, freeze-thaw cycles can be controlled as well as the water table level.

At the TU Dresden, a 5 m long and 2.5 m wide test track is located with a single point load device. The load can be applied by a circular plate with a variable diameter of 300 to 600 mm, the load ranges from 6 to 66.4 kN and the loading direction is variable without the possibility of transverse distribution. A sinusoidal load with a frequency of 5 Hz can be applied. This allows 12 million load repetitions per month with controlled air temperature.

Figure 9: CEDEX test facility in Spain – overall view (left), loading device (right) (pavetest.org 2008)

2.3.4 Test Tracks

Test tracks are full scale roads on a test area which is not part of the public road network. The pavement construction is loaded by normal trucks with a certain axle weight for a certain number of load repetitions. Pavement performance, climatic conditions and other boundary conditions are monitored. Test tracks provide a very realistic view on how pavement structures react to real trafficking but the acceleration of loading is quite low. Test tracks are not to be misinterpreted as test sections on the public road network. The latter one is an alternative expression for RLT.

CATRIN - Deliverable D7 – January 2009

21

Appendix B includes detailed datasheets for the European ALT facilities mentioned above. It was taken from the final report of Work package 1 of COST 347.

CATRIN - Deliverable D 7 – January 2009

Table 1: European ALT facilities with circular test tracks, according to (Dawson 2002)

Test track

General

Placement

Mean diameter

[m]

Loading characteristics

Nr. of arms

Axle/ wheel config.

Range of axle load

[kN]

Speed characteristics Pavement charact.

Transverse distribution*

[mm]

Test speed

(normal/m ax/min)

[km/h]

Test frequency

(max)

[passes/h/se ction]

Practical nr. of loads per month

[loads/ month/ section]

Nr. of sections

Pavement width

[mm]

Pavement thickness

[mm]

Environ. characteristics

22

Manège de fatigue -

LCPC Nantes, France fixed outside/ mobile (3 test tracks)

35 4

LIRA - Gh. Asachi,

Technical Univeristy

Iasi, Romania

CTT Circular Test

Track - VUIS-CESTY

Ltd. Bratislava,

Slovakia fixed outside fixed outside

15

32

*) distance between center of wheel paths

2

3

40 - 135 ± 500

45-70/

100/3.6

3,400 500,000 4 6,000 900

115 ± 300 20/40/0.5

1,700 50,880 8 3,000 2,000

85 - 130 ± 900 30/50/0.1

1,500 170,000 6 6,000 2,000

----or

-----

---------

… monitored

… partially controlled

… controlled

CATRIN - Deliverable D 7 – January 2009

Table 2: European ALT facilities with linear test tracks, according to (Dawson 2002)

23

CATRIN - Deliverable D 7 – January 2009

Table 3: European ALT facilities with pulse loading devices, according to (Dawson 2002)

24

CATRIN - Deliverable D7 – January 2009

25

2.3.5 Pavement Instrumentation

A survey among the European ALT facilities conducted within COST 347 found that the most common instrument in use is the thermocouple to measure temperature in different pavement layers. Also the strain-gauged ‘H’-bar to find out about vertical and horizontal strain in layers and the diaphragm pressure cells to measure vertical stress in non-asphaltic layers are used quite commonly. A fourth instrument to be used in many ALT facilities is LVDTs for vertical and horizontal strains in non-asphaltic layers. All other instruments are only used by a few

ALT facilities or even a single user.

Another result is that temperature and horizontal strain in bituminous bound layers as well as vertical strain in the subgrade and unbound layers are found to be the variables measured by most of the facilities. Also surface and sub-surface deflection, temperature in subgrade and unbound layers and vertical stress in subgrade is monitored by more than half of the ALT facilities. Interestingly enough, moisture and stresses in asphaltic layers are not measured by any of the facilities, although moisture sensors are commonly available and economic to purchase.

Many respondents reported reading fewer variables than they would like. Reasons for this are because the variables are not believed to be important for the application and that the data was believed to be too difficult to interpret and to process, respectively. Eight of the ALT facilities indicated that costs for an instrument were too high for it to be installed.

From these few statements it can be seen that there is no common standard or code of practice for instrumentation and data acquisition.

Also the way how instrumentation is installed differs from facility to facility. Some test sites place an instrument on top of a layer and bury it by the next layer of the construction while others cut a hole into the layer of the completed pavement and lower the instrument into the hole before backfilling the hole by hand.

Most research centres place instruments sparsely along the centreline of the wheel path. On the other hand some ALT facilities also install instruments at lateral positions. There is no standard for installation of instruments.

When it comes to data collection, the schedules are often based on the number of load application, being more frequent during early stage of load application. No facility uses a performance-related approach which is interesting as it would be important for operators to have instrumentation to tell him when the performance limit of a road section is reached.

Data collection is performed in different ways depending on the operator of the facility. Most test sites use electronic methods with software-driven control. Nevertheless, six respondents use manual data collection techniques occasionally. Half of the ALT facilities present and store data completely, while the other half removes outliers by one method or the other. This shows that no common practice or even standard for data collection can be found within

European ALT facilities.

2.3.6 Pavement Condition Evaluation

During construction of test structures most of the facilities monitor layer thickness, density of asphalt and granular layers. Also load tests are performed commonly on each layer during construction. Before and after the tests many respondents of the survey mentioned above, use

FWD to measure surface deflection. Still one facility uses Benkelman Beam and another one performs static plate loading tests.

During an ALT all facilities monitor permanent deformation and 11 facilities measure surface deflection. Cracking of the surface is measured by 10 test sites. Some ALT facilities take

CATRIN - Deliverable D7 – January 2009

26 longitudinal profiles during the test; some perform an investigation by exhumation after the test.

When it comes to the criterion used for test termination, it can be shown how much ALT

differ from facility to facility. Table 4 shows the termination criteria for each ALT test site.

There is no common factor, not even when it comes to the termination criterion. Even if the same criterion is used, the termination values vary, e.g. from 18 to 50 mm permanent deformation with many exceptions stated in footnotes. The same can be said about surface cracking and the combined termination criteria. For EURODEX to be successful, common

guidelines have to be created to prevent situations like the example in Table 4 reveals.

Table 4: Test termination criteria, according to (Dawson 2002)

1) 100,000 passes to 200,000 passes

2) Termination depends on the testing requirements of the pavement under investigation.

4) At least some visible cracks

5) The shape of the permanent deformation development. Subjective interpretation, but aimed to prevent uncontrollable pavement failure. Another subjective indicator is the development of longitudinal cracking over the heaves of the rutting profile.

6) 80 to 100% of the section cracked and at least 2 cm rut depth everywhere. In the contract with partners a number of loading is specified, if one of the sections is greatly damaged before the end, it is repaired to go on the others sections. If there is no damage to the pavement after the number of loading specify in agree with the partners, the number of loadings increase 10% more with a load higher to have significant damages. Otherwise the experiment stops when the machine cannot move on the damaged pavement.

7) For rutting, mechanical conditions decide for the end of the test: contact of the asphalt layer with the loading machine or the tires. For fatigue, we make an evaluation of the life duration expressed in ESAL

8) Increasing of bearing capacity.

9) Degradation index. Skid resistance and bearing capacity, Romanian Standard

10) Determined number of load cycles, for instance 1x10e6, 2,5x10e6, 5x10e6

It is also interesting to take a look at the methods, how pavement condition variables are measured. 11 out of 12 facilities use transverse profiles to get permanent deformation. Surface cracking is measured by inspection as “cracks in the loaded area”. 4 facilities don’t ever monitor or obtain any cracking. It is also common to use longitudinal profiles in the centre line of the loading as well as FWD for surface deflection.

The number of measurements varies between 3 to 40 points per test section at linear test tracks and 5 to 26 per test section for circular test tracks. Some data is recorded and stored on paper; some data is only recorded on paper but stored electronically or recorded and stored electronically. Only one facility has a full automatic data record and storage system.

2.3.7 Strengths and Weaknesses of European ALT facilities

COST 347 also prepared a Strength-Weakness-Analysis of the existing European ALT facilities. Therefore different significant factors were identified to asses the goodness of each

facility (Table 5).

CATRIN - Deliverable D7 – January 2009

27

As a next step each Strength-Weakness (S-W) had to be classified considering the relative importance of each factor for ALT facilities in general as well as for specific tests. The specific tests were divided into 7 different types of tests with different research objectives:

- material testing

-

- performance testing pavement design

-

-

-

- pavement maintenance wheel load effects test validation environmental effects

Table 5: Factors to be considered in the S-W analysis

Category Factor S-W

Magnitude 1,2

Speed 3,4,5

LOADING

Device 6,7

Axle configuration 8,9

Tracking 10

Wandering 11

Suspension 12

PAVEMENT

Propulsion 13

Loading direction 14

Dimensions 15,16

Subgrade 17

Existing pavements 18

CONSTRUCTION Procedure 19

Pavement temperature 20

Water table 21

ENVIRONMENT

Freeze-thaw cycles 22

Rainfall 23

Damage acceleration 24

OUTPUT

Number of sections 25

For ALT in general 6 out of 25 S-W were considered to be “very important” whereas the others are seen to be “important”. The 6 main important factors are:

-

Load magnitude is representative of real traffic load (S-W 1).

-

-

-

Load is applied by rolling wheels (S-W 6).

Wheels are full-scale (S-W 7).

Speed is representative of real traffic (S-W 3).

-

-

Width and thickness of the test sections are full-scale (S-W 15).

Length is representative (S-W 16).

Also the importance of S-W for specific tests was evaluated. Therefore four members of

COST 347 discussed about their opinion about the importance of different S-W for the

various test types. Table 6 shows the result of this discussion. In the table an “x” means, that 3

of 4 and 4 of 4 respectively agreed in this point, whereas a “·” means that only 1 or 2 out of 4

made the choice. This system also applies for Table 7 and Table 8.

As shown in Table 6 great importance for nearly every test type is given S-W 1, which states,

that the load magnitude should be representative of real traffic. Also highly important seem to be Strength-Weakness-factors connected to speed. Speed should be representative of real traffic (S-W 3), low speeds should be applicable for certain tests (S-W 4), e.g. rutting tests

CATRIN - Deliverable D7 – January 2009

28 and the speed should be variable (S-W 5). For 4 out of 7 test types it is crucial that the load is applied by rolling wheel (S-W 6) and not by pulse actuators. An automatic transverse distribution is considered to be fundamental for another 4 test types (S-W 11).

Table 6: Importance of S-W for specific tests, according to (Dawson 2002)

Interestingly enough certain test types have higher demands on the equipment of an ALT facility than others. Tests to develop performance models show high importance in 20 out of

25 S-Ws. Test validation or tests to find out about environmental effects on the other hand do not require as high standards.

As EURODEX will deal with validating existing and creating new performance models, as well as material testing in general, wheel and environmental effects, it is clear that Europe’s

ALT facilities need to provide high standards in every S-W factor. It does not mean that every single test site has to be adapted to be a high-tech laboratory but some additional equipment will be necessary to cover all essential parts of EURODEX.

Table 7 and Table 8 show the evaluation of S-W of European ALT facilities. Many strength-

factors are shared by nearly all facilities, like the load application by rolling wheels (S-W 6) or a representative load magnitude (S-W 1). But there are also some factors that are hardly covered by European facilities. This is especially true S-W 12 and 13. S-W 12 describes the use of several different suspensions systems. None of the European test sites has the possibility to change the suspension system. S-W 13 is about the use of different propulsion system, like wheels being towed or have an engine for themselves. The propulsion system is

fixed for each facility. As it does not seem to be of great importance for many tests (Table 6),

the lack of this is not crucial but still noteworthy.

Only one facility – LCPC, France – can use different axle types. All the other test sites use single axle configuration with no possibility to change this. LCPC’s circular test track can also be loaded by tandem and even tridem axles. There is a need for action as a shortage in this factor limits the possibilities of EURODEX.

CATRIN - Deliverable D7 – January 2009

29

Other mentionable shortcomings in the European ALT community are the possibility to control rainfall, to test existing pavements and to control or monitor freeze-thaw cycles. It has to be decided when a detailed layout for EURODEX is available whether these shortcomings are crucial or create a bottleneck-situation which would lower the quality of EURDEX. If so, one facility or the other has to be equipped with additional instruments or devices to overcome this deficiency.

Table 7: Evaluation of strengths for European ALT facilities, according to (Dawson

2002)

By taking a look at Table 8 a similar picture as the evaluation of the strength-factors can be

discovered by evaluating weaknesses of European ALT facilities. Especially the problem with fixed axle configuration, control of rainfall, freeze-thaw cycles and testing existing pavements can be mentioned as weaknesses of European test sites.

Even more important is the fact that non of the “very important” S-W factors – highlighted in

grey in Table 7 and Table 8 – is a problem in European ALT. 5 of 6 grave factors are covered

by more than 9 facilities. Only when it comes to S-W 3 about representative testing speed, only the 3 circular facilities and CEDEX achieve this goal. It is a general weakness of linear facilities as top speeds are usually below 25 kph.

Of course a strict S-W analysis can never constitute a rating of the quality of an ALT facility.

Elements like experimental design, instrumentation, condition testing, workmanship, experience, etc. have to be considered as well. But the factors mentioned in the analysis are basic requirements. The highest level of experience or quality in test section construction cannot overcome the shortcomings if basic requirements are missing.

Summing up the analysis it is obvious that loading, wheels, speed and dimensions of the test sections have to be representative of real traffic. Linear test tracks usually fulfil all main strengths except for the realistic speed. Circular test tracks fulfil all main strengths, whereas pulse loading devices lack of realistic loading, since rolling wheels are substituted by pulse loading. This limits the potential of pulse loading devices considerably.

The lack of flexibility to use different axle configurations as well as different suspension and propulsion systems are present weaknesses. Also, for northern member states pavement deterioration due to studded tyres is a significant problem, not only on HGVs but also when it

CATRIN - Deliverable D7 – January 2009

30 comes to passenger cars. To find out more about the effect of studded tyres even with low axle weights it will be necessary to adjust one facility or the other to be able to test axle weights on passenger car-level.

Table 8: Evaluation of weaknesses for European ALT facilities, according to (Dawson

2002)

An appropriate environmental control has not been achieved for any type of facility.

Pavement temperature is commonly monitored, sometimes controlled indirectly by air temperature, rarely directly controlled, as are freeze-thaw cycles and rainfall. If environmental influence on pavement deterioration should be considered within EURODEX there will have to be adaption to existing ALT facilities to manage this part of the experiment.

CATRIN - Deliverable D7 – January 2009

31

3 Factors for Pavement Deterioration, Distress and Performance

A review of current best knowledge of mechanisms that drive road deterioration is undertaken as a part of Deliverable 7 and is given in Appendix A. This chapter aims at summarizing the review of current knowledge on the vehicle and pavement characteristics that are important to take into account when studying (costs of) pavement deterioration. The review derive costs

from a chain of consequences, cf. Figure 10 below, starting with factors that lead to

deterioration such as traffic, climate and the pavement and subgrade itself. Then, models for pavement deterioration and subsequent needs for maintenance are depicted. Finally, the costs of different maintenance activities are investigated. This chain of consequences leading to costs for maintaining our road infrastructure is the key to understanding the mechanisms behind road user marginal costs.

Vehicles

Traffic

Design and Construction

Climate Pavement and subgrade

Pavement deterioration

Pavement performance

Maintenance activities

Road owner costs

Road user costs and benefits

Society

Figure 10: Vehicles generating costs for maintaining road network. Arrows indicate direction of consequence

The review also adds support to why EURODEX is necessary and why this project needs to be coordinated on a European level to gain potential of testing performed in different countries. Technical aspects of traffic, pavement deterioration and ALT testing to consider in analysis of costs for maintaining road network is given.

3.1 Pavement Deterioration Factors

Initially, factors responsible for deterioration of pavements are discussed. These factors are related to traffic, climate, pavement materials, geometry and construction. Furthermore, these factors often interact. Avoiding or mitigating future deterioration is the key to pavement design, and failure in this aspect will lead to increased maintenance costs.

The intensity of traffic is obviously very important to deterioration of pavements. Speed and changes in speed by braking and acceleration as well as turning (transversal acceleration) are also crucial factors that influence (particularly) the bituminous bound layers. Furthermore, the weight of vehicles, axle loads and widths, suspensions of vehicles, as well as tyre positions and properties will also affect the deterioration relationship.

Traffic loading is quite complex to describe, but shows interesting possibilities and pitfalls in the development of more pavement friendly vehicles and more wear resistant pavements.

CATRIN - Deliverable D7 – January 2009

32

Correct design and selection of maintenance treatments need to estimate the influence of the miscellaneous vehicle characteristics in fleet and overall traffic intensity.

The climate sets the conditions for pavement design. Temperatures ranging from low to high are important pavement deterioration factors for both rigid and flexible pavements. Asphalt concrete at low temperatures becomes stiff and less able to relax stresses, with increased risks of cracking. At high temperatures, permanent deformation is more likely, both as a result of compaction and changes in geometry. The load spreading capacity of the asphalt concrete layer is also reduced due to reduced stiffness. The rate of ageing increases exponentially with temperature.

To set exact figures of temperatures is not possible since these temperatures will vary across

Europe. The reason is that bituminous binders with different properties are used to account for the variation in temperatures, climate and traffic loads.

Water and moisture will strongly influence pavement deterioration in a number of ways by stone loss on wearing course, weakening of bound and unbound pavement layers or bonding between layers, and loss of bearing capacity due to poor drainage or during flooding.

Freezing and thawing is of course related to the previous factors, water and temperature, but is described separately due to the combined origin and specific effects. Cycling between freezing and thawing may for many mechanisms cause greater damage. Deterioration due to freezing and thawing is, in some regions, more severe than traffic loading, and selected as limiting design criteria.

The climate variation across Europe is by its own or combined with traffic deteriorating pavements. Factors described above need to be carefully selected, controlled during testing and described in documentation in order to be able to interpret results for EURODEX.

Parameters such as moisture content and temperature strongly influence the performance of pavement materials.

3.2 Pavement Distress

Pavements around Europe are built with a variety of materials and respond quite differently to traffic loading. Bituminous as well as cement bound materials are subject to ageing, which alter their properties and set a time limit for use. Stresses observed in different layers are given below.

The levelled foundation for the pavement is called subgrade and consists of on site soil. The performance of the subgrade can also be improved by drainage and protection from climate actions. The review shows that distress from several separate loads at the pavement surface will overlap further down in the pavement or subgrade. Narrow, heavy and dynamic loads at the surface, increase the risk of deterioration in the subgrade. The deformation behaviour of the subgrade strongly influences the mode of failure of the pavement as a whole and the subsequent needs of maintenance.

Unbound materials in pavements are normally crushed aggregate or uncrushed stone material. Unfortunate combinations of high loads or improper load configurations, weak or damaged bound layers and subgrade, and conditions in the unbound material itself may cause severe damage to a pavement. Overloading at the wrong time in the wrong place may lead to damage far beyond any 4th power law.

For performance reasons, bituminous bound materials for pavement construction are usually divided into

-

Wearing course – the surface layer with the purpose of withstanding wear and climatic variation, as well as creating a surface with good characteristics regarding friction, drainage, noise and visual properties

CATRIN - Deliverable D7 – January 2009

33

-

Binder course – an intermediate layer to distribute the concentrated wheel load from the surface to a larger stressed area

-

Base course – bottom asphalt concrete layer which can withstand both deformations and numerous cyclic loads (fatigue).

Asphalt concrete is used in layers with different performance conditions and distress types.

When comparing field observations, ALT testing and laboratory testing, it is important to keep in mind effects of time, for example due to bitumen ageing (long term), healing (mid term) and viscoelasticity (short term). Asphalt concrete deterioration can to some extent be modelled but may be strongly influenced by the structure as a whole. However, more accurate predictions are not possible yet but research is ongoing regarding distress mechanisms such as permanent deformations and fatigue.

Special products with unique features, such as noise reducing porous asphalt concrete, are usually more expensive and often more sensitive, leading to a substantial influence on maintenance costs and associated reasons of deteriorations.

In combination with heavy traffic, some critical conditions for flexible pavements

(bituminous or unbound materials) are:

-

Cold (stiff) surface and wet (weak) subgrade. Often the case during thawing. Great risk

-

- for cracking of asphalt concrete.

Hot (soft) asphalt concrete and wet (weak) subgrade. Often the case during rainy summers on poorly drained pavements. Great risks for excess rutting.

Slow traffic on hot (soft) asphalt concrete on pavements designed for low or medium volume traffic.

Cement concrete may also be subject to fatigue and ageing. Fatigue due to temperature and traffic loads is used as a pavement design criteria. Ageing and durability is an important topic since rigid pavements are usually demanded a long service life to be cost efficient compared to flexible pavement with much shorter maintenance intervals.

Interaction between factors related to traffic, climate and pavement structure is needed to explain some of the pavement deterioration. From the point of view of carrying out

EURODEX, it is of great importance to handle covariance.

A critical condition for rigid pavements containing cement bound materials is rapidly heated surface (creating a large temperature difference to lower layers) and heavy traffic. Slabs are poorly supported in the centre and experience great stresses.

Distress mechanisms are not always clearly associated with traffic. This means that they do not contribute to marginal costs for traffic in the manner that is suggested by the 4th power law. However, in some situations the 4th power law is certainly applicable and even not enough sensitive to the detrimental effects of heavy vehicles, for example on pavements with poor bearing capacity.

Stress conditions and stress history present in the field and in ALT testing is very difficult to reproduce exactly in laboratory. Models are therefore simplified. However, these simplified models clearly identify severe conditions that will lead to permanent deformations and that further loading will result in rapidly increasing deformations (which in turn will show as ruts).

For EURODEX purposes, it is of particular importance that material properties and actual stresses and strains are accurately selected, controlled and documented.

The marginal cost for a given vehicle will differ substantially on a strong pavement compared to a thin pavement designed for a low volume road. Another consequence is that increased traffic, traffic load, poor maintenance etc. (i.e. either increasing load or decreasing bearing capacity) may shift the rate of deterioration to become more severe. A pavement designed to be adequately strong in one decade may be considered weak in the next due to changes in traffic, unexpected damage or poor maintenance.

CATRIN - Deliverable D7 – January 2009

3.3 Pavement Performance

34

To achieve pavement performance is the goal of pavement construction and maintenance.

Examples of pavement performance related factors are

-

Traffic safety in terms of skid resistance, low risk of aqua planning, predictability and acceptable levels of evenness, and visibility;

-

Transport economy, efficiency and quality in terms of evenness, ruts, local faults, stone loss and aesthetics.

The efforts needed to maintain acceptable levels of service differ substantially depending on the amount of traffic (related to preferred standard and rate of deterioration), type of road

(standard and width), type of pavement, climate and geographical location etc. The level of pavement performance should be chosen based on road user and road owner costs, and road user benefits.

Pavement performance is measured on a regular basis by various equipment from which a number of performance related parameters can be generated, so called performance indicators.

Consequently, a vast amount of information exists regarding pavement surface characteristics present in the field, which could be used extensively to relate fundamental technical issues to performance for road users. The next step is to relate costs for constructing and maintaining a road with its level of performance, thus relating costs for maintenance and construction with the actual technical decisions made regarding design and long term maintenance strategies.

This information can be used in decision making based on Life Cycle Cost (LCC) Analysis.

Statistical models of pavement performance are often used to plan activities, select maintenance treatment and estimate future expenditures. Statistical models of pavement performance relate observable features related to a road with performance indicators such as rutting and extent of cracking. These models have a few things in common. They relate to traffic, structural strength of pavement (bearing capacity) and involve a number of calibration factors. Statistical models are easy to use, but lack the ability to handle new practices or changes in conditions. That is why more fundamentally based design models are needed, such as mechanistic – empiric design models.

The majority of European countries have adopted mechanical-empirical based design procedures. These procedures directly calculate pavement life based on a performance model, which relates to stresses or strains induced by wheel loads in a response model in which material properties are given as a function of parameters such as climate. Empirical data from in-service roads is used to calibrate mechanically determined design criteria in the performance model. Finally, the calculated performance is compared to the desired need for performance.

Performance models are the key to successful pavement design. Since the relationships between factors leading to road deterioration and development of pavement performance are very complex, the simplifications and empirical calibrations made in traditional pavement design are radical and inclusive. A complete mechanical analysis is not reasonable to expect and a great deal of empirics is needed. ALT testing is in this context a very valuable tool to bridge between laboratory and field observations.

The criticism against the AASHO 4th power law shows the need to find an improved foundation for assessing the damaging power of vehicles. The review of demands of pavement performance and means of assessing the road network reveals a great source of information not yet reached its potential for applications in this field. Together with the statistical and mechanistic-empiric models for pavement performance and design, a foundation is laid for a new system for assessing the damaging power of vehicles. However, there are large gaps in the knowledge structures. The majority of gaps, especially those relating field and laboratory observations, can be bridged by a well organised EURODEX.

CATRIN - Deliverable D7 – January 2009

35

4 Data Guidelines

This chapter is along the lines of the US NCHRP Report 512 (“Accelerated Pavement

Testing: Data Guidelines”, Saeed 2003) which was published in 2003. The findings of the report are used for this report considering the special situation in European ALT facilities.

For a European Road Damage Experiment to be carried out in a successful way scientifically as well as economically, it is of crucial importance to provide definitions and guidelines.

These guidelines will ensure proper interpretation of the data and facilitate their use by different pavement research facilities and the European scientific community. Data from different laboratories will be compatible and make sure that funds for EURODEX are spent efficiently. Duplication of research efforts can be reduced and therefore benefits pavement research will be enhanced.

The guidelines presented in this report are valid for every full-scale pavement test facility as it deals with all kinds of data elements associated with pavement testing. There are recommendation for the collection, storage and retrieval of these elements. The guidelines can be used for all full-scale facilities where full pavement structures are loaded by wheel loads either of machines or vehicles in a test facility or in-service pavements.

The data elements were categorized as follows:

-

Administrative: administrative details of a particular test facility and a particular experiment conducted at the facility.

-

-

-

-

Load application: wheel loadings applied to a test pavement and the characteristics of the applied loads.

Pavement description: information on pavement type, construction and geometric details

Material characterization: information about material types, composition, stiffness, strength and test methods.

Environmental conditions: information about environmental conditions above and within

- a pavement structure

Pavement response: deflections, stresses and strains measured at the pavement surface or within the structure when subjected to a given load or to changes in temperature and moisture.

-

Pavement performance: information on different types of pavement surface distress, pavement smoothness and longitudinal and transverse profiles.

Besides information about acquisition of the above mentioned data elements, there are also recommendations for state-of-the-art data storage, and a database as well as data collection frequencies are given.

If these guidelines are implemented in the practice of European pavement test facilities, present problems with data incompatibility can be overcome. Today data are generally facility specific or even project specific. Differences in definitions of test parameters and the format in which data are collected, recorded and stored make it difficult for others to use and interpret the data. This often leads to duplication of efforts.

4.1 Data Element Types

The following chapters provide guidelines for data collection and storage requirements.

Figure 11 shows the seven main data categories with examples of related sub elements.

CATRIN - Deliverable D7 – January 2009

36

Organizing the data elements into numerous data sets reduces the need for repeated data recording. Information need to be entered only once. Furthermore the use of multiple data sets makes it easy to retrieve a specific piece of information from a database.

Figure 11: Data categories (Saeed 2003)

4.1.1 Administrative Data

Administrative data can be facility related, i.e. unique to the test site or project related, i.e. unique to each project at the facility. These data contain the staff running the test site, the location of the facility, ownership and who operates the facility, as well as staff responsibilities and how to contact the personnel.

It is important, that only one set of facility data exists at each site and it should be maintained by the facility administrator. The data is to be provided to researchers for inclusion in the

researcher’s file. Table 9 gives information about data elements that should be recorded in the

facility administration file.

Table 9: Facility administration data elements, according to (Saeed 2003)

Data Element Definition

Name

Location

Objectives Set of goals to be achieved

Owner agency Agency who owns the facility

Operator

Unique name of facility

Physical location of facility; address of fixed facilities or the home office of a mobile facility

Key personnel

Agency who operates the facility

Name and phone number, fax number, address, email address, function within the facility and other relevant contact information for key personnel at the facility, including the director, administrator, owner point of contact and operator point of contact

Project administration data include the personnel running each project. Project information should be gathered at the beginning of the project and updated if there is a need to during the

project. The principal investigator is responsible for maintaining the information. Table 10

defines the data elements for project-level administration data.

4.1.2 Load Application Data

Load application data elements characterize the loads applied to a test pavement. Loading conditions must be recorded for each test. It is of crucial importance that the loads are measured, characterized and counted properly since ALT is a method of applying large numbers and magnitudes of load repetitions to a pavement in a short period of time to

CATRIN - Deliverable D7 – January 2009

37 accelerate its deterioration. The elements must describe how the loads are applied, the magnitude and the load patterns.

Load application data have to be collected for each test during the course of a project since they are specific to each test. The measurements should be reported to at least 1 % accuracy.

Some of the data elements such as wheel position and spacing or contact area shape are best described using drawings.

Table 11 gives detailed information about Load application data elements.

Table 10: Project administration data elements, according to (Saeed 2003)

Data Element Definition

Name and identification

Objectives

Status

Project name and identification information

Research goals and scope of work or the abstract of a completed project

Completeness of a project: proposed, % complete, or complete; specify if percent complete is in terms of time, effort or money

If the project is at a fixed facility, the name and address of the facility; if the project is mobile,

Location

Time frame the area where testing occurs, identified by highway number, kilometre marker or GPS coordinates

Anticipated or actual start and finish dates; also any planned or unplanned breaks in the testing schedule

Funding agency Agency providing funding for the project

Key personnel

Name and phone number, fax number, address, email address, function within the facility and other relevant contact information for key personnel on the project, including principal investigator, key researchers, key technicians, the funding agency point of contact and the facility point of contact

Table 11: Load application data elements, according to (Saeed 2003)

Data Element Definition

ALT machine

Loading method

Load type

Load monitoring

Load magnitude

Tire pressure

Contact shape/area

Load configuration

Load frequency

Load test duration

The equipment used to load the pavement

Source of the load (hydraulic, gravity, etc.)

Type of loading device (single wheel, dual wheel, etc.)

Type and placement of sensors to detect load on the pavement; also manufacturer, model numbers, and calibration dates of the load monitoring equipment, e.g., portable scales, weigh-in-motion system, in-pavement sensors

Load imparted to pavement; list both total load and contact pressures at each wheel

Inflation or actual pressure of the tire on the pavement

Area of the footprint of each tire on the pavement surface, and the shape of the contact area

Number of contact areas, wheel position (spacing) of each load; a drawing is the best format to present this information

How often a load pulse is applied to the pavement

Time period that load pulses are continuously applied to the pavement, e.g., an ALT machine runs 20:00 to 05:00 every day for 5 days a week and imparts 2,500 cycles/shift

Test duration

Pulse duration

Test start and stop dates and times and interruptions

Length of time load is applied during each load cycle

Cycles Total number of load cycles imparted over duration of test

Load application speed

Load travel speed over the test pavement; includes any variations in speed; method of measurement (radar, speedometer, etc.)

Tolerances and Units

N/A

N/A

N/A

N/A

±1%, report in kN

±5 kPa

±500 mm

Position: ±10 mm

±0.1%

N/A

N/A

±1 %

N/A

±1 kph distribution and measurement wheel, centreline-to-wheel, etc.), and method of load wander measurement (radar, GPS, etc.)

Loading direction

Load direction (unidirectional, bidirectional, static, impulse)

±10 mm

N/A

CATRIN - Deliverable D7 – January 2009

4.1.3 Pavement Description Data

38

Pavement description data elements characterize the physical properties of a test section. The

data sets are about the design and construction of a pavement section. Table 12 can only give

a suggested minimum of data elements. The objectives of a certain project may implicate the collection of more or different data elements. Drawings should be used to display data elements if the data are better expressed by graphical means.

To know about the age, construction and traffic history of an existing pavement that should be tested is essential for correct results. Details about each layer should be recorded separately.

There is also an element about a QC/QA programme. It should include results of all tests, such as density, modulus and strength tests that were conducted to ensure the quality of the pavement. Also the reported data for each test procedure should include frequency of sampling and testing, location of samples, description of test equipment, test procedure standards and details, number of tests per sample and any statistical analysis of the data. The test procedure standards and statistical analysis should be the same for every European test facilities.

Table 12: Pavement physical description data elements, according to (Saeed 2003)

Tolerances

Data Element Definition

and Units

Pavement Type Flexible, rigid, composite, etc.

Surface AC, SMA, etc.

Shoulder

Design Cross section

If present, type and method of tying to the pavement

Thickness and type of each pavement layer as designed

As-built cross section

Traffic lane dimensions

Test bed dimensions

Slab size

Thickness and type of each pavement layer as constructed and method used to measure thickness, variability of measurements

Size and location of the area where load is applied to the pavement and measurements are recorded

Length and width of the test pavement

Spacing between joints in rigid pavements

Joint details

Load transfer mechanism

Reinforcement

Grade

Cross slope

Pavement origin

Design method

Drainage provisions

Designer

Construction method

Construction agency

QC/QA information

Construction time frame

Weather

Typical joint construction method; may be shown in a drawing

Devices for transferring longitudinal and transverse loads between rigid pavement slabs; size and location of the devices

Size, location and type of tensile pavement reinforcement

N/A

N/A

N/A

Thickness: ± 1 mm

±1mm

±1mm

±1mm

Elevation change along the centreline of the test pavement

Elevation change from edge to edge of the test pavement

Special construction for a programme, existing pavement from another programme, or existing in-service pavement

Method used to calculate the layer thickness attributes of the pavement and the input values for the calculations (design wheel loads, design subgrade conditions, layer shear strengths and moduli, life span, etc.)

Type and location of drainage structures in pavement, e.g., free-draining aggregate, collector pipes, sump pumps, etc.

Name of the person who designed the pavement

Equipment and methods (temperature, compaction method/pattern, curing time/conditions, etc.)

Name, address, and phone number of the contractor or organization that constructed each element of the pavement

QC/QA plan, procedures, and test schedule for each pavement element

(each QC/QA test result should be recorded as a material property)

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Commencement and completion dates of construction for each layer; including any breaks in the construction schedule

N/A

Complete environmental and climate data for the duration of construction N/A

±1mm

Distance:

±1mm

Distance:

±1mm

Distance:

±1mm

±1%

±1%

CATRIN - Deliverable D7 – January 2009

4.1.4 Material Characterization Data

39

Material characterization data elements record information about the properties of the materials used in constructing the pavement layers. These data elements vary depending on the purpose of the project and the type of material being tested. To give examples, some standard materials for road construction are listed here:

-

-

Hot mix asphalt (HMA): Table 13

Portland cement concrete (PCC): Table 14

-

-

-

-

Reinforcement and load transfer device: Table 15

Bituminous stabilized base/subbase: Table 16

Cement, lime and fly ash stabilized base/subbase: Table 17

Unbound aggregate materials: Table 18

-

-

Subgrade: Table 19

Stabilized Subgrade: Table 20

Data regarding material characterization are taken at various stages during design, construction and testing of pavements, as well as after test completion. To every data element there are three components: the test method, the test results and the variability of the results.

The procedure used to perform the test should be standardized and the same for every laboratory. The results should be recorded, along with an indication of whether the reported value is a mean, a weighted mean, a median or some other form of average. Also the number of samples and the variability among samples are to be recorded using a confidence interval, standard estimate of error, standard deviation or other statistical measure.

Table 13: HMA characterization data elements, according to (Saeed 2003)

Data Element

Mix design method

Mix design parameters

Mix design and resulting job mix formula

Binder and modifier characteristics

Aggregate characteristics

Filler characteristics

Additive characteristics

Recycled AC pavement characteristics

Other salvaged or recycled materials

Mix stiffness

Strength

Definition

Mix design specification references (e.g. Marshall)

Gradation limits; volumetric limits; rutting, stiffness and strength criteria

Final job mix formula, gradation, volumetrics, moisture sensitivity, etc.

Varies by project, but should include viscosity, or penetration graded asphalt binder parameters

Source, gradation, particle shape, surface texture, mineralogy, specific gravity, porosity, toughness, hardness, etc. for each aggregate

Gradation, source, specific gravity, and chemical composition

Purpose (emulsifier, tensile reinforcement, etc.), type (liquid, powder, etc.), manufacturer, description, certification, quantity, blend method

Gradation, extracted aggregate gradation, residual binder characteristics, source

(milled surface, etc.), source characteristics, if known (age, etc.)

Type (crumb rubber, crushed glass, shingles, etc.), quantity, description

Resilient modulus, creep modulus, shear modulus, etc. of lab compacted and inplace material

Compressive, shear, or triaxial strength characteristics, etc. of lab compacted and in-place material

Results of an asphalt pavement analyzer, wheel tracking test, etc. of lab compacted and in-place material

Rutting data

CATRIN - Deliverable D7 – January 2009

40

Table 14: PCC characterization data elements, according to (Saeed 2003)

Data Element

Mix design method

Mix design parameters

Job mix formula

Cement characteristics

Course aggregate characteristics

Fine aggregate characteristics

Other pozzalon/fly ash characteristics

Mineral additives

Admixtures

Air content

Workability

Maturity testing

Strength

Definition

Mix design specification references

Workability, strength, and durability criteria

Material proportions (cement, aggregate, sand, additives, other)

Manufacturer, source, type, chemical composition

Source, mineralogy, gradation, specific gravity, absorption

Source, mineralogy, gradation, specific gravity, absorption

Material (lime, fly ash), source, type, chemical composition

Type, source, chemical composition, purpose

Type (air-entraining, water reducing, modulus enhancing, shrinkage reducing, etc.), source, composition, manufacturer certification

Entrained and entrapped air, both design (lab) and QC/QA (field) tests

Slump, etc., both design (lab) and QC/QA (field) tests

Maturity test results

Compressive and flexural strength, both design (lab) and QC/QA (field) tests

Table 15: Reinforcement and load transfer device characterization data elements, according to (Saeed 2003)

Data Element Definition

Type

Size

Material and grade

Coating

Tie bars, mesh, rebar, dowels, etc.

Material size

Material type (stainless steel, fibre glass, etc.) and material strength

Surface characteristics and treatments of the material

Capacity Capacity of a nonstandard load transfer device (shear capacity, etc.)

Manufacturer's certification information Standard manufacturer certification/warranty

Table 16: Bituminous stabilized base/subbase characterization data elements, according to (Saeed 2003)

Data Element Definition

Stabilizer type

Mix parameters

Strength

Stiffness

Aggregate

Type of stabilized material (aggregate, soil, etc.), stabilizer (lime, fly ash, etc.), source, type, chemical composition

Job mix formula and target mix properties (strength, stiffness, etc.), recommended curing procedures

Shear and/or compression strength, before and after stabilization, lab and in-situ samples

Resilient or other modulus, before and after stabilization, lab and in-situ samples

Source, mineralogy, gradation, specific gravity, absorption

Table 17: Cement, lime and fly ash stabilized base/subgrade characterization data elements, according to (Saeed 2003)

Data

Element

Stabilizer type

Mix parameters

Strength

Stiffness

Density

Aggregate

Definition

Type of stabilized material (aggregate, soil, etc.), stabilizer (asphalt, liquid asphalt, etc.), and properties of each (aggregate gradation, asphalt grade, etc.)

Job mix formula and target mix properties (strength, stiffness, etc.)

Shear and/or compression strength, before and after stabilization, lab and in-situ samples

Resilient or other modulus, before and after stabilization, lab and in-situ samples

Lab and in-situ densities, compaction methods

Source, gradation, particle shape, surface texture, mineralogy, specific gravity, porosity, toughness, hardness, etc. for each aggregate

CATRIN - Deliverable D7 – January 2009

41

Table 18: Unbound aggregate materials characterization data elements, according to

(Saeed 2003)

Data Element Definition

Material origin Source (quarried, natural, slag, bottom ash, etc.), and degree of crushing

Gradation Particle size distribution

Fines Percent fines, clay fraction, Atterberg limits

Moisture-density Proctor or Modified Proctor results

Toughness

Soundness

Abrasion loss tests

Environmental loss tests

Shear strength Shear properties or other strength parameters

Stiffness Resilient or other modulus

Other index tests CBR, etc.

Table 19: Subgrade characterization data elements, according to (Saeed 2003)

Data Element Definition

Classification

Gradation

Fines

Stiffness

Soil classification

Particle size distribution from sieve, hydrometer, or other test

Clay fraction, percent fines, Atterberg limits

Resilient or other modulus

Shear properties Cohesion and friction angle

Other strength properties Unconfined compression tests,

Moisture-density

Placement

Proctor or Modified Proctor results

Undisturbed soil or engineered fill

Table 20: Stabilized subgrade characterization data elements, according to (Saeed 2003)

Data Element Definition

Soil properties See subgrade characterization data elements

Aggregate properties See unbound aggregate characterization data elements

Mix parameters

Strength

Job mix formula and target mix properties

Shear and/or compression strength, before and after stabilization, lab and in-situ samples

Stiffness

Density

Resilient or other modulus, before and after stabilization, lab and in-situ samples

Lab and in-situ densities, compaction methods

Moisture-density Proctor, Modified Proctor, or other moisture-density relationship test

4.1.5 Environmental Conditions Data

For EURODEX to be carried out successfully, it will be necessary to at least monitor the environmental conditions as detailed as possible. Therefore each and every pavement test site should maintain a weather station for outdoor facilities or a recording climate control system for indoor facilities. The conditions should be recorded at least every 15 minutes, in case of

controlled environment (environmental chambers, etc.) continuously. Table 21 gives a

suggestion about environmental data that should be collected by the facility.

CATRIN - Deliverable D7 – January 2009

Table 21: Environmental and climate data elements, according to (Saeed 2003)

Data Element Definition

Air temperature Ambient air temperature

Temperature

Temperature at the pavement surface and various depths

Temperature sensor

Humidity

Type of temperature sensor (thermocouple, IR, etc.)

Relative humidity

Precipitation

Wind speed

UV Index

Water table

Instruments

Calibration

Daily amount of precipitation

Average wind speed, gusts

Measure of solar energy

Depth to water table and datum

Weather instruments and data acquisition equipment used

Last calibration of weather instruments, calibration factors

Tolerances and Units

±1°C

Temperature: ±1°C Depth: ±1cm

N/A

±1%

±1mm water, form (e.g., 5mm as

27mm snow)

±1m/s

Absolute

±0.01m

N/A

DD-MM-YYYY

42

4.1.6 Pavement Response Data

Pavement response is defined as the reaction of pavement to loading by a wheel placed on the pavement or by change in moisture content or temperature. As pavement response and the associated parameters are often related to the development of pavement distress and can be used to calculate material properties, it is of particular importance to make sure that data is collected correctly and that it is of high quality.

The collection schedule is generally set during the project planning phase but may be altered during the course of a project if occasion demands it. Pavement response is usually recorded intermittently throughout the project, often measured at intervals (e.g. every 10,000 load cycles). Post-mortem and baseline pavement response testing is also common and necessary to get a complete view. The response of pavements should be recorded in terms of deflection,

strain and stress. Table 22 defines the suggested pavement response data elements.

Pavement deflections are measured by a variety of means. Common methods are geophones on FWD devices, geophones, accelerometers or LVDT sensors in or on the pavement and

Benkelman beams. Deflection data are often used to determine the modulus of pavements and for assessing pavement structural capacity and change over time.

Strain is typically measured at the bottom of the bound layers of pavement by special strain gauges. Strains are also recorded at rigid pavement joints and vertical strains in unbound materials, especially at the top of the subgrade.

Stress is often measured in the vertical direction in unbound materials. Pressure gauges are located in the base or subgrade soil and the stress induced by ALT equipment is monitored.

No matter what kind of instrument is used for measuring, it is important that the values recorded are comparable and that there is a definite interrelationship between values from different types of measurement.

CATRIN - Deliverable D7 – January 2009

43

Table 22: Pavement response data elements, according to (Saeed 2003)

Data Element Definition

Load source

Load magnitude

Configuration

Contact area

Loading rate

Frequency, speed, time, or other measure of how "fast" the load is applied to the pavement

Response type Deflection, strain, stress, pore pressure, etc.

Sensor type Sensor mechanism (LVDT, etc.), type, model number, etc.

Sensor location

Location of the sensor in/on the pavement; a drawing may be the best method to display this information; include longitudinal, transverse, and depth data

Calibration factor

Calibration data

Raw sensor data

Source of the load, e.g., static, rolling, impact (FWD), vibratory; list specific device used to load pavement, including model numbers, etc.

Load imparted to pavement; list both total load and contact pressures at each wheel

Number of contact areas, wheel position (spacing) of each load; a drawing is the best format to present this information

Area of the footprint of each tire on the pavement surface and the shape of the contact area

Number used to convert raw sensor readings into useable engineering units; e.g., 1.2mV from the sensor corresponds to

0.01m deflection.

Date and place of last sensor calibration, technician that calibrated the sensor

Data as recorded by the sensor

Processed data Data in appropriate engineering units

Time stamp Date and time of each sensor reading

Tolerances and Units

N/A

±1%, report in kN

Distance: ±1cm

±5cm

Frequency: Hz

Speed: kph

N/A

N/A

±1mm

Absolute

N/A

Record sensor precision and units

Record precision and units

Record the storage format of the timestamp

Repetitions

Test type

Number of ALT loading cycles the pavement has experienced at time of data collection

QC/QA, in-service, post-mortem, etc.

Absolute

N/A

4.1.7 Pavement Performance Data

Pavement performance as the “serviceability” of pavement over time is invariably linked to the presence of numerous surface distresses, such as rutting, cracking or roughness.

Information about pavement distress function can be found shortly summarized in Chapter 3 of this deliverable or in greater detail in Appendix A.

Performance data are generally measured on a periodic basis (e.g. every 10,000 load cycles).

For in-service pavements, performance data should be measured before the start of load application and an estimate of accumulated traffic at the time of measurement is also essential.

Depending on the objectives of a certain test within EURODEX the type and amount of performance data differ from test to test. Some tests may focus on a particular distress, such as rutting. Other tests may address the overall pavement performance. Multiple surveys should be performed and the variability among results or the confidence interval should be

recorded. Table 23 gives an overview of pavement performance data elements.

For a detailed planning of EURODEX it will be important to define either European or international standards by which certain data elements are obtained. This will guarantee that data of different test facilities and laboratories are comparable and can be used to create a consistent database.

CATRIN - Deliverable D7 – January 2009

44

Table 23: Pavement performance data elements, according to (Saeed 2003)

Type of

Information

General

Performance

Measurement

Data element

Definition

Survey date

Pretest condition

Date(s) of survey

For in-service pavements, condition before load application and an estimate of traffic.

Number of repetitions

Variability

Number of load pulses applied to pavement

Survey purpose Property being measured (smoothness, rutting, etc.)

Survey method

Standard used to perform survey or the complete documentation for the survey protocol

Survey results Results of the survey after the data have been reduced

Standard deviation, range, confidence interval, standard error of estimate, etc.

Raw data

Longitudinal profile

Transverse cross sections

Rutting

Surface distresses

Raw performance data should be stored and the storage format defined; e.g., distress data from a visual survey or elevation data from a roughness survey should be stored

Elevation of the pavement surface in the longitudinal direction in relation to a datum or beam; data should be reduced to a strip chart of elevation vs. station or to an index number reflecting the condition of the pavement, such as the International Roughness Index

Elevation of the pavement surface in the transverse direction in relation to a datum or beam; data should be reduced to a strip chart of elevation vs. transverse location, or to an index number reflecting the condition of the pavement

Rutting is a special case of the transverse cross section

Surface distresses are recorded and reduced to an index number reflecting the condition of the pavement

4.2 Sampling Frequency of Data Measurements

The number of times a particular data element is collected strongly depends upon the type of data element and the objectives of the test. Especially administrative and pavement description data need to be collected only once and maybe updated if required. Other data, such as material characterization and pavement performance are collected at the beginning of a programme and at regular intervals throughout the test. Still others, such as climate and environmental data, should be collected continuously throughout a project. Take a look at

Table 24 for a detailed summary of suggested frequencies.

CATRIN - Deliverable D7 – January 2009

45

Table 24: Sampling frequency of data measurements, according to (Saeed 2003)

Data Category

Administrative information

Load application

Pavement description

Material characterization

Environment and climate

Pavement response

Pavement performance

Type of

Information

General information

Load type and configuration

Frequency

Collect once at the beginning of the test or project

Load magnitude

Tire pressure

Load direction(s) and speeds

Load movement or wander

Collect once at the beginning of the test or project

Record once every 4 hours; record variations in load along section length

Record once every 8 hours (start and end of load application day)

Record load direction at the beginning of the test; collect speed data every hour during load application

Collect continuously

Pavement features

Material properties

Above pavement data elements

Within pavement data elements

Collect once at the beginning of the test or project and confirm at test completion

Collect once at the beginning of the test, forensic-type tests may be conducted at the end of the test series and incrementally during testing if study is to look at changes in material properties

Collect once every 15 to 30 minutes during load application; collect hourly otherwise

Load-deflection

Pavement strain

Collect once every 15 minutes during load application

At the beginning, end, and after a predetermined number of load applications

At the beginning, end, and after a predetermined number of load applications

Soil pressure

Surface distresses

PCC associated distresses

At the beginning, end, and after a predetermined number of load applications

At the beginning, end, and after a predetermined number of load applications

At the beginning, end, and after a predetermined number of load applications

4.3 Data Storage and Retrieval (EURODEX-Database)

The present situation of data storage and retrieval in European pavement research is quite versatile. For small studies, spreadsheets are commonly used for data recording and the principal investigator keeps track of the data. For larger projects, special databases have been developed and a specific person has been assigned to data tracking and storage. The design and layout of the database differs from on facility to the other. Different needs led to different answers to questions like what data to be stored, in which format and whether raw data or already processed data to keep in the database. Many facilities use photographs to document testing and changes in pavement conditions, some even keep video record of activities.

In the future, a shared European pavement research database will be the core of EURODEX.

This database will ensure that raw data produced within this comprehensive damage experiment is stored safely, that data evaluation and statistical methods are standardized and thus comparable. Without creating this database, EURODEX will not be able to use funds efficiently and the output would be useless for a European analysis. Therefore great effort has to be laid upon the design and creation of the database from the very beginning.

As a first step data and results from previous full-scale pavement tests have to be collected and analyzed. The final report of work package 2 of COST 347 (Gourdon 2004) provides an impressive bibliography on previous ALT research topics in Europe and around the world.

Over 760 papers were found and can be used as a basis from where to start a literature review

CATRIN - Deliverable D7 – January 2009

46 of the present knowledge. This should also be the starting point for the database. If available, data from relevant, previous pavement research projects should be gathered and by collecting data, the database can be developed. A database always needs to grow with the data collected for it. Besides that, gaps and missing knowledge can be identified and analysing existing data also ensures that preceding results are not duplicated and funds are spent efficiently.

(Saeed 2003) also gives information about data storage and retrieval. Databases are always composed of hardware and software. With regard to hardware, data have been stored on devices ranging in simplicity from paper to complex optical disks and flash memory cards.

With regard to software, data storage has ranged from written information filed in folders and stored in cabinets to electronic text files and spreadsheets for small data amounts to dedicated databases for large data amounts.

4.3.1 Hardware

A very familiar form of data storage and retrieval is paper. Observations are recorded on paper and stored for later use and analysis. These data are often transferred to an electronic form before being analyzed. The vast majority of data elements are collected in some electronic form on electronic storage media. Storage media range in simplicity from floppy

disks to hard drives and optical disks for larger databases. Table 25 shows common storage

media types and also the suggested use for a European pavement research database.

4.3.2 Software

Electronic text (*.txt) files are simple ASCII files that can be read by most word processing, spreadsheet and database software files. Therefore non processed raw data should always be stored in a txt-format because it is the most compatible file-type. Data fields stored in text files are separated by a comma, a tab or a space. Each row then represents a data record.

Spreadsheet programmes can be used to store and manipulate large amounts of data. A spreadsheet allows the user to organize information into columns and rows. Each cell of the spreadsheet can contain a label, value or a formula. Spreadsheets are commonly used as databases for small amounts of data, but they are generally difficult to verify and audit and do not provide good tools for managing data, whether in terms of consolidation or searching for specific details. When used as a database, spreadsheets are unable to display one record at a time and do not allow multiple-report format. Relational links to other tables and data are also not supported.

Table 25: Data storage media characteristics, according to (Saeed 2003)

Media

CD-ROM

Capacity Advantages

640MB

DVD-ROM 17GB

Common, relatively long life span, small, cheap

Common, relatively long life span, small, cheap

CD-RW 640MB Common, small, cheap

Hard disks

(single)

Hard disks

(RAID)

Tape

Flash disk

300GB

500TB

20GB

16GB

High capacity, long life span, rugged media

Extremely high capacity, extremely rugged media, extremely long life span

High capacity, long life span

Small, rugged media, fast

Disadvantages

Delicate, non-reusable

Delicate, non-reusable

Delicate, relatively short life span

Not portable

Not portable, very expensive

Not common, slow

Not common

Suggested Use

Transport archive

Transport archive

Working copy

Working copy archive

Archive

Network — Fast, common Not really storage Transport

CATRIN - Deliverable D7 – January 2009

47

Dedicated databases that arrange information in tables and records are best suited for a largescale database such as the EURODEX database. Traditional databases are organized as fields, records and files. A field is a single piece of information, a record is a complete set of fields and a file is a collection of records. A database management system (DBMS) consisting of a collection of programmes that enable entering, selection and organizing data in a database is used to access information from a database.

The purpose of recording and storing data is to make them available at a later date for use in analysis. An electronic database is a collection of information optimized for quick selection of desired data using a DBMS. An automatic navigation system, or relational database, is stateof-the-art. It does not require the user to specify how to retrieve the data but what should be retrieved. Data stored in a relational database system are manipulated using standard database manipulation language, Structured Query Language (SQL).

The main purpose of a data storage system is to store and retrieve data. But there are other factors that need to be considered. This includes data safety, ease of use, storage capacity, cost, performance, reliability and manageability.

As the database, its efficiency and longevity are of crucial importance for the success of

EURODEX, IT and database experts have to be involved for the installation of the research database. This report can only show the present state-of-the-art in computer technology.

Experts in this field will have to decide which database system is the optimum for

EURODEX data and results. Besides that it is considered to be very important to store data in its raw format (ASCII,…) to make sure it can be used by different systems as well as in a processed format for the database itself.

CATRIN - Deliverable D7 – January 2009

48

5 Strategic Plan for EURODEX

Previous and present research in the field of pavement is mostly laboratory specific and produces results for materials and designs used by the lab’s member state. This may be sufficient for the development of new materials and pavement design, but when it comes to a complex matter like deterioration models the resources of a single laboratory regarding time, financial and human capacity will always be too scarce to find satisfying results.

The problem we are facing now is that the data produced by hundreds of research projects in the past cannot be used for a European analysis as the data – although high in quality – are not comparable. Nearly every laboratory has its own code of practice and common European standards for material testing are implemented in individual ways by the member states.

The question arises why previous attempts to implement a common code of practice in the

European pavement research community have failed and what makes us sure that EURODEX will not be just another well-intentioned yet unrealizable effort. As an example, COST 347

(“Improvements in Pavement Research with Accelerated Load Testing”) had the objective to develop a European code of practice to optimize the use of ALT facilities and improve the

European application of results. Started in 2000 the project never made it to a complete final report after the scheduled 3.5 years. For one reason or the other the work on the final report of

WP 4 which was the core of the project was stopped in the middle of progress. Still significant findings were found by the other WP. In order not to lose the work of more than 3 years we extracted the most important results and implemented them in CATRIN. So we manage to follow up where COST 347 stopped its work.

On the other hand we found excellent examples how such a comprehensive testing programme can be carried out successfully including many different partners with various interests in the project.

First, the US Long-Term Pavement Performance (LTPP) programme started in 1984. The objective was to find out why some pavements perform better than others and therefore to find improved ways to build and maintain the US highway system. After 3 years of planning the actual test programme started in 1987 with 20 years of data acquisition on over 2,000 test sections on the public road network across the US and Canada. Besides a better understanding of pavement performance another major objective was to strengthen the co-operation of different partners in road construction and research. The LTPP-website

1

provides comprehensive information about the organisation, partners, objectives and the progress of the project as well as papers with results and findings.

The WesTrack accelerated pavement testing experiment – Accelerated Field Test of

Performance-Related Specifications for Hot-Mix Asphalt Construction – is another example of successful long-term pavement research programmes. It was started in 1994 and consisted of a test track which includes 26 HMA test sections. Again a long planning phase of 2 year was the basis for this project. Traffic was initiated in March 1996 and was completed in

February 1999. The WesTrack experiment had two primary objectives. The first was to continue the development of performance-related specifications for HMA construction by evaluating the impact of deviations in materials and construction properties from design values on pavement performance in a full-scale, accelerated field test. The second was to provide some early field verification of the Superpave mix design procedures. The total experiment yielded clearly differentiated levels of permanent deformation and fatigue cracking among the experimental sections.

1

http://www.fhwa.dot.gov/pavement/ltpp/index.cfm

CATRIN - Deliverable D7 – January 2009

49

The experimental results were analyzed to develop the performance models for permanent deformation and fatigue cracking. Also a comprehensive database with test results was developed within the WesTrack experiment. (Epps 2002) describes every stage of the project including results and findings in great detail.

In Task 4.4 of CATRIN we gathered information about each of these projects mentioned above and analyzed their strengths and weaknesses to find out how we can optimize the way to EURODEX. All successful research projects have in common that the planning was carried out in great detail and lasted for a relatively long time – at least 2 years.

Especially from COST 347 we started to understand that it will be of crucial importance to involve potential scientific and financial partners from an early stage on. Only if they know that they can participate in the planning process – in the creative part of EURODEX – they are willing to contribute. Another finding from the analysis of COST 347 was that the more details we know about European pavement research facilities the more efficient we can find out about what European facilities can account for EURODEX and where we lack in equipment at the present point. We have to concentrate on the most important performance factors, as well as on a certain number – 10 to 12 – of pavement materials and designs that should be tested within EURODEX. This is the most serious matter: to strictly limit the size of the testing programme. Otherwise the project will get unmanageable in size regarding time and financial resources.

The US research projects mentioned above show that a common database for data and results is absolutely necessary to make sure that storage, evaluation and analysis is done in a common way. Also we learned from LTPP as well as from WesTrack that it does not make sense to restart research from zero. Even a comprehensive project like EURODEX can only close gaps in knowledge. That is why we have to collect useable data and results from previous pavement research and find out where we stand. What do we already know when it comes to pavement performance and deterioration and where are the gaps we need to close to find improved performance models?

We analyzed the strengths and weaknesses of various large research projects and took into account the present situation in European pavement research. On this basis we provide a profound strategic plan for EURODEX. It is the framework and we believe that it will lead to a research project with consistent findings on a European level if it is carried out step by step.

First steps have been carried out in this EU-project; on this basis further actions have to be

taken. Figure 12 outlines the necessary elements that – if assembled correctly and with high

quality – will lead to a successful European Road Damage Experiment. The way to

EURODEX consists of a foundation with basic and most important elements and four levels of planning, each relying on the level below, producing more pertinence and serving as input for the next level.

5.1 Foundations of EURODEX

First of all economists, researchers in the field of pavement testing, owners and operators

as well as researchers concerned with full-scale pavement test facilities have to be involved

from an early stage on. Meetings and workshops should be held to bring together people from different fields of research and to let them know that they will have a say in the process of EURODEX if they are willing to. It will also bring personal and more informal contacts and it will help to strengthen the co-operation of European facilities. This is a major necessity for the damage experiment. Furthermore the idea of EURODEX can be spread and promoted.

Especially for the European database to be filled with the most important previous test data and results meetings and workshops can be helpful. A more detailed inventory of test

CATRIN - Deliverable D7 – January 2009

50 facilities, equipment and instrumentation can be gathered and information about the present code of practice and used standards for pavement construction, testing, evaluation and statistics will be obtained. The sooner the idea of the database and EURODEX itself is promoted, the more people involved in pavement research can be convinced to work together for it.

EURODEX

Detailed planning

Define mission statement and objectives

Evaluate and analyse existing data

Identify need for additional equipment and instrumentation

Create European database for pavement research data & results

Inventory of European pavement test facilities

Define standards for

EURODEX including

QC/QA

Inventory of RLT* on a

European level

Decide about most relevant pavement designs and materials

Collect data and results from previous pavement research

Isolate performance and distress indicators of pavements

Install European committee concerning the co-operation and co-ordination of EURODEX

Involve economists, researchers, owners and operators of pavement test facilities, EC and national representatives, private and state road managers by organizing meetings and workshops

*test sections on public road network

Figure 12: Strategic plan for EURODEX

From these meetings a second foundation of EURODEX should be developed: a European

committee concerning the co-operation and co-ordination of EURODEX. This committee will be the central point to co-ordinate the scientific and financial planning and realization of

EURODEX. Representatives of ALT owners and operators, researchers concerned with pavement testing, economists, representatives from the EC, from the national ministries of transport, as well as private and state road managers responsible for test sections on the public road network (RLT) should be part of it. All further steps to EURODEX will be discussed, negotiated and decided by the committee. This central organization is necessary to ensure that the approach of the damage experiment is co-ordinated and research efforts are not being duplicated. Also the design of the European database should be worked out in the committee, as well as the decision which standards to use for construction, instrumentation, measurement, evaluation and statistics for EURODEX. The QC/QA-system should be established by the members, the question which pavement designs and materials to use for the experiment and whether there is any need for additional equipment for European ALT facilities. In addition the detailed planning will be co-ordinated by the committee and the research assignments will be split into small-scale laboratory testing by university or private labs, ALT and RLT programmes. To spend funds efficiently, lab testing should cover as much research work as possible, as it is the most inexpensive way of testing. Nevertheless as this is a road damage experiment, a comprehensive ALT programme will be necessary to provide useful data. To verify results derived from ALT, RLT on in-service pavement under real traffic conditions will be inevitable. Since the tasks of the committee are most important for EURODEX, it is crucial that it is manned by a variety of experts and people who have the authority to decide from as many different EU-countries as possible. It is of particular importance that the EU-

CATRIN - Deliverable D7 – January 2009

51 countries with ALT-facilities and a large public road network are represented in this organization.

5.2 First Level - Basics

The first level contains basic knowledge about the present situation in full-scale pavement testing in Europe and consists of four elements.

The inventory of European ALT facilities can be found in this report in Chapter 2 as well as

in Annex A in greater detail. It may be necessary to update the inventory at the beginning of

EURODEX to check whether presently existing facilities have been closed or new test sites have been commissioned. There will also be the need to contact the facilities and promote the idea of EURODEX to the operators and research teams. At the same time, invitations for meetings and workshops mentioned above can be extended and potential members for the committee can be found.

The inventory of RLT on a European level will be another important task. Since the lack of such an inventory on a European or even national level, it will take time and research resources to complete this task. Still it is important to prepare this inventory with pavement material used, research objectives, etc. to get an idea about the existing data and knowledge in this research area. RLT on the public road network is often co-ordinated by national ministries of transport or road administration. By contacting these ministries and administrations, people concerned with RLT can be informed about EURODEX and potential members from the ministries’ of transport as well as private and state road managers for the committee can be found.

To gather data and results from previous pavement research projects is the first step for a

European pavement research database. A lot of research on this topic has already been done by COST 347. The final report of work package 2 (Gourdon 2004) includes a bibliography on previous ALT tests around the world. It contains over 760 papers on pavement research. This is a perfect starting point for a literature review. Also the NCHRP Synthesis 235 (Metcalf

1996) and 325 (Hugo 2004) give detailed information about ALT and RLT programmes carried out over the past 20 years. To find out about recent and ongoing European ALT and

RLT projects, the facilities have to be contacted. Considering the abundance of data and results, the committee will have to sort out generously and keep only the most important findings and also only the results that can be used for further evaluation for EURODEX.

Since EURODEX will have to concentrate on the most important research topic to prevent it from getting unmanageable in size, most important performance indictors and distress

functions should be isolated. This work has been completed by this project; the most important finding can be found in Appendix A of this deliverable. It is still necessary to critically review these factors before starting detailed planning of EURODEX to find out whether one indicator or the other has to be added or deleted from the list.

The higher the quality of the four research items given in this level is, the easier it will be for the next steps to be carried out.

5.3 Second Level - Framework

The second package contains large pieces of research and administrational work. First of all the European database for pavement research results will be established on this level.

Basic requirements about a database like this can be found in Chapter 4 and also in the

NCHRP Report 512 (Saeed 2003). The database will be the core of EURODEX since it will contain every piece of data and findings of the damage experiment as well as significant results from previous research. The last-mentioned data as well as the inventory of European

CATRIN - Deliverable D7 – January 2009

52

ALT facilities and RLT which will also find its way into the database can be taken from

Level 1.

Since the database, its efficiency and longevity are of highly important for the success of

EURODEX, IT and database experts have to be involved for the installation of the research database. Experts in this field will have to decide which database system is the optimum for

EURODEX data and results. It is considered as essential to store data in its raw format

(ASCII,…) to make sure it can be used by different systems as well as in a processed format for the database itself.

To make sure that the code of practice is the same for every test no matter if it is from smallscale lab testing of pavement materials or from full-scale ALT or RLT test sites, one package of European standards has to be defined which will be obligatory for every participant of

EURODEX. COST action 347 was started in 2000 with the objective to provide a ”common code of good practice for the application of ALT” but never made it to a complete final report.

Nevertheless, findings from the different work packages were brought together in different final reports and can be used. Standards must cover every step of the damage experiment from pavement construction and material testing, to instrumentation, data acquisition, evaluation

and storage up to statistical methods. Findings from (Saeed 2003) summarized in Chapter 4 of

this report give suggestions about the minimum requirements of data elements to be recorded.

Standards for nearly all parts of the damage experiments already exist on a European level but with different implementation in the member states. Therefore the participants of EURODEX have to find an agreement upon which standards to use. To assure the quality of data and results a QC/QA system will have to be implemented.

The first concrete planning will happen in Level 2, when it comes to decide about most

important pavement designs and materials. There are hundreds of materials and designs for layered pavements in use all over the European Union. It can never be the objective of

EURODEX to test and evaluate all available material and design. We rather have to concentrate on the most important materials and designs and find correlations between results of materials and design tested within EURODEX and materials and designs used in different member states. European engineers and experts in pavement testing have to discuss this topic thoroughly and find an agreement on 10 to 12 designs representing the European road network in the best possible way. EURODEX should contain rigid and flexible pavements. As semi-rigid pavements are not common for high-level road infrastructure, this kind of pavement will not be part of the research project. The decision about designs to be tested within EURODEX is clearly a critical point as every member state has its own interest in testing its own materials and designs. But only if an agreement is found on this topic,

EURODEX can be carried out at all. A design or material not represented in the damage experiment does not mean that there will be no finding for this specific material or design. It rather means that correlations between tested materials and non-tested materials need to be and can be found. Finding these correlations will be an important part of EURODEX.

5.4 Third Level – Preparation

By creating the database and defining common standards for the damage experiment in the level below, now the existing, relevant data from previous full-scale tests can be evaluated

and analyzed. As there are guidelines for evaluation and statistics and a database for storage of processed data at this point of time, existing test data can be processed. By feeding the database with these data, shortcomings of the database can be detected and it can be adjusted to actual demands. This task brings to light what we actual know about pavement performance and deterioration, in other words the existing knowledge of the European pavement research community.

CATRIN - Deliverable D7 – January 2009

53

Before actual test programmes of EURODEX can start we also have to isolate the need for

additional equipment and instrumentation for the existing ALT-facilities and European pavement research laboratories. With the inventory from Level 1 which includes a Strength-

Weakness-Analysis from this report, there will be a precise knowledge about what European

ALT-facilities can do and what they cannot do at this point. As COST 347 has already found out, from the present point of view there is no need for a new European ALT facility. If the existing facilities strengthen their co-operation, some features will have to be added to one facility or the other, but compared to what we already have, the need for additional equipment are relatively small.

5.5 Fourth Level – Launching EURODEX

As a last step before EURODEX can be started with a detailed project and test programme

gaps and missing knowledge have to be identified. On the level below, data from previous full-scale pavement tests have been evaluated and analyzed. Scientists, pavement research experts and economists now have to discuss about what we lack in knowledge, what we need to know to find out more about pavement performance and deterioration and also what we need to create a sustainable and fair concept for cost allocation on European roads.

On this stage it has to be decided how to link laboratory small-scale testing, ALT and RLT

(instrumented roads, road service measurements,…) in an optimal way to get data for a wide field of analysis, such as maintenance strategies, marginal cost analysis, optimization of vehicles, etc. Finally, a mission statement and all objectives have to be defined distinctively and precisely to guarantee the success of this pioneering European road infrastructure experiment.

We are positive that various stakesholders in European transport infrastructure will profit from EURODEX’ findings and the implementation of results. EURODEX will provide a solid basis for a European Transport Policy and a source-related cost allocation on European roads.

With the database, the EC will have strong tool for the review of future project propsals and

EU research funds can be spent even more efficiently. In the same way, member states and taxpayers will be on the winning side. Results can be used by road operators to make maintenance works more efficient and economic, the road industry can improve its competitiveness in international bidding. Road users will find a fair pricing principle on roads and travel safer if the findings of EURODEX are realized. Details about benefits for stakeholders can be found in the next chapter.

CATRIN - Deliverable D7 – January 2009

54

6 Benefits from EURODEX

EURODEX will rely on time, human and financial resources. In detail we consider 2 to 3 years of planning followed by 4 to 5 years of test programmes. Considering the budget, LTPP for example had a budget of 50 m$ for 20 years of comprehensive data acquisition, evaluation and analysis.

From the present point of view we are sure that all players in European road infrastructure will profit from EURODEX, its findings and the implementation of results.

6.1 Politics

6.1.1 European Commission

Today we rely on rough estimations about pavement deterioration due to vehicles when it comes to source-related cost allocation. EURODEX will overcome the 4 th

power rule and find improved models for the complex material-vehicle-environment interaction. In other words an enhanced and more reliable pavement performance and deterioration model adequate for the road infrastructure of the 21 st

century is one of the major objective of EURODEX. The experiment will result in a more detailed knowledge about how much each factor (e.g. temperature, moisture content, axle weight, axle configuration, etc.) contributes to the damage of a certain pavement material/design. This provides a solid basis for a European Transport

Policy. Furthermore, it gives legal certainty, and a fair system of transport pricing can be implemented on European roads; fair in terms of source-related cost allocation. Road users will have to pay for the damage caused by their vehicle taking parameters like axle and wheel configuration, suspension system, axle weight, etc. into account.

Another output for EU-legislation from EURODEX is to provide guidelines for the optimal

weights and dimensions of HGVs. For example, today forwarding companies tend to use their vehicle fleet with highest possible tyre pressure to reduce fuel consumption and CO

2

emissions. The problem is that high tyre pressures reduce the contact area between tyre and pavement and this leads to higher contact pressures and increases deterioration. With

EURODEX adequate axle weights and tyre pressures to optimize road life expectancy versus

CO

2

-emissions can be investigated.

A basic requirement of EURODEX will be for the participants to agree on one package of standards used for each step of the experiment as a common code of practice. Furthermore a

European database for pavement research will be developed within the project. This ensures that research data and results from EURODEX and future projects will be comparable and can be used for a European analysis. With the database the EC has a strong tool for the review

of future project proposals. The reviewer can find out whether a proposed research item has already been covered by previous programmes by simply taking a look into the database.

Therefore EU research funds will be spent more efficiently. The common code of practice for European pavement research labs can be used for future projects as an obligation for participating partners. This ensures that data are collected, evaluated and analyzed in a comparable way.

6.1.2 Member States

As the EU will give guidelines for a common pricing principle on European roads based on the findings from EURODEX, the member states have a strong basis upon they can

CATRIN - Deliverable D7 – January 2009

55

implement national transport pricing. As an analytical deterioration model will be the basis for transport pricing, today’s common criticism about arbitrary tolls for road users will be

obsolete. Another advantage for the member states is that costs for road construction and especially maintenance will be covered in a source-related way. Financing for these matters can be split in a more accurate and fair way between road-users and other tax sources.

It will be more certain to decide which fraction of costs is related to deterioration by vehicles and which fraction is related to other sources such as climate or construction quality.

As all member states will implement their transport pricing based on a common pricing

principle, the present problem of shifting HGV-traffic from one member state’s road network to another’s due to different transport pricing can be solved.

Similar to the EC, the research database and the package of standards as a common code of practice in pavement research can be used by the member states to make more efficient use of

their research funds. The database can be used to check whether objectives of project proposals have already been covered by previous programmes and the common code of practice can be set as an obligation for project partners to ensure comparable data and results.

6.1.3 Taxpayers

Taxpayers will also be on the winning side of EURODEX. Even before source-related transport pricing based on an improved deterioration model will be installed in the member states, the findings of EURODEX will be used to develop more efficient strategies for road construction and maintenance. Through an enhanced understanding of pavement performance and why certain pavement types perform better then others by then, the application of materials and renewal of road infrastructure can be optimized. Therefore especially pavement

rehabilitation will require less tax money.

With the implementation of transport pricing on European roads taxpayers will no longer have to pay for road infrastructure costs related to deterioration by road-users. Taxpayers will only have to cover costs for road construction and maintenance that is related to general factors such as the specific climate. All other costs will be paid by actual source of damage – by each user or – if the construction quality was found to be poor – by the construction consortium.

More efficient spending of research funds on a European and national level means that each

Euro paid in taxes will bring more payback.

6.2 Road Infrastructure

6.2.1 Road Managers

Private and state managers of road infrastructure will benefit from EURODEX in several ways, strategically as well as operationally.

EURODEX will provide a pavement performance model that allows road managers to update and optimize their design guides and maintenance standards. Pavement management systems will get improved knowledge how to divide funds up into maintenance, rehabilitation and renewal of pavements in the most efficient way.

Today monitoring of pavement condition is mostly carried out by measuring parameters only on the surface, such as rutting or road grip, etc. Full-scale pavement tests carried out within

EURODEX will show which parameters are significant to predict pavement performance

correctly. From that improved monitoring systems can be obtained that measure relevant

CATRIN - Deliverable D7 – January 2009

56 performance indicators on the surface as well as within a structure, such as moisture content or strains in different layers.

For the first time the effects of material variability, construction quality and maintenance levels on pavement distress and performance will be determined in a consistent way. Road operators will have a strong tool at hand to decide when the perfect time for maintenance

works in terms of safety and cost efficiency arrives. They will know more about how construction quality (e.g. compaction rate, etc.) influences the lifespan of a certain material or design and can use this knowledge for tendering. This fact combined with the improved deterioration models will lead to more sustainable and efficient strategies for the

rehabilitation of existing pavements on an enhanced life-cycle cost approach.

A constantly updated pavement research database keeps road operators up-to-date and

findings from research can be implemented more quickly.

6.2.2 Road Industry

The road industry includes all parts of road infrastructure works from planning and project management to the actual construction and maintenance. It also contains research facilities of private enterprises.

EURODEX will evaluate existing designs and pavement materials on a European basis for the first time. Therefore the road industry will learn more about designs and materials used in

different member states. This is especially important when it comes to international

tendering and bidding.

As the road industry can enhance their knowledge about the effects of various factors (e.g. loading, environment, construction quality, etc.) on pavement performance, more efficient strategies for construction and maintenance can be developed based on this knowledge. It

improves the competitiveness of the road industry and makes road works more economic.

In case of private research labs which are willing to share their results within the research

database, they will be granted access to data and findings from other public and private research projects which means a significant advantage in competition. In addition enterprises’ spending of research funds will be more efficient as they can find out easily if a certain research item has been covered by previous research.

6.2.3 Road Users

Last but not least in road infrastructure the actual users will have their share of benefits from

EURODEX. The project will provide the basis for an improved pricing principle on European roads. This does not mean that EURODEX turns against European road users. In fact most of the member states have already implemented their tolling systems for passenger cars and

HGVs. The problem nowadays is that in many cases these systems are based on rough estimations of cost allocation. EURODEX will provide an analytical deterioration model taking into account all major distress factors. A fair pricing principle can be installed based on the findings of EURODEX. Road users will only have to pay their share of costs caused by the vehicle type.

EURODEX will also show in which ways tyres, suspension systems and other parts of

vehicles can be optimized to reduce vehicle and pavement deterioration and therefore

increase the life expectancy of vehicles and pavement structures.

Moreover as road operators know more about the optimal time to start road maintenance and improved strategies and materials for rehabilitation this will increase the safety on

European roads significantly for all users.

CATRIN - Deliverable D7 – January 2009

6.3 Research Community

57

Since the European road damage experiment will be a shared research project, the European research community will profit strongly by EURODEX.

First of all the pavement research database will be accessible to all research facilities that are willing to share their data. It is a strong tool to stay up-to-date with the latest findings and make sure that already existing results are not being duplicated. The progress in the

European research will considerably speed up which permits an advantage over other research regions in the world.

EURODEX will strengthen the co-operation of European test facilities and the pool

resources of laboratories. This advantage of European research can be used by future projects to enhance the outcome.

With the improved pavement performance and deterioration model the European research community has the chance to overcome the 50-year-old 4 th

power rule with its well-known shortcomings. The improved model will set new standards when it comes to pavement deterioration in Europe and throughout the world. Europe will assure and strenghten its

leading position in pavement research.

6.4 Benefits from improved performance model

Besides from the perspective of different stakeholders, there is yet another way to look at benefits from EURODEX.

Figure 13 gives an overview of benefits from the improved performance model and how different benefits are connected and interact. It also shows, who of the stakeholders benefits from a certain item.

Determine effects of vehicle-, environment-, material-parameters, material variability, construction quality and maintenance level on pavement performance and deterioration

Provide improved analytical model to describe pavement performance and deterioration

Tool to determine perfect time for maintenance works

RM, RI

More efficient, sustainable strategies for construction and maintenance

TP, RM, RI, RU

Road maintenance needs less tax money

EC, MS, TP

More efficiency in tendering

EC, MS, RM, RI

Improve competitiveness of highway industry

MS, RM, RI

Solid basis for European

Transport Policy

EC, RM

Legal certainty of transport pricing on European roads

EC, MS, RM

Strong basis for national transport pricing in the member states

MS, RM

Overcome 4th power rule

RC

Set new standards with improved model worldwide

EC, MS, RC

Assure EU’s leading position in pavement research

EC, MS, RC

Enhanced safety on

European roads

Common pricing principle in the member states

MS, RM, RU

EC

TP

EC, MS, TP, RM, RU

European Commission

MS Member States

Taxpayers

Overcome shifting of

HGV-traffic due to

RM Road Managers

RI

Road Industry

RU

Road Users

RC

Research Community different pricing principles

EC, MS, TP, RM, RU

Source-related cost allocation – fair to all users

TP, RM, RU

Figure 13: Benefits from EURODEX findings (performance model)

On the one hand, an improved pavement performance and deterioration model will give a good tool to determine the optimal time for maintenance or renewal works. This leads to more

CATRIN - Deliverable D7 – January 2009

58 efficient and sustainable strategies for road construction and therefore maintenance and reconstruction will consume less tax money. Besides that, this will enhance the safety on

European roads.

On the other hand, we will know more about how construction quality (e.g. compaction rate, etc.) influences the lifespan of a certain material or design. This brings more efficiency for tendering. The road industry can enhance their knowledge about the effects of various factors

(e.g. loading, environment, construction quality, etc.) on pavement performance, more efficient strategies for construction and maintenance can be developed based on this knowledge. It improves the competitiveness of the road industry and makes road works more economic.

As EURODEX will quantify the influence of different factors on pavement deterioration, the share of different vehicle types in the process of deterioration is determined. This gives a solid basis for a European Transport Policy and legal certainty of transport pricing on roads. It is an excellent basis for the implementation of a common pricing principle in the different member states and will overcome today’s problem with shifting HGV-traffic from one member state to the other due to different pricing principles. Moreover this type of transport pricing is sourcerelated and therefore fair to all users.

6.5 Benefits from research database

Figure 14 shows benefits from the research database and the common code of practice

developed within EURODEX.

European Pavement Research Database

Common Code of Practice

Strengthen co-operation, pool resources of

European laboratories

EC, MS, RC

Efficient tool for project proposal review

EC, MS

Spend research funds more efficiently

EC, MS, TP

Comparable data for

European analysis

EC, MS, RI, RC

Evaluation of existing materials and designs on

European basis

RM, RI, RC

Tool to stay up-to-date with latest findings

RM, RI, RC

Implement researchfindings more quickly into construction practice

RM, RI, RU

Get more results for each

Euro paid in taxes

EC, MS, TP, RM, RI

Knowledge about materials/designs used in different member states

RI, RC

Speed up progress in

European pavement research

RM, RI, RU, RC

Advantage in worldwide research competition

MS, RC

Advantage for international tendering and bidding

RM, RI

Assure EU’s leading position in pavement research

EC, MS, RC

EC

European Commission

MS Member States

TP

Taxpayers

RM Road Managers

RI

Road Industry

RU

Road Users

RC

Research Community

Figure 14: Benefits from EURODEX findings (research database)

As the database will contain all findings from EURODEX itself, important previous projects and should be constantly updated, it is an efficient tool for project proposal review. Just by taking a look into the database the reviewer can find out whether a proposed research item has already been cared about in a different project. The common code of practice and EURODEX strengthens the co-operation of European pavement research laboratories. All of this means that research funds can be spent more efficiently and we will obtain more results for each

CATRIN - Deliverable D7 – January 2009

59

Euro paid for research. It will speed up the progress of European pavement research and – again – assure EU’s leading position in pavement research.

The common code of practice is a precondition to get comparable data for a European analysis. Data from different test programmes carried out within EURODEX lead to the evaluation of existing materials and design on a European basis. We can enhance our knowledge about materials and designs used in the different member states. Road managers and the road industry can use this knowledge as an advantage, when it comes to international tendering and bidding.

The database is also a tool for different stakeholders who are willing to share their results in the future to stay up-to-date with the latest findings in pavement research. These findings can be implemented more quickly into the practice of road construction.

CATRIN - Deliverable D7 – January 2009

60

7 Conclusions

In order to overcome the deficient 4 th

power rule and find improved relationships between material, environment and vehicle parameters, it is necessary to collect already existing knowledge in this field of pavement research, find out where gaps in knowledge are located and fill these gaps by a systematic research programme. This programme will consist of small-scale laboratory testing whenever possible and necessary, full-scale ALT and RLT as the central point of EURODEX combined with numerical simulation and statistical methods.

Since accelerated load testing will be the main part of EURODEX, the report concentrates on this test type.

This report outlines a strategic plan for a first empirical EUropean ROad Damage EXperiment

– EURODEX and draws its framework and preliminary design. An inventory of European

ALT facilities is provided and guidelines for common data acquisition, data storage and retrieval are given. Most important performance indicators and distress functions of road pavements can be found in Appendix A of this deliverable.

As the 12 active European ALT facilities up to this day mainly served national research topics, there are few co-operations between different test sites to bundle their strengths and less co-ordination of the various research programmes carried out in the different national institutes. To launch EURODEX successfully it will be necessary to strengthen the cooperation between European ALT-facilities and co-ordinate the research efficiently. Only if comparable data is produced by the test sites, reliable pavement performance and deterioration models can be obtained. Respective guidelines will ensure proper interpretation of data and facilitate their use in different, participating laboratories. Duplications of research efforts can be avoided as an additional benefit from EURODEX. There are also recommendations for the collection, storage and retrieval of data in a common European pavement research database. This database shall be the core of EURODEX. It will contain all relevant data, results and findings from literature review, small-scale lab tests, full-scale pavement tests and numerical simulation related to pavements.

The way to EURODEX consists of a foundation with different basic elements and four levels of planning, each relying on the level below, producing more pertinence and serving as input for the next level.

For EURODEX to be carried out in the forthcoming FP it is of grave importance to prepare this comprehensive project thoroughly. First steps have been carried out within the respective project CATRIN; on this basis further actions have to be taken. The foundation of

EURODEX is to involve people who will participate and work for this project from an early stage on by organizing workshops and meetings. Furthermore a European committee concerning the co-operation and co-ordination of EURODEX has to be installed. The first level contains basic elements to be planned for the damage experiment. Inventories of

European test sites have to be set up – for ALT-facilities this task has already been done in

this EU-project (Chapter 2). Data from previous pavement research projects have to be

collected and performance and distress indicators isolated. The last-mentioned point has been completed by Task 4.3 and can be found in Appendix A. A short summary on pavement

deterioration, distress and performance factors is provided in Chapter 4. On the basis of the

tasks mentioned above the European database for pavement research should be developed.

Common standards for all types of tests being carried out within EURODEX have to be defined to gain comparable data and results. A decision about pavement designs and materials to be tested within EURODEX has to be made. Finally, the collected data and results have to be evaluated and analyzed, needs for additional equipment for European test facilities should

CATRIN - Deliverable D7 – January 2009

61 be isolated and gaps in knowledge concerning pavement performance and deterioration should be identified. These gaps will be closed by EURODEX.

From the research and literature review the following insights have been gained:

-

With 12 active full-scale ALT facilities combined with small-scale pavement research laboratories and an unknown number of test sections on the public road network (RLT),

-

Europe is excellently equipped for a new empirical European road damage experiment

(EURODEX).

When it comes to RLT, there are no statistics or even numbers about test sections on the

European level and in hardly any of the member state. It is of crucial importance to inventory these test sections with as many pieces of information about location, materials,

-

-

- objectives, etc. as possible to find out about the state-of-the-art in RLT. This is necessary to carry out EURODEX efficiently.

The full-scale ALT facilities mainly work for national research purposes. However, there are only few co-operations and a lack of co-ordination on a European level. An exception is the HVS-Nordic that is jointly owned and operated by Sweden and Finland. The research done in each of the facilities has high quality, but for EURODEX it is important to strengthen the co-operation and install a co-ordinating committee for full-scale pavement testing as the core of EURODEX.

Since ALT facilities work in particular for national research, every test site uses different standards when it comes to constructing pavements, pavement material testing and the collection of pavement performance and deterioration data. For example, there are 12 different test termination criteria in the 12 European ALT-facilities. For EURODEX, the participants have to agree on which standards to use for each step of the project.

Respective data guidelines are presented in this report (Chapter 4).

A Strength-Weakness-Analysis carried out in COST 347 found that the European ALT

-

-

- facilities when working together and bundling their strengths are excellently equipped.

There is no need to build any new facilities in the EU but to adapt one facility or the other and install additional equipment according to a detailed plan.

A lot of research effort has been made in the last 20 years in European pavement research.

The problem today is that data and results have not been collected in a common European pavement research database. To develop such a database is crucial for EURODEX to be carried out successfully. Relevant data from previous research programmes have to be collected, evaluated and analyzed. The data guidelines provided in this report will help to evaluate existing data in a uniform manner and store and analyze data from EURODEX and other pavement research projects in a common way. Therefore a QC/QA system for

EURODEX should be implemented.

The objective of EURODEX is to find improved models for pavement performance and deterioration by means of laboratory and full-scale pavement testing, as well as by means of numerical simulation. The derived models will be a reliable and improved basis for a sustainable and fair transport pricing on European roads, they will contribute to make road construction and especially maintenance more efficient on an innovative life-cycle analysis approach. It will provide important insights for the pavement research community. Furthermore a database for pavement research shall be developed. To make

EURODEX a strong tool, it is necessary to find participants from many different fields, like economists, researchers, owners and operators of testing facilities, EC representatives, private and state road managers, etc.

It is important for EURODEX to concentrate on most important performance indicators and distress functions as well as on a particular number of pavement materials and designs commonly used by the member states. This is necessary to stay within acceptable

CATRIN - Deliverable D7 – January 2009

62

-

- limits regarding time and financial efforts. Therefore the participants of EURODEX must find an agreement especially when it comes to pavement materials and designs. Thus one objective of EURODEX is also to find correlations and relationships between the tested materials/designs and those used by member states.

For a comprehensive research project like EURODEX it is essential to carry out the planning systematically and thoroughly. Therefore a strategic plan is presented in this

report (Chapter 5) to plan each step of the experiment.

To link laboratory small-scale testing, ALT and RLT (instrumented roads, road service measurements, etc.) in an adequate manner is of crucial importance to have data for a wide field of analysis, such as maintenance strategies, marginal cost analysis, optimization of vehicles, etc.

All stakeholders in European road infrastructure will substantially profit from EURODEX, its findings and the implementation of results. Politics, the research community as well as players in road infrastructure will benefit from the project:

EURODEX will provide a new solid basis for a European Transport Policy and a sourcerelated cost allocation on European roads. With the database, the EC will possess a strong tool for the review of future project propsals and EU research funds can be spent even more efficiently. In the same way, member states and taxpayers will be on the winning side. Results can be used by road managers to make maintenance works more efficient and economic, the road industry can improve its competitiveness in international bidding. Road users will find a fair pricing principle on roads and travel safer if the findings of EURODEX are realized.

Results and findings from EURODEX will form a strong tool to improve not only materials and pavement design, it will also set the directives in which manner tyres, suspension systems and other parts of vehicles may be optimized to reduce pavement and vehicle deterioration and therefore improve the life expectancy of roads. A detailed analysis of benefits from

EURODEX is given in Chapter 6.

CATRIN - Deliverable D7 – January 2009

63

8 References

American Association of State Highway Officials (AAHSO) (1961): The AASHO Road Test:

History and Description of the Project, Special Report 61A, Highway Research Board,

Washington D.C., 1961

Dawson A., Hildebrand G., et al. (2002): COST 347: Improvements in Pavement Research

with Accelerated Load Testing, Final Report WP1, Brussels, 2002

Epps A. E., Hand A., et al. (2002): NCHRP Report 455: Recommended Performance-Related

Specification for Hot-Mix Asphalt Construction: Results of the WesTrack Project,

Washington D.C., 2002

Golkowski G., Hofbauer M., et al. (2004): COST 347: Improvements in Pavement Research

with Accelerated Load Testing, Final Report WP3, Brussels, 2004

Gourdon J.-L., Balay J.-M., et al. (2004): COST 347: Improvements in Pavement Research

with Accelerated Load Testing, Final Report WP2, Brussels, 2004

Hildebrand G., Stryk J., et al. (2004): COST 347: Improvements in Pavement Research with

Accelerated Load Testing, Final Report WP4, Brussels, 2004

Hugo, F., McCullough B.F., Van der Walt B. (1991): Full-Scale Accelerated Pavement

Testing for the Texas State Department of Highways and Public Transportation,

Transportation Research Record 1293, Transportation Research Board, National Research

Council, Washington D.C., 1991, pp. 52–60.

Hugo F., Martin A.L.E., et al. (2004): NCHRP Synthesis 325: Significant Findings from Full-

Scale Accelerated Pavement Testing, Washington, 2004

HVS International Alliance (2002): Minutes from the International Heavy Vehicle Simulator

Workshop, CSIR, Pretoria, South Africa, 2002.

CATRIN - Deliverable D7 – January 2009

Mahoney, J.P. (1999): Accelerated Pavement Testing: An Overview, Proceedings from the

First International Conference on Accelerated Pavement Testing, Reno, Nevada, 1999

Metcalf J.B., et al. (1996): Synthesis of Highway Practice 235: Application of Full-scale

64

Accelerated Pavement Testing, Washington, 1996

Mn/DOT Office of Materials and Road Research (1999): Load testing of instrumented

pavement sections – Literature review, Minnesota, 1999

Molenaar A.A.A., Groenendiijk J., Van Dommelen A. (1999): Development of Performance

Models from APT, Proceedings from the First International Conference on Accelerated

Pavement Testing, Reno, Nevada, 1999

National Physical Laboratory (2007): NPL’s History Highlights, online: http://resource.npl.co.uk/docs/educate_explore/history/history_of_npl.pdf

(01.09.2008)

Saeed A., Hall J.W., et al. (2003): NCHRP Report 512: Accelerated Pavement Testing: Data

Guidelines, Washington, 2003

Turtschy J.-C., Sweere G. (1999): Long-Term Pavement Performance Modeling Based on

Accelerated Pavement Testing, Proceedings from the First International Conference on

Accelerated Pavement Testing, Reno, Nevada, 1999

White T.D., Hua J., Galal K. (1999): Analysis of Accelerated Pavement Tests, Proceedings from the First International Conference on Accelerated Pavement Testing, Reno, Nevada,

1999

WWW-Site: http://www.pave-test.org/altp-pic.htm

(02.09.2008)

CATRIN - Deliverable D7 – January 2009

65

Appendix A: Relationships between traffic and pavement maintenance costs

SIXTH FRAMEWORK PROGRAMME

PRIORITY 1.6.2

Sustainable Surface Transport

CATRIN

Cost Allocation of TRansport INfrastructure cost

Relationships between traffic and pavement maintenance costs – Deliverable D7 (Appendix A)

Outline of a New Empirical Road Damage Experiment

Version 1.0

January 2009

Authors:

Robert Karlsson (VTI) with contribution from partners

Contract no.: 038422

Project Co-ordinator: VTI

Funded by the European Commission

Sixth Framework Programme

CATRIN Partner Organisations

VTI; University of Gdansk, ITS Leeds, DIW, Ecoplan, Manchester Metropolitan University, TUV

Vienna University of Technology, EIT University of Las Palmas; Swedish Maritime Administration,

University of Turku/Centre for Maritime Studies

CATRIN - Deliverable D 7 – January 2009

66

CATRIN

FP6-038422

Cost Allocation of TRansport INfrastructure cost

This document should be referenced as:

Robert Karlsson (VTI) with contribution from partners, CATRIN (Cost Allocation of

TRansport INfrastructure cost), Relationships between traffic and pavement maintenance costs - Deliverable D7 (Appendix A) Outline of a New Empirical Road Damage Experiment.

Funded by Sixth Framework Programme. VTI, Stockholm, January 2009

Date: January 2009

Version No: 1.0 above.

PROJECT INFORMATION

Contract no: FP6 - 038422

Cost Allocation of TRansport INfrastructure cost

Website: www.catrin-eu.org

Commissioned by: Sixth Framework Programme Priority [Sustainable surface transport]

Call identifier: FP6-2005-TREN-4

Lead Partner: Statens Väg- och Transportforskningsinstitut (VTI)

Partners:

VTI; University of Gdansk, ITS Leeds, DIW, Ecoplan, Manchester Metropolitan University,

TUV Vienna University of Technology, EIT University of Las Palmas; Swedish Maritime

Administration, University of Turku/Centre for Maritime Studies

DOCUMENT CONTROL INFORMATION

Status:

Distribution:

Availability:

Filename:

Quality assurance:

Co-ordinator’s review:

Signed:

Draft/Final submitted

European Commission and Consortium Partners

Public on acceptance by EC

Chris Nash

Gunnar Lindberg

Date:

CATRIN - Deliverable D 7 – January 2009

67

Table of Contents

0 Executive Summary .............................................................................................................. 69

1 Introduction ........................................................................................................................... 70

2 Pavement engineering and practices ..................................................................................... 71

2.1 Concepts, abbreviations and notions.............................................................................. 72

3 Pavement deterioration factors.............................................................................................. 73

3.1 Traffic loads ................................................................................................................... 73

3.1.1 Traffic pattern.......................................................................................................... 73

3.1.2 Vehicles................................................................................................................... 74

3.1.3 Tyre configuration................................................................................................... 74

3.1.4 Tyre properties and studs ........................................................................................ 74

3.1.5 Consequences for EURODEX ................................................................................ 75

3.2 Pavement geometry ........................................................................................................ 75

3.3 Climate ........................................................................................................................... 75

3.3.1 Temperature ............................................................................................................ 75

3.3.2 Precipitation ............................................................................................................ 76

3.3.3 Freezing and thawing .............................................................................................. 76

3.3.4 Consequences for EURODEX ................................................................................ 77

3.4 Pavement materials ........................................................................................................ 77

3.4.1 Subgrade.................................................................................................................. 77

3.4.2 Unbound materials .................................................................................................. 79

3.4.3 Bituminous bound materials.................................................................................... 81

3.4.4 Cement bound materials.......................................................................................... 82

3.5 Factor interaction............................................................................................................ 83

4 Pavement distress .................................................................................................................. 83

4.1 Distress on flexible pavements....................................................................................... 83

4.2 Distress on rigid pavements ........................................................................................... 84

4.3 Distress on Semi-rigid pavements.................................................................................. 84

4.4 Consequences for EURODEX ....................................................................................... 84

5 Modelling of pavement performance .................................................................................... 85

5.1 AASHO test and the Fourth power law ......................................................................... 85

5.2 Pavement performance................................................................................................... 86

5.3 Statistical models............................................................................................................ 87

5.4 Mechanistic - empiric design ......................................................................................... 87

5.4.1 Flexible pavements.................................................................................................. 87

5.4.2 Rigid pavements...................................................................................................... 89

5.4.3 Semi-rigid pavements.............................................................................................. 89

5.5 Consequences for EURODEX ....................................................................................... 89

6 Maintenance and reconstruction............................................................................................ 89

6.1 Triggers for maintenance ............................................................................................... 89

6.2 Maintenance practices for flexible pavements ............................................................... 89

6.3 Maintenance practices for rigid pavements.................................................................... 91

6.4 Maintenance practices for semi-rigid pavements........................................................... 91

7 Discussion on construction and maintenance costs due to traffic......................................... 92

7.1 Importance of different maintenance activities.............................................................. 92

7.2 Costs of pavement deterioration caused by vehicles...................................................... 93

7.2.1 Flexible pavements.................................................................................................. 93

7.2.2 Rigid pavements...................................................................................................... 94

7.3 Summary of consequences for EURODEX ................................................................... 95

CATRIN - Deliverable D 7 – January 2009

68

8 References ............................................................................................................................. 96

9 Annex 1 – Distress types found on flexible pavements ........................................................ 96

10 Annex 2 – Distress types found on rigid pavements......................................................... 100

11 Annex 3 – HDM models of flexible pavement performance indicators rut depth and cracking .................................................................................................................................. 101

CATRIN - Deliverable D 7 – January 2009

69

0 Executive Summary

• Traffic loading is quite complex to describe but shows interesting possibilities and pitfalls in the development of more pavement friendly vehicles and more wear resistant pavements.

• Correct design and selection of maintenance treatments need to estimate the influence of the miscellaneous vehicles characteristics in fleet and overall traffic intensity.

• The climate varies across Europe and is setting the conditions for pavement design. Climate is by its own or by combined action with traffic deteriorating pavements. Climate is also of immediate interest due to effects of climate changes.

• The more, closer, heavier and dynamic loads at the surface, the greater is the risk of deterioration in the subgrade. The deformation behaviour of the subgrade strongly influences the mode of failure of the pavement as a whole.

• Permanent deformations in granular materials are very sensitive to stress conditions and stress history. Unfortunately, stress conditions and stress history present in the field and in ALT testing is very difficult to reproduce exactly in laboratory.

• When comparing field observations, ALT testing and laboratory testing, it is important to keep in mind effects of time, for example due to bitumen ageing (long term), healing (mid term) and viscoelasticity (short term).

• Special products with unique features, such as noise reducing porous asphalt concrete, are usually more expensive and often more sensitive, leading to a substantial influence on maintenance costs and associated reasons of deteriorations.

• In order to catch the true origin of pavement deterioration costs caused by vehicles, both common conditions as well as rarely appearing conditions with high rates of deterioration, need to be considered.

• Most distress mechanisms are not clearly associated with traffic. This means that they do not contribute to marginal costs for traffic in the manner that is suggested by the fourth power law. However, in some situations the fourth power law is certainly applicable and even not enough sensitive (enough value of power) to the detrimental effects of heavy vehicles. At least four stages of rate of deterioration have been identified in the report.

• Looking at the history of the fourth power law and the conditions for the AASHO Road Test is necessary to establish a more refined model for pavement deterioration.

• Performance measurements could be used to relate fundamental technical issues to performance for road users.

• The next step is to tie costs for constructing and maintaining a road with its level of performance, thus relating costs for maintenance and construction with the actual technical decisions made regarding design and long term maintenance strategies. This information that can be used in decision making based on Life Cycle Cost Analysis.

• The review of demands of pavement performance and means of assessing the road network reveals a great source of information not yet reached its potential for applications in this field.

Together with the statistical and mechanistic-empiric models for pavement performance and design, a foundation is laid for a new system for assessing the damaging power of vehicles.

However, large gaps in the knowledge structures exist. Some gaps, especially those relating field and laboratory observations, can be bridged by well organised ALT. More fundamental pavement design methods are essential in this case to be able to not only predict performance of yesterdays pavement design, but to invoke new and more efficient pavement designs into practice.

CATRIN - Deliverable D 7 – January 2009

70

1 Introduction

The EU-project CATRIN (Cost Allocation of Transport Infrastructure Cost) supports the European

Transport Policy, specifically to assist in the implementation of transport pricing for all modes of transport. Deliverable 7 is an outline of a new empirical road damage experiment and as a part of that, a review of current best knowledge of mechanisms that drive road deterioration is undertaken. The review is written for a popular audience in order to explain the important relationships that need to be taken into account when planning a large scale road damage experiment. Since we are aiming at a broader audience, the scientific depth is sometimes avoided, although the text is relying on sound, accepted scientific knowledge. To reach a popular audience, simplifications are sometimes made in theories and written presentations that tend to stretch the boundaries of common views among pavement engineers.

This review aims at summarising current knowledge on the vehicle and pavement characteristics that are important to take into account when studying costs of pavement deterioration. The report derive costs from a chain of consequences, cf. Figure 1 below, starting with factors that lead to deterioration such as traffic, climate and the pavement and subgrade itself. Then, models for pavement deterioration and subsequent needs for maintenance are depicted. Finally, the costs of different maintenance activities are investigated. This chain of consequences leading to costs for maintaining our road infrastructure is the key to understanding the mechanisms behind road user marginal costs.

Vehicles Design and Construction

5.

Traffic

3.1.

Climate

3.3.

Pavement and subgrade

3.2., 3.4.

Pavement deterioration

4.

Pavement performance

5.2.

Maintenance activities

6.

Road owner costs

7

Road user costs and benefits

Society

Figure 1: Vehicles generating costs for maintaining road network. Section numbers in italic. Arrows indicate direction of consequence.

Technical aspects of traffic, pavement deterioration and ALT testing to consider in analysis of costs for maintaining road network is given. The report also shows why Accelerated Load Testing (ALT) is necessary and why these tests need to be coordinated on a European level to gain potential of testing performed in different countries.

Costs related to choice of standard, rather than deterioration, are excluded from this report. Examples of costs related to standard are costs associated with road type, maintenance standard, response time to operational action, noise reducing measures, etc. These standards are mainly set based on total traffic flow (ADT) and location (rural or urban).

CATRIN - Deliverable D 7 – January 2009

71

Consequences and remarks of importance to planning of the EURODEX project is summarised after each section in the report and then finalised in the last section. Special emphasis is given to parameters related to the fourth power law and the foundation it is based on. The hypothesis is that the fourth power law is not generally applicable to all vehicle and pavement types. Or rather, from a scientific viewpoint, it is easier to test and falsify the alternative hypothesis, that the fourth power law is generally applicable. Alternative models for assessing damaging power of vehicles are discussed.

2 Pavement engineering and practices

Counted from the surface, pavements in general consist of a number of layers made of different materials for different purposes. A general rule of thumb is that the lower down in the material, the lower is the demand for performance of the material, which also means a reduction in price. Below is a brief description of each layer. Pavement design is the process in which the layer thickness and properties are selected by optimisation between economy and desired function. In practice, the most important design factors are related to climate, traffic loads and available materials.

Wearing course

Surface layer optimised to give surface properties such as wear resistance, skid resistance, visibility, macro texture against aqua planning and light reflections, handling climate action (lead off rain, withstand sun and heat) and distributing traffic and point loads. Asphalt or cement concrete.

Binder layer

Only on more trafficked flexible pavements. Optimised to withstand permanent deformations and distribute loads. Asphalt concrete.

Bound base layer

Optimised to distribute loads and withstand cracking. Asphalt or cement bound aggregate.

Unbound base layer

Distributing loads and being an even layer for placement of bound materials (otherwise the thickness of the bound layers will vary). Unbound aggregate of moderate size and high quality.

Sub base

Further distributing loads. Drain water and break water movements from below. Unbound aggregate, often from site. Often course crushed rock or other available aggregate/granular materials.

Protection layer

Not always needed. Solve problems with inferior subgrade materials by replacement down to a depth where risks for distress are acceptable. Can be on-site soils. Geotextiles may be used to avoid penetration of fine subgrade materials into a coarser sub base.

Subgrade

Existing in situ soils.

CATRIN - Deliverable D 7 – January 2009

Wearing course, binder and bound base layers

Subgrade

Base layer

72

Sub base layer

Figure 2: Layers in a flexible pavement.

Pavements are usually characterised as flexible, rigid or semi-rigid. The distinction between these types is very important when discussing deterioration, pavement design and maintenance activities and their associated costs. Flexible pavements are asphalt concrete or gravel roads, in which the materials to some extent can adjust to movements without failure. Rigid pavements are cement concrete roads.

Semi-rigid pavements are either asphalt concrete on cement bound base layers or concrete block pavements. The latter is common for esthetical and traffic behaviour reasons on city streets and for good resistance to point loads on port or airfield surfaces, but not used on the road network.

2.1 Concepts, abbreviations and notions

ADT

Annual Daily Traffic

ALT

Accelerated Load Testing. Pavement subject to large number of vehicles passes in short duration by purpose built equipment

Bitumen

End residue from distillation of heavy crude oils used in asphalt concrete (not to be confused with AE word for heavy oils)

Distress and deterioration

Damage and damage process, where damage is defined as an alteration of original properties with a future negative impact

Flexible,

Rigid and

Semi-rigid pavements

(or semi-flexible)

Flexible = Only unbound materials and asphalt concrete

Rigid = Cement concrete layer as wearing and base course

Semi-rigid = Not defined by above. For example asphalt concrete on cement base (a rigid layer) or concrete block pavements (rigid elements with flexing joints)

Fourth power law

The damaging power of one load is related to another by power four of the quotient. (Double the load means (2

4

=) 16 times more damage per passage)

Granular materials

Coarse stone materials, whose strength is dependent on stresses, cf. figure in Section 3.4.2. (opposite is cohesive materials such as clay)

CATRIN - Deliverable D 7 – January 2009

73

Plastic and

Viscous deformations

Resilient and permanent response

Stress and

Strain

Plastic deformations are permanent and governed by the level of loading (stress), often with a threshold (cf. soft metals that can be bent)

Viscous deformations are also permanent and governed by the time of loading. The load is dependent on the rate of deformation (cf. liquids)

Stresses and strains as a consequence of loading is jointly called response. Resilient response is associated with stresses and strains during a loading sequence after which the response is reversed after unloading. Permanent strains are left after a loading sequence

Stress = Load distributed over an area or cross section of a material body. If equal in all directions the same as pressure. Usually units N/m

2 which corresponds to Pa (Pascal)

Strain = Ratio of total deformation to the total initial dimension. Strain defines the amount of stretch, compression or distortion in a material body. Dimensionless

3 Pavement deterioration factors

In this section, factors responsible for deterioration of roads are discussed. These factors are related to traffic, climate, and pavement materials, geometry and construction. Furthermore, these factors often interact. Avoiding or mitigating future deterioration is the key to pavement design, and failure in this aspect will lead to increased maintenance costs. The factors and their implications will be further discussed below.

3.1 Traffic loads

Wear from vehicles is dependent on parameters that can be described from macro level down to a micro level, i.e. from general traffic intensity down the actual properties of the tyre in contact with the pavement.

3.1.1 Traffic pattern

The intensity of traffic is obviously very important to deterioration of pavements. Traffic intensity can be expressed as Annual Daily Traffic (ADT), Average Annual Daily Traffic (AADT) or vehicle passes per hour. Vehicle passes are measured in sections, for example if an investment in the road network is planned. Road lane configuration is of great importance together with subsequent division of vehicles in each lane. The time during the day for passes cannot be neglected in some cases. The reason is the viscoelastic properties of flexible pavements (high temperature softening and inadequate relaxation of stresses at low temperatures) and temperature related stresses in rigid pavements. A worst case example for a flexible pavement is a slope facing sun with afternoon traffic jams (low speed, much traffic and high temperatures).

Speed and changes in speed by braking and acceleration as well as turning (transversal acceleration) influences particularly the bituminous bound layers. The reasons are related to the viscoelastic properties, extra shear forces, dynamic loads and surface wear.

Lateral or transversal wander of vehicles, i.e. the distribution of vehicle positions in a lane, is very important to road deterioration. Lateral wander depends on factors such as lane width, distance to objects (e.g. guardrails), and speed.

CATRIN - Deliverable D 7 – January 2009

74

3.1.2 Vehicles

The weight of vehicles is a distribution from the smallest vehicles up to extreme transports of indivisible goods. The total weight of vehicles is a problem on bridges while it is less important on pavements since the load can be distributed on several axles and tyres. However, it is important to keep in mind that loads distributed over several axles and tyres will yet overlap deeper down in the pavement and subgrade, cancelling some of the effects of load distribution. Axle configurations and maximum axle loads are prescribed by EU directive 96/53/EG, as well as by national legislations. EU directive 96/53/EG allows maximum weight on drive shaft axle of 11.5 ton and other axles 10 ton.

Numerous maximum weights for bogies and vehicle configurations are also prescribed and may influence deterioration deep down in the subgrade, cf. section below on subgrade materials. Vehicle and axle loads have been measured during recent years, for example in relation to the European Bridge

Weight in Motion project (often referred to as Bridge-WIM), which has given new insights in the actual loading patterns on some specific roads.

Axle loads at travelling speed often substantially exceeds the mean static loads due to dynamic effects.

Longitudinal unevenness, sometimes in combination with the axle configuration, is one important factor causing vehicles to oscillate. The suspension of vehicles will in this case determine the resulting vertical forces.

3.1.3 Tyre configuration

Tyre positions and axle width will determine where the actual loads are applied onto a pavement cross section. Ruts caused by heavy vehicles are normally further apart compared to wear from cars, due to their wider axles. Narrow roads may be more sensitive to wide axles because the load is applied closer to a weak edge.

The axle load is normally distributed on two or four tyres on each axle. Pairs of tyres will distribute the load more evenly in the bound layers while the effect is negligible further down due to overlapping of stresses and strains. Loading from subsequent passing axles such as triple bogie may also be more severe than three passes with intermediate resting periods due to viscoelastic properties of asphalt concrete, cf. section on bituminous bound materials.

3.1.4 Tyre properties and studs

The actual properties of tyres are also of great importance, especially for the local levels of stresses and strains in the vicinity of the contact area.

• Tyre pressure, optimum

• Tyre pressure below or above optimum causing uneven stress distributions

- below: excess stresses on the sides

- above: smaller contact area and higher stresses

• Tyre tread in contact area

• Stiffness of tyre rubber

The difference in damaging power between for example double wheels and super single wheels has been much debated, since the super single wheels generally has higher pressure and greater contact area. In the energy saving debate another cause of pavement damage is spoken for. Higher pressures are put forward as a mean of reducing rolling resistance and save energy.

Studded tyres on cars are still used in many countries in colder climates. Studded tyres can be extremely detrimental to some stone material in wearing courses, which have lead to ban in many

European countries.

CATRIN - Deliverable D 7 – January 2009

75

3.1.5 Consequences for EURODEX

Traffic loading is quite complex to describe but shows interesting possibilities and pitfalls in the development of more pavement friendly vehicles and more wear resistant pavements. Legislations and observance of laws and consequences of overloading are not very well understood. The great effect of tyre properties on pavement deterioration is one example that shows potential savings for society. Tyre design and pressure have large influence on fuel consumption and thereby on CO2 and transport economy. However, raising tyre pressures on ordinary tyres to save fuel may be very detrimental to asphalt concrete. Modern suspensions are a good example of vehicle components which save pavements from unnecessary deterioration.

Correct design and selection of maintenance treatments need to estimate the influence of the miscellaneous vehicles characteristics in fleet and overall traffic intensity. The status of current knowledge on traffic loading is as follows: traffic is measured on a standard basis; vehicle and axle load measurements are in its infancy; while much other information is missing to get the whole picture.

Factors described in this section need to be carefully selected, controlled during testing and described in documentation in order to be able to interpret results for EURODEX.

3.2 Pavement geometry

The geometry of a pavement and its layers will influence both where the loads are applied, as mentioned in the previous section, and the stresses and strains developed in each layer. Generally, a flexible and rigid pavement consists of the following layers:

• Wearing course

• Bound base layer

• Unbound base layer

• Subbase

• Subgrade

Changes in pavement geometry should not be abrupt but smoothened over an adequately long distance so that differences in pavement geometry will not develop uneven characteristics of the pavement, for example showing as uneven surface or reflective cracking.

3.3 Climate

The climate set the conditions for pavement design. Parameters related to climate are generally collected by meteorological authorities but some data is also collected specifically for pavement design, maintenance and operations purposes. Models also exist to describe climate. An important issue is to relate climate parameters to pavement performance and deterioration to estimate and mitigate effects of global warming.

3.3.1 Temperature

Temperature is an important pavement deterioration factor in several ways. For rigid pavement, stresses from temperatures are so significant that they are taken into account in many design procedures. In flexible pavements, stresses from temperatures can be relaxed to some extent but may significantly contribute to the overall stresses, especially at low temperatures. Flexible pavements are generally more affected by temperatures compared to rigid ones because temperatures strongly influence the material properties of asphalt concrete. Asphalt concrete is influenced by temperature in the following ways

CATRIN - Deliverable D 7 – January 2009

76

• Low temperatures (cold winter periods)

Asphalt concrete is stiff and less able to relax stresses. Cracking is more likely while

permanent deformations are not.

• High temperatures (hot summer days and/or sunshine reaching pavement surface)

Asphalt concrete is softer and more viscous. Permanent deformation is more likely, both as a result of compaction and changes in geometry. The load spreading capacity of the asphalt concrete layer is also reduced due to low stiffness. The rate of ageing in asphalt concrete

increase exponentially with temperature.

• Intermediate temperatures

The most frequent temperatures will be important to the performance of pavements and the

distress developed such as rutting and cracking.

To set exact figures of temperatures for the intervals above is not possible since these temperatures will vary across Europe. The reason is that bituminous binder with different properties are used to account for the variation in temperatures, climate and traffic loads.

3.3.2 Precipitation

Water and moisture will strongly influence pavement deterioration in a number of ways:

• Stone loss on wearing course

Asphalt concrete is sensitive to water. There are numerous examples of where stone aggregate are stripped of bitumen in the presence of water, promoted by traffic, high

temperatures or frost-thaw cycles.

• Weakening of bound pavement layers or bonding between layers

Similar to above but maybe not noticed until holes are appearing.

• Weakening of unbound pavement layers

Fine materials, especially some minerals such as mica, are very sensitive to moisture. If sufficiently large amounts of fines are present in the unbound layers, the material properties

such as stiffness or propensity for permanent deformation will be affected.

• Loss of bearing capacity due to poor drainage or during flooding

Similar to above, at higher water content, pore pressures develop that reduces the strength of

unbound materials.

Water and moisture is generally considered very detrimental to pavements. Therefore, pavement engineers carefully seek to secure proper drainage of pavements as their number one priority.

3.3.3 Freezing and thawing

Freezing and thawing is of course related to the previous factors, water and temperature, but is described separately here due to the combined origin and specific effects. This description of freezing and thawing effects are in chronological order. Cycling between freezing and thawing may for many mechanisms cause greater damage. Some effects of freezing and thawing are:

• Frost expansion damage in concrete. Particularly after a large number of freeze - thaw cycles.

• Frost damage in stabilised layers.

• Frost heave in subgrade causing uneven surface (poor road performance) and cracking of bound layers

• Thawing on top of frozen soil may lead to heavy excess of water in subgrade and subsequent poor bearing capacity, if the released water cannot drain.

CATRIN - Deliverable D 7 – January 2009

• Frost and subsequent thawing reduces stiffness and resilience to permanent deformation in unbound materials and soils with fine material.

Deterioration due to freezing and thawing will, in some regions, be more severe than traffic loading, and selected as limiting design criteria.

77

3.3.4 Consequences for EURODEX

The climate varies across Europe and is setting the conditions for pavement design. Climate is by its own or by combined action with traffic deteriorating pavements. Climate is also of immediate interest due to effects of climate changes.

Factors described in this section need to be carefully selected, controlled during testing and described in documentation in order to be able to interpret results for EURODEX. Parameters such as moisture content and temperature strongly influence the performance of pavement materials.

3.4 Pavement materials

Properties of crushed aggregate or stone materials vary substantially throughout Europe. Many geographical areas lack coarse material of high strength and performance. Transportation of aggregate is very costly and energy consuming. Consequently, pavements around Europe are built in a variety of manners and respond quite different to traffic loading.

Bituminous as well as cement bound materials are subject to ageing, which alter their properties and set a time limit for use. The consequences are at least twofold: (1) there is an equivalent annual cost without traffic running on the pavement, (2) materials with properties altered due to ageing will not perform very well at some time in the future.

Road markings need a brief comment: they are affected by all kinds of traffic, especially vehicles with studded tyres. However, the deterioration of road markings is also dependent on road design and climate.

In the following sections, details on how pavement materials perform to carry traffic loads are given.

3.4.1 Subgrade

The levelled foundation for the pavement is called subgrade and consists of in situ or on site soil.

Inferior subgrade soil may be replaced or improved. The performance of the subgrade can also be improved by drainage and protection from climate actions.

By tradition, pavement engineering consider configuration of pavement layers (cf. section on pavement design) and geotechnical engineers consider settlements and slope stability. There might be a grey zone in between these disciplines in which the traditional design factors such as traffic loading affects the deterioration of the subgrade (mainly contribution to rutting and evenness). In an attempt to illustrate this, contributions to vertical stresses from the weight of materials and traffic loading is calculated by simplified models and assumptions based on an ordinary pavement and traffic loading

(25 kN load on each tyre). This simplified model shows that traffic load will generate substantial contributions to the stress levels in the pavement well below 2 meters; not accounting for other factors such as dynamic contributions to loads. Hence, the development of rutting in subgrade due to heavy traffic is substantial but maybe not getting the attention it deserves in the pavement design procedures.

Obviously, heavier traffic loads and unfortunate tyre configuration, together with dynamic loads, will increase the depth at which traffic loads become significant and rutting is occurring.

CATRIN - Deliverable D 7 – January 2009

78

0,5

1

0

0 20 40

1,5

2

Stresses from weight of soil and pavement

2,5

3

3,5

Load stress from

2.5 ton tyre

(Boussinesq)

4

Vertical stress [kPa]

60 80 100 120

Total stress, one tyre

Total stress, two tyres

Total stress, three tyres

140

4,5

5

Figure 3: Simplified illustration of contribution from static traffic load and pavement weight on vertical stresses at different depths.

Although geotechnical problems is generally a matter for static analysis not including effects of heavy vehicles, there are sensitive soils that may be severely affected by dynamic loads and vibrations. In this case, heavy vehicles may cause speed restrictions or costs for securing the area as well as even putting people and built assets at great risk.

Cebon summarised a model of the relationship between incremental permanent deformation (

δ) and vertical strain (

ε) in subgrade and unbound granular materials during one axle passage as [1]:

δ

i

=

L

1

⋅ ε

i

L

2

(1) where L

1 and L

2

are material constants. Cebon refer to a PhD thesis (Hardy, Cambridge, UK, 1990) in which a review is done showing L

2

to vary within the range 1.85 to 7.14, depending on pavement design and test conditions. Later, Cebon concludes that for thin asphalt pavements (less than 125 mm), the mode of failure is determined by the L

2

(L

2

. A low power (L

2

= 1) means fatigue while at a high power

= 3.5), the predicted mode of failure is rutting. Furthermore, Cebon concludes that pavement life becomes very sensitive to variations in asphalt concrete thickness at high power (L

2

= 3.5).

EURODEX remarks

Above is shown how distress from several separate loads at the pavement surface will overlap further down in the pavement or subgrade. However, at some level, the pavement and subgrade weight will take over so that no effects of traffic can be expected. The more, closer, heavier and dynamic loads at the surface, the greater are the risks of deterioration in the subgrade. The deformation behaviour of the subgrade strongly influences the mode of failure of the pavement as a whole and the subsequent needs of maintenance.

CATRIN - Deliverable D 7 – January 2009

79

3.4.2 Unbound materials

Unbound materials in pavements are normally crushed aggregate or uncrushed stone material taken from site or transported from quarries, so called granular materials. In some areas of Europe, coarse granular materials are scarce and finer materials such as sands are used instead. From a performance point of view, unbound materials are fundamentally dependent on its gradation (size distribution), particle properties, water content and surrounding pressure on the material (confining pressure, which is function of previous compaction and support from surrounding materials and above loading).

From a heavy load point of view, it is of particular interest to investigate how heavy loads influence the development of permanent deformations and, consequently, the important contribution from unbound layers to rutting.

Several researchers have concluded that at low levels of stress, an equilibrium level of permanent strain can be reached. At higher stresses, however, permanent strains may gradually increase until failure is reached in the material (i.e. rapid growth of deformations). It appears that unbound granular materials experience a load threshold level, as reported by many researchers. The so called

“Shakedown theory” is an analytical approach to applying a limiting stress condition to calculation of

permanent strains. Lekarp and Dawson [2] extended the approach to modelling of granular materials,

cf. Figure 4 below. The figure illustrates the sensitivity of unbound granular materials to stress conditions under traffic loading and the presence of a threshold limit for loading with respect to development of permanent deformations contributing to rutting. Stresses applied to a sample in loading cycles with slightly different stress paths (a.) as defined from ordinary normal stresses (b.) generates very different permanent strains and deformations during cyclic loading (c.) so that a threshold level defined by the peak q/p is indicated (d.). Not shown in the figures are the negligible effects of loading with stress paths in the lower left quarter of Figure 4a. Even though peak q/p is high, the stress levels are not great enough to cause any damage to the material. Consequently, the threshold level also need to consider the level of the stresses (not just the ratio q/p but also q and p themselves).

A major drawback of the above methods of testing is that loading is only orientated in fixed directions and not allowed to rotate as happens in practice when a wheel is passing. If one tries to compact a granular material one should apply forces in varied, kneading manner.

CATRIN - Deliverable D 7 – January 2009

800

700

600

500

400

300

200

100

0

0

I.

II.

III.

IV.

100 200 p [MPa]

300 400

a. Stress paths I-IV in diagramme deviator stress (q) vs. mean normal stress (p) (or confining pressure).

σ

1 b. Definition of p and q from normal stresses (

σ

1

and

σ

3

).

σ

3

500

c. Development of permanent strain during load cycles.

σ

σ

1 q =

σ

1

− σ

3 p = (

σ

1

+ 2σ

3

)/3

3 d. Permanent strain at 20 000 cycles divided by stress path length vs. ratio of maximum q over p.

80

Figure 4: Dependency of stress conditions on development of permanent deformations and the load threshold limit, after Lekarp and Dawson [2] and Lekarp [

3

].

The above fundamental description of behaviour of granular materials cannot be directly converted into a fourth power law without major simplifications but points out several important features of unbound granular materials. Unfortunate combinations of high loads or improper load configurations, weak or damaged bound layers and subgrade, and conditions in the unbound material itself may cause severe damage to a pavement. It is for example the reason why pavements with steep ditch slopes cannot be loaded by trucks near the edge without edge faults (no pressure holding back from pavement side). Overloading at the wrong time in the wrong place may lead to severe damage far beyond any fourth power law. However, on high strength, high volume roads, the likelihood of being close to the threshold levels are probably very low and in that case associated with an already developed damage.

EURODEX remarks

Permanent deformations are very sensitive to stress conditions and stress history, as indicated in the

Figure 4 above. Unfortunately, stress conditions and stress history present in the field and in ALT testing is very difficult to reproduce exactly in laboratory. Models are therefore simplified. However, these simplified models clearly identify severe conditions that will lead to permanent deformations and that further loading will result in rapidly increasing deformations (which in turn will show as rutting). For EURODEX purposes, it is of particular importance that material properties and actual stresses and strains are accurately selected, controlled and documented. Otherwise, the EURODEX results will not be comparable. The figure above also illustrates the risk of overloading, which need to be further investigated.

CATRIN - Deliverable D 7 – January 2009

81

3.4.3 Bituminous bound materials

Bituminous bound materials consist of aggregate, filler, bitumen and sometimes different types of additives. These materials are in popular parlance called “asphalt concrete” while in practice, the products and methods used are actually quite broad and complex. Any further technical details are not given here but it is important to state that performance of different types of “asphalt concrete” differ quite a lot and that extra performance costs accordingly (as always). For performance reasons, bituminous bound materials for pavement construction are usually divided into:

• Wearing course – the surface layer with the purpose of withstanding wear and climatic action, as well as creating a surface with good characteristics regarding friction, drainage, noise and visual properties

• Binder course – an intermediate layer which can withstand deformations

• Base course – bottom asphalt concrete layer which can withstand both deformations and numerous cyclic loads (fatigue).

Bitumen is affected by ageing, which gradually alters the material properties, mainly as a consequence of bitumen becoming stiffer. The rate of ageing depends on temperature and exposure to air during the history of production and service. Asphalt concrete for low volume roads is designed slightly different to withstand effects of ageing and climate rather than effects of traffic.

Fatigue and rutting are two deterioration mechanisms occurring directly in the asphalt concrete as a consequence of traffic. There exist numerous publications on both issues. A review is impossible to fit into this context and the follow only aims at picturing mechanisms.

Fatigue in asphalt concrete is often described in terms of number of cycles to failure at a certain stress or strain level. Typically, after laboratory testing of asphalt concrete, linear relationships are obtained for the parameters log (N = number of cycles) vs. log (

ε or σ = strain or stress level). Reported slopes

(b) of this relationship is usually between 3 and 5 (i.e. log N = b

⋅log ε + a, where a is a constant). With some mathematics one can conclude that

N

=

a

ε

b

(2) which is a direct parallel to the fourth power law (if b = 4). It appears that the fourth power law is feasible for pavements which are maintained due to fatigue deterioration. Another proof of the effects of cycling loading on asphalt concrete is given by fracture mechanics. Fracture mechanics is a discipline in science where the development of cracks in materials during repeated loading are modelled and explained. It has been theoretically shown by different researchers that b is between 3 and 4 [

4

,

5

,

6

]. These models also show some of the complex interactions between the stiffness of the asphalt concrete, stiffness of subsequent layers, asphalt concrete layer thickness as well as loading characteristics.

Rutting on the other hand show a different response to traffic loading, compared to fatigue. In the recently implemented US M-E Design Guide, the following relationship between resilient strain (non permanent),

ε r

, and plastic strain (permanent),

ε p

[

7

]:

ε

ε

r p

=

k

1

10

3 , 4488

T

1 .

5606

N

0 .

479244

(3) where k

1

is a parameter dependent on asphalt concrete thickness, T is temperature and N is number of vehicle passes. Keeping these parameters equal, one can conclude that the permanent deformation is linear to the resilient response, which is linear to the traffic load. Hence, there is no fourth power law in this case. It is no power at all, but a linear relationship between axle load and permanent deformations leading to rutting.

Asphalt concrete is viscoelastic, which means that its response to load is time dependent and develops a time history. Asphalt concrete is also believed to heal some damage over a period of time. Therefore,

CATRIN - Deliverable D 7 – January 2009

82 there is a slight difference between for example passes of three single axles and one pass of a triple axle with the same axle loads. This is illustrated in Figure 5 below where the increasing top transversal strain should be noted. Another consequence of viscoelasticity is the considerable time needed for the transversal strain to return to near its original level (which is never totally reached).

Figure 5: Calculated strains at the bottom of asphalt concrete layers during passing of triple bogie [

8

].

The deterioration mechanisms on noise reducing wearing courses may be slightly different. Particles clog the pores of some types of noise reducing wearing courses (porous asphalt) and wear will reduce the service life of grooves with noise reducing abilities. Wear from studded tyres are particularly troublesome.

EURODEX remarks

Asphalt concrete is used in layers with different performance conditions and distress types. When comparing field observations, ALT testing and laboratory testing, it is important to keep in mind effects of time, for example due to bitumen ageing (long term), healing (mid term) and viscoelasticity

(short term). Asphalt concrete deterioration can to some extent be modelled but may be strongly influenced by the structure as a whole. However, more accurate predictions are not possible yet but research is ongoing regarding distress mechanisms such as permanent deformations and fatigue.

Special products with unique features, such as noise reducing porous asphalt concrete, are usually more expensive and often more sensitive, leading to a substantial influence on maintenance costs and associated reasons of deteriorations.

3.4.4 Cement bound materials

Similar to asphalt bound materials, cement bound materials are used in wearing courses and base courses. Furthermore, cement, lime or other hydraulic binders can be used to improve properties of subbase and subgrade materials. Cement concrete has no relaxation abilities and consequently no viscous deformations. Flexible joints at sufficiently short spacing are needed to account for the always appearing deformations in lower layers and keep internal stresses at an acceptable level. Temperature variations and vertical gradients are one source of internal stresses. Cement concrete may also be subject to fatigue and ageing. Fatigue due to temperature and traffic loads is used as a pavement design criteria. Ageing and durability is an important topic since rigid pavements usually are demanded a long service life to be cost efficient compared to flexible pavement with much shorter maintenance intervals.

CATRIN - Deliverable D 7 – January 2009

83

3.5 Factor interaction

Interaction between factors related to traffic, climate and pavement structure is needed to explain some of the deterioration showing on pavements. Development of new pavement design and maintenance practices should also go hand in hand with development and legislations regarding vehicles, especially heavy vehicles. From the point of view of creating an ALT testing programme, it is of course of great interest to handle co-variance. Pavement deterioration is generally not suitable for determining a marginal cost per extra vehicle passing. Critical situations and periods may be responsible for a substantial part of the deterioration. Since pavement management generally assess the pavement surface condition, at the best, once a year, it is difficult to quantify the importance of these critical events.

In combination with heavy traffic, some critical conditions for flexible pavements are:

• Cold (stiff) surface and wet (weak) subgrade. Often the case during thawing. Great risk for cracking of asphalt concrete.

• Hot (soft) asphalt concrete and wet (weak) subgrade. Often the case during rainy summers on poorly drained pavements. Great risks for excess rutting.

• Slow traffic on hot (soft) asphalt concrete on pavements designed for low or medium volume traffic.

One example of critical condition for rigid pavements is:

Rapidly heated

surface (creating a large temperature difference to lower layers) and heavy traffic. Slabs are poorly supported in the centre and experience great stresses.

EURODEX remarks

In order to catch the true origin of deterioration costs caused by vehicles, both common conditions as well as rarely appearing conditions with high rates of deterioration, need to be considered.

4 Pavement distress

All pavement distress types are subject to maintenance activities and subsequent costs. Understanding distress mechanisms enables tracking of costs, but more important, reveal means of avoiding costs.

4.1 Distress on flexible pavements

In annex 1 is a review presented of common distress mechanisms found on flexible pavements, more or less common knowledge among pavement engineers. The purpose of this review is to tie pavement distress to their underlying causes, which is a crucial part of assessing the damaging power of vehicles and subsequent marginal costs. This presentation is by necessity quite long since one of the purposes is to show the great complexity, which need to be considered in a future EURODEX. Therefore, the review is put in annex. From Annex 1, rutting and cracking is selected as being caused by heavy traffic. Wear is caused by traffic in general. Additionally, many types of distress can be related to poor production, resulting in pavements with subsequent poor bearing capacity or resistance to traffic wear.

However, most distresses are probably related to traffic in one way or another. Suburban streets with very low traffic volumes tend to last more than 30 years.

The importance of different distress mechanisms is difficult to determine. It is even more difficult to estimate how today’s changes in pavement design and maintenance practices will influence the future, especially in the light of inadequate maintenance budgets.

CATRIN - Deliverable D 7 – January 2009

84

4.2 Distress on rigid pavements

Rigid pavements are not the main focus of this work. However, a brief description of distresses is given in Annex 2.

A major difference between rigid and flexible pavements is that rigid pavements show hardly any deformations in the bound layers. This may appear as a beneficial property only, but in practice, rigid pavements lack flexibility and needs more attention to deformations or movements in the pavement and subgrade to avoid cracking. Rutting on concrete pavements is mainly due to wear of the surface.

Longitudinal uneveness is mainly attributed to initial uneveness and development of uneveness as allowed by the joint design, given that no unexpected faults are developing in the slabs (continuously reinforced concrete pavements being an exception since no joints exists). Joint design is consequently of great importance with parameters such as joint spacing, restrain of vertical displacement by dowel bars and rebars, as well as sealing of joints. Dowel bars allow horizontal movements but restrain shift of adjacent slabs to avoid typical concrete pavement bump effects. Sealing of joints is beneficial to protect lower layers from water related distress. Cement bound base layers may be sensitive to water

(especially in combination with frost). Free water underneath the concrete slabs may lead to pumping of fines under the impact of heavy traffic.

The distress caused by traffic is difficult to single out. These distress mechanisms are considered in the design and construction process and any premature failure is caused by interacting deterioration factors.

4.3 Distress on Semi-rigid pavements

Semi rigid pavements are often associated with reflective cracking. Semi-rigid pavements can include a number of different setups. The traditional semi-rigid pavement consists of an asphalt concrete layer on top of a cement bound base layer. A more rigid cement bound base need to develop cracks to account for shrinkage in the continuous layer. These cracks may reflect up and through the asphalt layer. Softer cement bound base layers may not be stable enough to account for heavy traffic.

4.4 Consequences for EURODEX

As can be observed above, most distress mechanisms are not clearly associated with traffic. This means that they do not contribute to marginal costs for traffic in the manner that is suggested by the fourth power law. However, in some situations the fourth power law is certainly applicable and even not enough sensitive to the detrimental effects of heavy vehicles, for example on pavements with poor bearing capacity.

In an attempt to summarise the findings so far, four stages of deterioration is identified, depending on the loading situation and pavement conditions:

• Negligible distress from vehicles (climate effects etc. overshadow)

• Linear relationship – i.e. Rutting in bound layers on strong foundation

• “3-5” power law – i.e. Fatigue in bound layers and acceptable levels of loading in unbound layers.

• Threshold level (“Large” power law) – i.e. loading during unacceptable conditions such as overloads, bad tyre and load configurations in combination with for example poor drainage, thawing or cracked pavements.

This means that the marginal cost for a given vehicle will differ substantially on a strong pavement compared to a thin pavement designed for a low volume road. Another consequence is that increased traffic, traffic loading, poor maintenance etc. (i.e. either increasing load or decreasing bearing capacity) may shift the rate of deterioration to become more severe. A pavement designed to be adequately strong in one decade may be considered weak in the next due to changes in traffic, unexpected damage or poor maintenance.

CATRIN - Deliverable D 7 – January 2009

85

5 Modelling of pavement performance

5.1 AASHO test and the Fourth power law

The origin of the fourth power law is the AASHO (American Association of State Highway Officials) road test. The test road was constructed between August 1956 and September 1958. Then test traffic was subject to the road between October 1958 and November 1960 [

9

].

The hot mix asphalt (HMA) mixes used at the Road Test consisted of crushed limestone coarse aggregate, natural siliceous coarse sand, mineral filler of limestone dust and a penetration grade bitumen of 85-100 pen. The pavements were constructed on clayey soils. The total thickness of bitumen bound layers (asphalt concrete) was between 2,5 cm (1 in.) and 15 cm (6 in.). Comparing with ordinary pavement designs, these pavements would be considered thin or medium. Thin pavements are likely to show distress during accelerated loading, which leads to observable results.

Consequently the underlying distress mechanisms in the AASHO Road Test were associated with high levels of stresses and strains in pavement materials and in the subgrade. Fatigue of asphalt materials after repeated loading is one of the major distress mechanisms appearing in thin asphalt pavements.

In the original equations modelling Equivalent Standard Axle Loads (ESALs), the so called Structural

Number (SN) was also included, indicating that the bearing capacity of the pavement was also a factor

to consider. However, the effect of SN was somehow odd and has subsequently been lost [10].

Table 1: Outline of the AASHO Road Test (axle loads in lb and thickness in inch). 1 lb = 0,454 kg,

1 in. = 2,54 cm.

CATRIN - Deliverable D 7 – January 2009

86

EURODEX remarks

Looking at the history of the fourth power law and the conditions for the AASHO Road Test is necessary to establish a more refined model for pavement deterioration. It is important to pin point

valuable experiences and identify shortcomings in today and future applications. In [11], some

criticism against the application of the AASHO Road Test to derive Load Equivalency Factors (LEF) today is:

• In accelerated tests, environment, age and mixed traffic patterns are not considered.

Furthermore, a limited number of pavement designs were constructed on the same soil in one climate.

• It did not consider vehicle characteristics (suspension, tyres, axle config etc.), which have also changed significantly since the test. Dynamic effects, loaded steering axles, and tridem axles other vehicle related topics not taken into account.

• Lateral distribution was not considered and is important for (both flexible and: authors comment) rigid pavements.

• Pavement design has significantly departed from the practice used at the time of the test.

• The LEFs derived from AASHO Road Test has not been shown to be applicable to specific distress elements, such as rutting.

• Pavement type and structure is needed information in a model for LEF. This is excluded in the simplified form (the fourth power law).

5.2 Pavement performance

To achieve pavement performance is the goal of pavement construction and maintenance. Examples of pavement performance related factors are:

• Traffic safety

- skid resistance

- low risk of aqua planning

- predictability and acceptable levels of evenness

- visibility

• Transport economy, efficiency and quality

- evenness

- rutting

- local faults

- stone loss

- aesthetics

The efforts needed to maintain acceptable levels of service differ substantially depending on the amount of traffic (related to preferred standard and rate of deterioration), type of road (standard and width), type of pavement, climate and geographical location etc. Level of pavement performance should be chosen based on road user and road owner costs, and road user benefits.

Pavement performance is measured on a regular basis by various equipment from which a number of performance related parameters can be generated, so called performance indicators. Today, road networks can be measured at traffic speed by mounted laser cameras (continuously measuring cross sectional profile), profilometers (instruments keeping track of evenness) and sometimes video cameras to catch images for further processing of for example incidence of cracks. Friction (skid resistance) is measured by specific equipment. Additionally, research and development has created a variety of more fundamental performance indicators which more accurately can be related to both performance required by road users and fundamental pavement design parameters.

CATRIN - Deliverable D 7 – January 2009

87

EURODEX remarks

Performance measurements covering road networks are performed on a regular basis. Consequently, today a vast amount of information exists regarding pavement surface characteristics present in the field, which could be used extensively to relate fundamental technical issues to performance for road users.

The next step is to tie costs for constructing and maintaining a road with its level of performance, thus relating costs for maintenance and construction with the actual technical decisions made regarding design and long term maintenance strategies. This information that can be used in decision making based on Life Cycle Cost Analysis.

5.3 Statistical models

Statistical models of pavement performance are often used to plan activities, select maintenance treatment and estimate future expenditures. Statistical models of pavement performance relate observable features related to a road with performance indicators such as rutting and extent of cracking. Annex 3 contains examples of statistical models, the Highway Design Manual (HDM) models of rutting and cracking. Similar models exist in great numbers, developed in different countries with different pavement designs, climates and traffic characteristics. These models have a few things in common. They relate to traffic, structural strength of pavement (bearing capacity) and involve a number of calibration factors.

The purpose of enclosing HDM models in Annex 3 is to show the structure of statistical models. The

HDM model for rutting divides the different origins of rutting and relate them to each underlying factors. In this case, the origins of rutting are initial densification, structural deformation, plastic deformation (permanent deformations in asphalt concrete) and wear.

Statistical models are easy to use, but lack the ability to handle new practices or changes in conditions.

Statistical data on current roads are also biased due to human thinking to avoid problems of deterioration. As an example, statistical models can rarely be used to correlate performance to subgrade properties since the subgrade is an important input in pavement and geotechnical design.

That is why more fundamentally based design models are needed, such as mechanistic – empiric design models.

5.4 Mechanistic - empiric design

5.4.1 Flexible pavements

According to COST 333 [12], the majority of European countries have adopted mechanical-empirical

based design procedures. In principle, these procedures follow the same procedure as outlined below, but they are not incrementally calculating continuously accumulated damage per load cycle. Instead, these procedures directly calculate pavement life based on a performance model, which relates to stresses or strains. For a specific pavement geometry, a mechanical model is used to calculate stresses or strains induced by wheel loads in a response model in which material properties are given as a function of parameters such as climate. These responses are used to relate to performance or design life in a performance model. Empirical data from in-service roads are used to calibrate mechanically determined design criteria in the performance model. Finally, the calculated performance is compared to the desired need for performance.

A survey of pavement design criteria used showed that most European pavement design guides used criteria for base course fatigue cracking and rutting in the subgrade while few considered rutting in bituminous and granular layers. Three countries used Serviceability index, which is a weighted rating from different performance related variables or condition ratings. Surface cracking and low temperature cracking was considered by only one country each.

CATRIN - Deliverable D 7 – January 2009

88

Performance models are the key to successful pavement design. Since the relationships between factors leading to road deterioration and development of pavement performance are very complex, the simplifications and empirical calibrations made in traditional pavement design are radical and inclusive. Today, much research exists that could be implemented to more fundamentally describe pavement deterioration. Below is a more fundamental procedure outlined for pavement design by

COST 333, which is based on fundamental deterioration models describing how traffic loads incrementally damage pavements during their service life. Each distress mechanism needs to be treated in parallel in order to catch the relationships between different types of distress, especially at later stages of deterioration.

Δt

Geometry Traffic

Response model

Stresses or strains

Performance model

Damage /cycle,

ΔD

Climate

Material properties

Cumulative damage, D

Figure 6: Outline of fundamental procedure for pavement design suggested by COST 333 [12].

Another drawback of today’s pavement design methods is the lack of connections to future performance and maintenance needs. From a life cycle cost and society point of view, cost benefit analysis of investments may need to consider more aspects than number of heavy vehicles, and focus more on the actual performance / services generated by the road investment. This line of thinking is also making marginal cost calculations for different road users easier and benefits of different pavement performance levels clearer.

Additional to pavement design with respect to traffic, a number of other aspects are taken into consideration. For example effects of frost and thawing, material properties (durability, heat transfer properties, water sensitivity, water permeability, risk of bleeding), pavement surface properties

(friction, macro texture, colour) etc. as well as economical and environmental considerations.

EURODEX remarks

More fundamentally based design models are developed which will enable also more refined marginal cost estimations. Still, there is long way to go before these more mechanistic models will enable accurate predictions without a great empirical material to back up performance models. This problem reflects back on the previous sections showing the complex relationships between different parameters leading to pavement performance and distress patterns. A complete mechanical analysis is not reasonable to expect and a great deal of empirics is needed. ALT testing is in this context a very valuable tool to bridge between laboratory and field observations.

CATRIN - Deliverable D 7 – January 2009

89

5.4.2 Rigid pavements

In this presentation of rigid pavement design, only differences are considered in comparison with flexible pavement design, relevant to the topic of the report. The overall layer design and pavement design procedure outlined above may be applicable also to rigid pavements.

The design criteria and, of course, the construction of each layer differ compared to flexible pavements. Common design criteria concern fatigue due to temperature and traffic induced stresses.

Examples of rigid pavement construction types are:

• Jointed plain concrete pavement (joints every 4.5-9 m)

• Jointed reinforced concrete pavement (joints every 4.5-9 m)

• Continuously reinforced concrete pavement (no joints but microcracks)

5.4.3 Semi-rigid pavements

The challenge in designing and constructing semi-rigid pavements is to avoid, on the one hand, durability problem, or on the other hand, cracking of the rigid base layer. It is merely a problem of mix design and production control. If too stiff, the rigid layer will crack. If too soft, problems with durability from combined effects of climate and traffic will cause deterioration.

5.5 Consequences for EURODEX

The criticism against the AASHO fourth power law shows the need to find another foundation for assessing the damaging power of vehicles. The review of demands of pavement performance and means of assessing the road network reveals a great source of information not yet reached its potential for applications in this field. Together with the statistical and mechanistic-empiric models for pavement performance and design, a foundation is laid for a new system for assessing the damaging power of vehicles. However, large gaps in the knowledge structures exist. Some gaps, especially those relating field and laboratory observations, can be bridged by well organised ALT. More fundamental pavement design methods is essential in this case to be able to not only predict performance of yesterdays pavement design, but to invoke new and more efficient pavement designs into practice.

6 Maintenance and reconstruction

6.1 Triggers for maintenance

Maintenance activities are triggered by observations of inferior performance or great risk for future inferior performance. These critical observations may not cover an area as a whole but decisions may be that a whole road link is maintained anyway for reasons of efficiency and rationality. Sometimes, different kinds of maintenance can also be co-ordinated to reduce costs and traffic disturbance.

Observations can be manual or judged from performance indicators. Typical indicators are IRI

(International roughness index) and rutting. Using models for IRI and rut development, such as implemented in HDM IV, many maintenance activities can be planned well ahead.

6.2 Maintenance practices for flexible pavements

Maintenance practices for flexible pavements can be categorised in localised, surface and strengthening treatments. During reconstruction the serviceability of an existing pavement is restored when its remaining structural life is insufficient [

13

].

Examples of localised treatments are:

CATRIN - Deliverable D 7 – January 2009

90

• crack sealing,

• patching and

• pothole filling.

Generally, these treatments are not costly but rather cost saving for both road users and road owners under the right circumstances. Obviously, localised treatments are used on local damages where not the entire surface is damaged.

Inlays are a special case in which the material in the wheel paths is replaced down to a certain depth.

The replaced material can either be pronounced to cracking or permanent deformations.

Examples of surface treatments are:

• surface dressings (single and double surface treatments),

• coat seals,

• milling (both texturing and increasing evenness),

• heating,

• remixing (hot, warm and cold),

• repaving and

• thin asphalt overlays and friction courses.

These treatments are selected when the distress is localised at the surface, for example low friction, inadequate macro texture, reasonably low degree of rutting and unevenness. In this case, cost of materials is becoming important. Therefore, the costs of surface dressings are much lower compared to new overlays. Thin asphalt overlays and recycling methods are not contributing substantially to bearing capacity but can account for rutting, depending on the depth of treatment. Table 2 below show how treatment life and costs differ between different treatments in the US, which probably holds as a general picture also in Europe. An open question of great importance for long term sustainable pavements is how different treatments influence the long term performance of pavements, e.g. if the treatments contribute to ageing of materials or if they are able to rejuvenate/heal materials or reduce rate of ageing. With more knowledge, surface treatments could become a powerful ingredient in long term maintenance strategies.

Table 2: Cost and life of common maintenance treatments in the US reviewed and summarised from several studies [

14

]. (Mean of running prices 1995-2002)

Preventive Maintenance Treatment

Crack Sealing

Thin Overlay

Chip Seal (Single)

Chip Seal (Double)

Microsurfacing

Cold In-Place Recycling

Ultrathin Friction Course

Fog Seal

Slurry Seal

Cape Seal

Scrub Seal

5

7

1

1

6

1

Treatment Life (years)

Cost per Lane Mile

Min Average Max

2

2

4.4

8.4

10

12

$5,300

$14,600

1

4

4

5.9

7.3

7.4

12

15

24

$7,800

$12,600

$12,600

10.6

9.8

2.2

4.8

9.8

3.7

20

12

4

10

15

8

$17,700

$31,100

$2,200

$6,600

$16,700

$5,800

Examples of strengthening treatments are thick asphalt overlays (>40 mm) and remixing plus, The cost of these activities are substantially higher compared to surface treatments. However, their contribution to increased pavement strength and consequence of reduced needs for surface repair must be considered in the long run, not just their rehabilitation of surface performance and their service life

CATRIN - Deliverable D 7 – January 2009

91 to the next maintenance activity. In principle, increasing traffic and vehicle loads should be accompanied by corresponding pavement strengthening, if a road cannot be assumed to be taken out of service in a time span of a few maintenance cycles. Pavement strengthening by additional overlays has previously been common when asphalt concrete has been laid during maintenance activities. However, future increasing bitumen prices and recycling efforts may, together with pavement level restrictions, lead to generally lower rate of increasing bearing capacity (due to increasing asphalt concrete thicknesses). Another aspect of strengthening by overlays is that bearing capacity increases before severe cracking occur, which is more difficult to rehabilitate in the bottom asphalt layers. There are several other factors that restrict the suitability of simply adding new overlays.

Reconstruction refers to all works that require the re-specification of the surfacing and road base layers. Reconstruction can either be aimed at restoring an old pavement to its intended standard or improving its standard in terms of bearing capacity and geometry (e.g. widening). Reconstruction is needed when the capacity of the layers beneath the surface layers are inadequate. Inadequate capacity can be defined as when it is cheaper to reconstruct compared to maintain often (including effects on road users and society). Costs of reconstruction can be in the range of new construction depending on the seriousness and depth of distress, and also related to the site conditions such as traffic situation on site. However, the expected service lives of reconstruction activities are in the range of new construction, which consequently lower their equivalent annual cost.

Generally, costs for maintenance are generated from needs of material, equipment and personnel for each type of maintenance activity. The total costs also involve the maintenance interval, i.e. the service life of each activity. The needs for resources are related to the depth of the treatment, i.e. compare surface treatment, 40 mm overlay and reconstruction. Pavement management means to monitor pavements and plan maintenance activities in an optimum way, i.e. cost efficient, timed, fulfilling performance standards for road users and long term sustainable; all included in a maintenance budget.

Planning of pavement maintenance activities need to consider a number of aspects but are obliged to fulfil budget restrictions and current pavement performance standards, which are both short term goals. Consequently, there is no strong driver in pavement maintenance to improve long term performance of the road network, e.g. increase bearing capacity.

EURODEX remarks

A concluding remark is that the costs of different treatments are closely related to the required amount of materials and the depth at which the origin of distress is localised. Damages due to increased loads or poor bearing capacity are consequently very expensive. However, their equivalent annual cost may soon become lower compared to frequently reoccurring activities on the pavement surface.

6.3 Maintenance practices for rigid pavements

Maintenance on rigid pavements can be divided in activities on the surface, at joints, in the concrete slabs and actions below the concrete layer. Activities on the surface include grinding for better friction or even out ruts as well casting thin overlays. Joints are carefully maintained both to keep detrimental water to penetrate and to keep slabs in place to avoid inferior performance and concrete damage. Parts or whole slabs can be replaced by casting if more severe cracking or damage appears.

6.4 Maintenance practices for semi-rigid pavements

Maintenance is often associated with reflective cracking if cracks develop to become broad. Cracks can be sealed. New overlays are often placed on top of the old pavement. In that case, measures are taken to break the emergence of reflective cracks by invoking layers to even out deformations over a larger zone.

CATRIN - Deliverable D 7 – January 2009

92

7 Discussion on construction and maintenance costs due to traffic

The aim of this discussion is to derive relationships between vehicles and maintenance costs that can be used in a EURODEX outline. First, a brief review of maintenance expenditures is presented.

7.1 Importance of different maintenance activities

Table 3 below shows the share of pavement types in European countries. Flexible pavements are the majority, as expected, but rigid and semi-rigid pavements play an important role in some countries and possibly an even greater role if share had been calculated based on transport work. Rigid pavements may be more common on highways compared to less trafficked roads. It should be noted that the figures concern the primary road network only.

Table 3: Share of pavement types in European countries [12].

In COST 324, maintenance expenditures with regard to distress mechanism were collected in a number of countries, cf. Table 4 below. In some categories, the variations between countries are very large, e.g. expenditures for transversal profile and surface layer cracking. One reason for variations is of course the problem of collecting and sorting data. However, variations are expected due to differences in traffic, climate and traditions of pavement construction and maintenance, including budget restrictions, considering the facts given in previous sections. Also in this case, the figures originate from the primary road network with high traffic volumes.

CATRIN - Deliverable D 7 – January 2009

Table 4: Maintenance expenditures on European road networks [15].

93

7.2 Costs of pavement deterioration caused by vehicles

Based on the facts presented so far in this work, the picture of marginal costs for vehicles is scattered. It is however clear, that the type of pavement, the pavement design and quality of construction is very important, perhaps the most important factors, in determining the level of maintenance needs, and hence total costs.

7.2.1 Flexible pavements

Costs for vehicles can be derived from the logical chain (as in Figure 1, Section 1):

Vehicle loading → Pavement deterioration → Maintenance activities → Costs of maintenance today and in the future, or by looking at actual costs for maintaining road networks and tracking these costs back their origins.

In this case, opposite to the presentation so far, the latter is chosen.

The maintenance expenditures for rutting (“Trans”) and cracking (“Crack” and “Struct”) in Europe mainly varies between 60 % and 90 %, cf. Table 4 above. The share of rutting and cracking varies substantially between countries. Some countries have a share of rutting also originating from studded

CATRIN - Deliverable D 7 – January 2009

94 tyres. The category related to Structural adequacy (“Struct”) directly focused on capacity for heavy vehicles is mainly between 20 % and 45 %. These figures originate from high volume roads, i.e.

European primary road networks.

An attempt was made to rate the different distress mechanisms with respect to heavy vehicles. The result is presented below. Hence, the important deterioration factors should be rutting in bound and unbound layers (importance of subgrade contribution to rutting should be subject of further studies) and fatigue cracking in bound layers.

Table 5: Rough estimation of importance of impact from heavy vehicles on deterioration of flexible pavements. Roman numerals from Annex 1.

Cracking (XX, XXI)

Medium or Low Rutting (VIII, I, II, III, V), Cracking (XXII, XXV, VIII), Unevenness

(XI, XIII, XIV)

The co-variance between deterioration caused by vehicles and deterioration caused by other factors is very strong, as stated in the annex and in the review in Sections 3 to 5. Deterioration caused during short but critical conditions cannot be neglected. This may be one of the reasons why future pavement design procedures aim at a complicated incremental procedure in which deterioration is simulated over its whole pavement service life, cf. Section 5.

In order to track origins of distress, models need to be refined where distress mechanisms are related to their factors related to loading, climate and pavement materials. This can be achieved by combined activities of ALT, laboratory studies and analysis of pavement surface measurements. The outcome can be used in a range of applications from pavement design, maintenance planning, LCC (both technical and strategic levels) to marginal cost analysis.

Loading from heavy vehicles differ in amount of annual daily traffic on a statistical basis but the vehicle fleet is becoming more uniform since EU legislations are put into practice and long distance freight is increasing. The differences in annual daily traffic can be assumed to be taken into account in the pavement design and maintenance process. An open question is the overloaded vehicles and the extent of exemptions granted. Increasing traffic and loads becomes a problem if not accounted for in pavement design and maintenance strategy. Development of structural damage leads to needs for costly maintenance and reconstruction activities. There is evidence that structural deterioration can accelerate, for example structural cracking, thus leading to disproportionately higher costs for maintenance at a later stage if postponed. If the traffic and loading situation was not foreseen in the pavement design stage, measures can be taken afterwards, although often more expensive. A particular problem is if any legal changes take place, such as allowed axle weights or regarding policies for transport of heavy goods and freedom of movement for goods, persons and services. Abrupt changes are much more difficult to account for and may also bring problems in, for example, current PPP

(Public Private Partnerships) projects when basic contract conditions are changed. However, the same contractual aspect applies to many other ongoing projects or current agreement.

7.2.2 Rigid pavements

Costs on rigid pavements are more difficult to relate directly to vehicles. Wear is a direct consequence of vehicles. The other distress mechanisms are strongly dependent on other factors.

Increasing traffic and weights is more probably more expensive to account for after construction of rigid pavements, compared to flexible pavements. Consequently, Life Cycle Cost analysis including risk analysis may show statistically based benefits of slightly stronger pavements and more carefully handled production.

CATRIN - Deliverable D 7 – January 2009

95

7.3 Summary of consequences for EURODEX

The following remarks have been derived in previous sections:

• Traffic loading is quite complex to describe but shows interesting possibilities and pitfalls in the development of more pavement friendly vehicles and more wear resistant pavements.

• Correct design and selection of maintenance treatments need to estimate the influence of the miscellaneous vehicles characteristics in fleet and overall traffic intensity.

• The climate varies across Europe and is setting the conditions for pavement design. Climate is by its own or by combined action with traffic deteriorating pavements. Climate is also of immediate interest due to effects of climate changes.

• The more, closer, heavier and dynamic loads at the surface, the greater are the risks of deterioration in the subgrade. The deformation behaviour of the subgrade strongly influences the mode of failure of the pavement as a whole.

• Permanent deformations in granular materials are very sensitive to stress conditions and stress history. Unfortunately, stress conditions and stress history present in the field and in ALT testing is very difficult to reproduce exactly in laboratory.

• When comparing field observations, ALT testing and laboratory testing, it is important to keep in mind effects of time, for example due to bitumen ageing (long term), healing (mid term) and viscoelasticity (short term).

• Special products with unique features, such as noise reducing porous asphalt concrete, are usually more expensive and often more sensitive, leading to a substantial influence on maintenance costs and associated reasons of deteriorations.

• In order to catch the true origin of pavement deterioration costs caused by vehicles, both common conditions as well as rarely appearing conditions with high rates of deterioration, need to be considered.

• Most distress mechanisms are not clearly associated with traffic. This means that they do not contribute to marginal costs for traffic in the manner that is suggested by the fourth power law. However, in some situations the fourth power law is certainly applicable and even not enough sensitive (enough value of power) to the detrimental effects of heavy vehicles. At least four stages of rate of deterioration have been identified in the report.

• Looking at the history of the fourth power law and the conditions for the AASHO Road Test is necessary to establish a more refined model for pavement deterioration.

• Performance measurements could be used to relate fundamental technical issues to performance for road users.

• The next step is to tie costs for constructing and maintaining a road with its level of performance, thus relating costs for maintenance and construction with the actual technical decisions made regarding design and long term maintenance strategies. This information that can be used in decision making based on Life Cycle Cost Analysis.

• The review of demands of pavement performance and means of assessing the road network reveals a great source of information not yet reached its potential for applications in this field.

Together with the statistical and mechanistic-empiric models for pavement performance and design, a foundation is laid for a new system for assessing the damaging power of vehicles.

However, large gaps in the knowledge structures exist. Some gaps, especially those relating field and laboratory observations, can be bridged by well organised ALT. More fundamental pavement design methods are essential in this case to be able to not only predict performance of yesterdays pavement design, but to invoke new and more efficient pavement designs into practice.

CATRIN - Deliverable D 7 – January 2009

96

In order for EURODEX to become a fruitful effort, test conditions need to be carefully selected, controlled and described in documentation. Otherwise interpretation of results will be difficult. This is a consequence of the complex relationship between the summarised remarks presented above.

8 References

1

Cebon D ”Handbook of vehicle-road interaction” Swets & Zeitlinger Publishers. 1999

2

Lekarp F, Dawson A ”Modelling permanent deformation behaviour of unbound granular materials” Construction and building materials, vol. 12, no. 1. 1998

3

Lekarp F, Isacsson U, Dawson A ”State of the art. II : Permanent strain response of unbound aggregates” Journal of transportation engineering, vol. 126, no. 1. 2000

4

Molenaar A A A ”Prediction of Fatigue Cracking in Asphalt Pavements – Do we follow the right approach?” TRR No. 2001. 2007

5

Medani T A, Molenaar A A A ” A Simplified Practical Procedure for Estimation of Fatigue and

Crack Growth Characteristics of Asphaltic Mixes” IJRMPD 1,4. 2000

6

Lee H J, Kim Y R, Lee S W ” Prediction of Asphalt Mix Fatigue Life with Viscoelastic Material

Properties” TRR 1832. 2003

7

”Guide for Mechanistic Empiric Design – Part 3. Design Analysis” NCHRP report 1-37A, TRB.

2004

8

Nilsson R ”A viscoelastic approach to flexible pavement design” Licentiate thesis. Royal

Institute of Technology, Sweden. 1999

9 ”The AASHO road test: History and description of project” Highway Research Board. Special report 61A. Washington DC. 1961

10

Huang Y H ”Pavement Design and Analysis” Prentice Hall. 1993

11

”Load testing of instrumented pavement sections – Literature review” Mn/DOT Office of

Materials and Road Research, MN, USA. 1999

12

COST 333, Development of New Bituminous Pavement Design Method, Final Report. 1999

13

Selection of Maintenance Techniques and Procedures for Implementation, Deliverable rep. D2,

EC-funded project FORMAT, 2003

14

Cuelho E, Mokwa R and Akin M ”Preventive Maintenance Treatments of Flexible Pavements:

A Synthesis of Highway Practice” FHWA/MT-06-009/8117-26, Western Transportation

Institute. 2006

15 COST 324, Long-Term Performance of Road Pavement, Final report. 1997

CATRIN - Deliverable D7 – January 2009

9 Annex 1 – Distress types found on flexible pavements

Rutting Surface

I.

Wear  

Wear from vehicle tyres. Studded tyres being the worst.

Dependent on wear resistance of wearing course.

Bound layer

II.

Loss of stones

Inferior material, aggravated by traffic.

III. Traffic compaction  

Traffic compacts asphalt concrete more efficiently, despite that compaction during production is aimed at full compaction.

 

IV.

Deformations ‐ changes in layer shape and  redistribution of material  

Permanent deformations under traffic can be identified as viscous (time dependent) or plastic (load dependent).

Generally, the heavier and slower the load, the more permanent deformations. Higher temperatures normally substantially decrease the abilities of asphalt concrete in this sense.

Changes in layer shape are also a function of the asphalt concretes ability to bear load and distribute on the layers below. In this case the stiffness of the asphalt concrete layer is crucial.

Unbound layer V.   Compaction  

Traffic loading may further compact unbound materials after initial compaction during production. High contents of water and fine aggregate or particles, as well support from surrounding materials, is critical in this context.

 

VI.   Deformations  

Traffic loading and deformations in supporting layers may lead to geometrical changes in the unbound layers.

 

VII.

Redistribution and blending of materials  

Fine materials may penetrate into adjacent materials.

Usually, geotextiles or filter layers are used to stop this process. The process may lead to inferior performance and accelerating rate of deterioration. Fine materials may also be removed (e.g. by water) leading to volume contraction.

Subgrade

VIII.

Settlements and deformations 

Subgrade settlements or deformations may prolong for a very long period of time, for example if extra weight is put on a clay by constructing a pavement. In order of 2-3 meters (geometrically dependent) down in the pavement, traffic loading is no longer significant compared to the pavement intrinsic weight. In the upper subgrade, the same traffic induced distress may appear as described in the unbound materials section above.

97

CATRIN - Deliverable D 7 – January 2009

Un‐ eveness

Long range

IX.

Uneven settlements  

Long range uneven settlements are likely due to differences along the road far down in the subgrade with respect to soils, geotechnical measures or handling of soils.

Medium range

X.

Frost heave  

Related to frost sensitive subgrades, climate

XI.

Uneven settlements  

Uneven subgrade material or geotechnical measures along the road will to some extent show in different settlements at the surface. To minimise the effect, the pavement itself will equalise some and the rest by shifting the change in properties continuously over a long stretch.

Local

XII.

Bumps

Faults high up in the pavement, e.g. large rocks or pipes too close to surface (related to compaction and uneven settlements).

Cracks Wide

Medium

XIII.

Pot holes  

Multiple origins but often associated with poor production, traffic action and water.

XIV. Local damages from vehicles and cargo handling  

Probably more common on weak low volume roads.

 

XV.

Repair 

Damages related to road works and patching.

XVI. Frost heave cracks  

Related to frost sensitive subgrades, climate and inadequate flexibility of pavement.

 

XVII. Low temperature cracking (transversal

)

 

Inadequate low temperature flexibility at low and falling temperatures.

 

XVIII.

Reflective cracking  

Related to cracks or deformations zones below asphalt concrete layer directing all deformations to a narrow zone.

XIX.

Subgrade sliding

Sliding shear zones deep down due to geotechnical problems may reflect up to the surface (also causing unevenness and edge faults).

XX. Top‐down fatigue cracking  

Related to repeated loading from mainly heavy vehicles but probably also influenced by surface phenomenon such as temperature variations and extensive ageing at the more exposed surface.

 

XXI.

Bottom up fatigue cracking  

Related to repeated loading from mainly heavy vehicles but probably also influenced by other superimposed stresses such as temperature variations and uneven settlements (heave).

Fine

XXII.

Alligator cracking

Appearing on thin asphalt pavements or on spots with poor supporting layers. A sign of exceeded life.

98

CATRIN - Deliverable D 7 – January 2009

Other

XXIII.

 Stone loss

Occur when bitumen can no longer hold stone particles at the surface. Combined reasons are low bitumen content, low ductility of bitumen, poor adhesion stone-bitumen, traffic action (shear), other damages (fine cracking), stone gradation, etc. Lead to rutting but also damage vehicles.

XXIV.

 Bleeding

Bitumen will increase its volume by temperature expansion. Asphalt concrete need enough air voids to allow bitumen to expand at hot sunny summer days.

XXV.

 Shear damage

On heavy horizontally loaded spots the risk of shear damage is great, for example at bus stops, roundabouts, and stopping points for airplanes. However, shear damage is always a risk if not considered (by proper use of e.g. tack coat).

99

CATRIN - Deliverable D 7 – January 2009

10 Annex 2 – Distress types found on rigid pavements

• Linear (panel) cracking [strong influence from heavy traffic]

• Wear [strong influence from traffic]

• Polished aggregate [strong influence from traffic]

• Corner break [likely influenced by heavy traffic]

• Spalling [likely influenced by heavy traffic]

• Pumping [likely influenced by heavy traffic]

• Joint load transfer system deterioration [likely influenced by heavy traffic]

• Blowup (buckling)

• Durability cracking ("D" cracking)

• Faulting (level difference at joint)

• Popouts (surface faults)

• Punchout (deep local failure, often at steel reinforcement)

• Reactive aggregate distresses

• Shrinkage cracking

100

CATRIN - Deliverable D 7 – January 2009

101

11 Annex 3 – HDM models of flexible pavement performance indicators rut depth and cracking

Highway Design Manual HDM model for rutting

RDM

=

RD

0

+

RDST

+

RDPD

+

RDW

RDM

RD0

RDST

RDPD

RDW

Mean rut depth [mm]

Rut due to initial densification [mm]

Rut due to structural deformation [mm]

Rut due to plastic deformation [mm]

Rut due to wear [mm]

RD

0

= ⋅

0

(

⋅ −

6

)

(

a

1

+

a

2

⋅ + ⋅ ⋅

)

(1)

SNC a

4

COMP a

5

(2)

Kid a0 – a5

(1.0)

Parameter describing pavement structure type

YE4

DEF

MMP

ACX

Accumulated load cycles per lane and year in millions

Sinking under Benkelman-Beam [mm] (80 kN standard axle laod)

Mean, monthly rainfall [m]

Damaged surface by cracking [%]

SNC

COMP

Modified structural number

Relative compaction [%]

For all pavement structure types used in Austria a0 = 51740, a1 = 0.09, a2 = 0.0384, a3 =

0.0016, a4 = -0.502, a5 = -2.3. If there is no data for DEF available, it can be estimated from the following relationship:

DEF

=

6 5

SNC

1 6

(3)

The relative compaction varies between 100 (very good) and 85 (very poor).

Δ

RDST

=

(

0

+ ⋅

DEF

+

a

2

⋅ ⋅

AGE

3

+

a

3

MMP

⋅ Δ

ACX

)

AGE

3

YE

4

))

RDMa

⋅ Δ

T

(4)

Krst (1.0) a0 – a3

AGE3

ΔACX

Parameter describing pavement structure type

Pavement structure age since last structural rehabilitation [years]

Predicted increase of damaged surface by cracking [%]

RDMa rut depth at the beginning of the year [mm] year

For all pavement structure types used in Austria a0 = 0.333, a1 = 0.0494, a2 = 0.0021, a3 =

0.0285.

To describe rutting due to plastic deformation a number of parameters is used in HDM.

ΔVIM a0 – a2

Sh

Δ

VIM

=

a

0

YE

4

Sh a

1

PT

( )

a

SP

2

(5)

Decrease of voids in the mix from one year to the next

Parameter describing pavement structure type

Speed of heavy load traffic [km/h]

CATRIN - Deliverable D 7 – January 2009

102

PT Pavement

SP Softening Point [°C]

For the first year the parameters for Austrian pavement structure types are a0 = 43.558, a1 = -

0.616, a2 = 2.231, for the following years they are a0 = 5.27, a1 = -0.716, a2 = 3.225.

The change in the Softening Point (SP) can be evaluated from the following equation

SP

=

SPi

+

SPm

+ Δ

SP

(6) with

Δ

SP

=

a

0

⋅ ⋅

a

1

(7)

SP

SPi

SPm

ΔSP

Softening Point [°C]

Softening Point [°C], laboratory test, unaged bitumen

Change in Softening Point due to pavement construction [K]

Change in Softening Point per year [K]

PT Pavement

VIMa Voids in the mix at the beginning of the analysis [%]

On this basis the plastic part of the deformation of bituminous bound layers can be computed

Δ

RDPD

= ⋅

0

YE

4

Sh a

1

HS a

2

PT a

3

( )

VIM

SP

ΔRDPD

Increase of rut depth due to plastic deformation [mm]

Krpd a0 – a4

SP

Parameter

Softening Point [°C]

a

4

⋅ Δ

T

(8)

HS Dimension of bituminous bound layer [mm]

No parameters for Austrian pavement structure types have been evaluated yet.

To determine the rut due to wear HDM presents the following relationship

RDW

PASS Number of vehicles with studded tires passing the road per year

CW Parameter

S

SALT

RDW

= ⋅

5

Rut depth due to wear [mm]

Vehicle speed [km/h]

S

SALT

(9)

2 for roads where salt is spread in winter, 1 where no salt is spread in winter

There are no indications for how to choose the parameter CW in the HDM.

The HDM-method is quite complex when it comes to predicting rutting. The development of ruts is divided into 4 main aspects:

− initial densification

− structural damage

− plastic deformation

− wear

For the Austrian high-level pavement structure type (Lastklasse S) the HDM-method shows a total of 8 mm rut depth in the first 10 years regarding the initial densification, structural damage and wear. For rutting due to plastic deformation the HDM does not provide a practical relationship, it is far to complex to be used on a day-by-day routine.

CATRIN - Deliverable D 7 – January 2009

103

Highway Design Manual HDM model for cracking

The HDM

2

divides cracks into 4 classes depending on the width of a crack. The Maximum-

Likelihood-Method and the statistics of Weibull are used to evaluate data from different test road sections. The following equation represents the density function:

=

α − β β

t

exp(

α − β β

t

/

b

)

(10)

The related distribution function is defined as

exp(

1

β

α

β β

t

)

(11)

In these equations,

α can be expressed as a function of different independent variables

α

=

exp(

y

0

+

y x

+

y x

+

...)

(12) y i x i

Coefficient of Parameter x i

Parameter

As an example the following relationship for parameter

α is given in the report

α

= ⋅

SNC

− ⋅

YE

4

2

)

(13)

SNC

SNC modified structural number

YE4 accumulated load cycles per lane and year in millions

Parameter

β is given with 2.28. Paterson

3

shows 38 more relationships for

α and β. The difference between each relationship is the choice of parameters (e.g. deflection instead of structural number) and on the other hand the statistical significance.

In the handbook for the HDM-III-model an explicit equation for the start of cracking on a road surface is given:

ICX

SNC

YE4

ICX

=

4 21 0 14

SNC

− ⋅

YE

4

2

SNC

time when cracking occurs for the first time [years] modified structural number accumulated load cycles per lane and year in millions

)

(14)

2

N.D. Lea International Ltd.: Modelling Road Deterioration and Maintenance Effects in HDM-4, November

1995

3

Paterson, W.D.O.: Road Deterioration and Maintenance Effects, Models for Planning and Management. World

Bank, Washington D.C., 1987

CATRIN - Deliverable D7 – January 2009

104

Appendix B: Datasheets of European ALT facilities

CATRIN - Deliverable D 7 – January 2009

105

CATRIN - Deliverable D 7 – January 2009

106

CATRIN - Deliverable D 7 – January 2009

107

CATRIN - Deliverable D 7 – January 2009

108

CATRIN - Deliverable D 7 – January 2009

109

CATRIN - Deliverable D 7 – January 2009

110

CATRIN - Deliverable D 7 – January 2009

111

CATRIN - Deliverable D 7 – January 2009

112

CATRIN - Deliverable D 7 – January 2009

113

CATRIN - Deliverable D 7 – January 2009

114

CATRIN - Deliverable D 7 – January 2009

115

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement