Mechanistic-Empirical IRI Model Accounting for Potential Bias

作者：Aguiar Moya Jose P; Prozzi Jorge A^*; Smit Andre de Fortier

来源：Journal of Transportation Engineering-ASCE, 2011, 137(5): 297-304.

DOI：10.1061/(ASCE)TE.1943-5436.0000200

• Mechanistic-empirical design
• M-E PDG
• Roughness
• IRI
• LTPP
• Econometrics
• Random-effects
• Fixed-effects

The roughness prediction models in the International Roughness Index (IRI) currently incorporated into the Mechanistic-Empirical Pavement Design Guide (M-E PDG) have been developed by means of ordinary least squares (OLS). However, some of the variables used in predicting future IRI were previously estimated by means of separate performance models. This could cause bias because of the correlation between the previously estimated distress types and unobserved components on the IRI model. The bias can be corrected by considering additional variables that are correlated with the distress types causing the bias, thus eliminating the correlation to the unobserved terms in the model. Bias in the IRI model can also be generated by unobserved factors not included in the model. If these factors are section-specific, the bias can be removed considering variations in the performance history of different pavement sections. The writers have used updated Long-Term Pavement Performance (LTPP) data consistent with the data set originally used to fit the M-E IRI model for flexible pavements over thick granular bases contained in the M-E PDG. The data were then used in modeling IRI by means of OLS and instrumental variable (IV) regressions analyzing the data as pooled and as a panel data set (by random-effects, fixed-effects, and joint random-effects approach) to check for possible bias in the model. It was found that the current IRI model, as estimated by OLS, exhibits several types of biases attributed to heterogeneity and incorrect assumptions in the modeling process. It was identified that the preferred IRI model was the joint random-effects approach and, therefore, the model parameters were estimated by correcting the omitted-variable bias and simultaneous-equation bias. Estimating the model by accounting for possible bias in the data indicated considerable changes in the effects of different parameters affecting IRI through time, mainly rutting of the pavement structure. DOI: 10.1061/(ASCE)TE.1943-5436.0000200.

• 出版日期2011-5