The Canadian Strategic Highway Program (C-SHRP) has developed and demonstrated Bayesian statistical methods and software as a tool to assist in the development of pavement performance models. The unique features of these methods are that they allow the analyst to encode expert judgement and then combine this judgement with field data. In the case of the Canadian Long Term Pavement Performance Program (C-LTPP), this is most helpful in that it explicitly addresses the problem of small sample size within C-LTPP and facilities the development of early modelling results. This paper presents an overview of the methodology of implementing Bayesian regression within the context of the Bayesian template. The template outlines a ten step process designed to facilitate the development of Bayesian regression models. A description of the underlying mathematical solution to the posterior model is also presented. (A)
Samenvatting