The Development of Models to Predict Pavement Performance in Frost Conditions

JOINT C­SHRP/QUEBEC BAYESIAN APPLICATION

Prepared for: Canadian Strategic Highway Research Program (C­SHRP)

Prepared by: Guy Dore Quebec Ministry of Transportation Pavement Branch

October, 1995

Executive Summary:

With freezing indices ranging from 1000 to over 2000 °C day and frost penetration reaching up to 3 meters under pavement surface, frost action has always been a major concern for pavement engineers in the province of Quebec (Canada). Faced with increasing needs and limited technical tools to cope with frost action on pavements, the Ministry of transportation has undertaken a major study with two primary objectives. The first is to revisit the whole approach of pavement design in frost conditions and to develop a rational pavement design method that deals with frost action. The second is to evaluate alternative techniques to protect pavement against the detrimental effects of frost action. In order to support these two objectives, a specific pavement monitoring program has been developed.

This paper presents the work done to develop first generation performance models using Bayesian statistics. This work was sponsored in part by the C­SHRP under the auspices of the (Joint C­SHRP/Agency Bayesian applications) project. The performance models will be the heart of the new design approach. Bayesian statistics compensate for the lack of good quality well controlled data needed for conventional statistical analysis of pavement performance, by providing a means whereby prior information (old databases and/or expert judgment) can be used to supplement the field data. The (XLBayes) statistical package, developed under the Canadian Strategic Highway Research Program (C­SHRP), will be used in the development of four distress specific models that will eventually feed the design model with information on pavement performance in freezing and thawing conditions.

Table of Contents:

1. INTRODUCTION

2. JOINT C­SHRP/AGENCY APPLICATION PROGRAM

3. PROBLEM STATEMENT

3.1 FROST ACTION ON PAVEMENTS

3.2 CURRENT DESIGN PROCEDURES

3.3 PROPOSED DESIGN PROCEDURE

4. THE DEVELOPMENT OF PERFORMANCE MODELS

4.1 MODEL TYPE

4.2 PERFORMANCE INDICATORS

4.3 METHODOLOGY

5. LITERATURE REVIEW

6. ROUGHNESS MODEL

6.1 SELECTION OF VARIABLES

6.2 DEVELOPMENT OF THE PRIOR

6.3 NEW DATA

6.4 QUALITY CONTROL OF THE DATA SET

6.5 FUNCTIONAL FORM OF THE MODEL

6.6 REGRESSION ANALYSIS

6.7 DISCUSSION

7. CRACKING MODEL

8. MODELS APPLICATION

9. CONCLUSION

ACKNOWLEDGMENT

REFERENCES

APPENDIX A

• Figure 1
• Figure 2
• Figure 3
• Figure 4
• Figure 5
• Figure 6
• Figure 7
• Figure 8a
• Figure 8b
• Figure 9
• Figure 10
• Figure 11
• Figure 12a
• Figure 12b
• Figure 13
• Figure 14
• Roughness Model: Prior
• Roughness Model : Posterior
• Roughness Model: Data 1
• Roughness Model: Data 2
• Roughness Model: Data 3
• Cracking Model: Prior
• Cracking Model: Posterior
• Cracking Model: Data

Return to Main Page