Developing a road performance index using a Bayesian belief network model

There is a high demand to develop and incorporate decision support tools, by the transportation sectors and other concerned agencies, to improve their infrastructure assets management under shrinking budgets and increasing demands. This paper develops a proof-of-concept Bayesian belief network (BBN) model to rank roads in a network system based on several key performance indicators (KPI) using a probabilistic framework. For a given road network, the proposed tool is capable of ranking or prioritizing the segment of roads for high level management objectives. To demonstrate the application of the proposed model, various scenarios are elaborated and discussed in detail. Finally, the sensitivity analysis is carried out to evaluate the effects of different KPIs.