Optimal pavement management strategy development with a stochastic model and its practical application to Ugandan national roads

This study examines the applicability to developing countries of a pavement management system (PMS) with less data (two-point) requirement to complement current practice. Due to limited capacity, many developing countries like Uganda have adopted a PMS after specified intervals regardless of deterioration rate (i.e. time-dependent policy). Maintenance planning based on deterioration rate (i.e. condition-dependent policy), as proposed, would be more ideal because pavement deterioration is stochastic. A PMS was developed using the Markov hazard model and then an empirical study was conducted for surveyed Ugandan national road network comparing both policies considering a basic plan and a fixed budget. A greedy algorithm was used to select sections for repair based on worst condition first (for time-dependent) or fastest deterioration first (for condition-dependent). A shift to the condition-dependent policy increased percentages of network in good and fair condition by 8.6% and 2.5%, respectively, and reduced percentages in poor and bad condition by 8.5% and 2.6%, respectively. Additionally, maintenance plans with and without preventive maintenance were compared considering a fixed budget. Preventive maintenance further increased percentage in good condition by 27.4% and reduced percentages in poor and bad condition by 11.6% and 4.2%, respectively, with a 53.5% reduction in life cycle costs.