2019 CSCE Annual Conference - Laval (Greater Montreal) Conference
Dr. Luis E Amador-Jimenez , Concordia University
Performance-based contracts (PBC) for highways are increasingly becoming an attractive mechanism for transferring traditional public sector activities to private duties. Increased financial pressures on governments, user demands for improved service levels, and the operational efficiencies offered by the private sector, all create a strong business case for PBC. To enable municipalities and private sector investors engage in PBC, there is a need for quantitative tools that allow both entities: (1) structure the PBC in terms of risk allocation, (2) develop appropriate levels for service penalties and incentives in the contract, (3) define appropriate targets for highway level of service, and (4) determine the most cost-effective set of maintenance and rehabilitation (M&R) activities. This paper develops a series of mathematical optimization models that allow municipalities (pre-contract) to define: (1) performance indicators; (2) their threshold levels; and (3) appropriate penalties’ and incentives’ levels. Furthermore, its ability is expanded for post-contract decisions such that; it aids maintenance contractors in selecting the optimal M&R plan for both project and network-levels while minimizing the Life-Cycle Costs (LCC) and meeting the performance indicators’ limits. The developed system extends the typical functionality of traditional pavement management systems to cover specific PBC contractual requirements. It revolves around four models: (1) asset inventory, which includes all the necessary physical, climatic, and traffic information, (2) deterioration models; where defect-specific pavement deterioration models are developed using multivariate regression and stochastic network-level deterioration models are developed using markov chains, (3) life cycle costing models; which are developed to cover specific financial obligations in PBC like penalties and incentives, in addition to traditional M&R expenditures, and (4) optimization engine; where genetic algorithms was used to trade-off various decisions. The models were applied to a 100-km rural highway in the Northeastern Egyptian governorate and the results showed the drastic effect of the penalties/incentives limits on the LCC and Pavement Condition Index (PCI), displaying a 12% increase in the LCC with 4% improvement in the PCI. Furthermore, the sensitivity analysis exposed the considerable effect of the limits’ variability among the M&R costs, penalties/incentives values, and PCI. A 17% increase in the penalties and an 8% decrease in the incentives resulted in a 17% increasing in the LCC while reaching 91% PCI. In conclusion, the developed system is an effective tool for municipalities and contractors to make informed pre-contract and post-contract decisions on their approach to contractual risk allocation and M&R planning respectively.