Highway Maintenance Scheduling Using Genetic Algorithm with Microscopic Traffic Simulation

This paper introduces a hybrid genetic algorithm-microscopic traffic simulation methodology for scheduling of pavement maintenance activities involving lane closures aiming to minimize network traffic delay. Genetic algorithm is implemented as an optimization tool for the generation and selection of maintenance schedules. For each possible schedule, a microscopic traffic simulation model is used to simulate traffic operations in the network to estimate the delay caused by the lane closures. This concept has been implemented with the genetic algorithm written in the C language and simulation carried out by PARAMICS. The hybrid model has been tested with an example problem that requires the scheduling of three maintenance crews to 10 links within a 24-hour period, in a network comprises 160 links. It recommended a schedule that has avoided peak periods of traffic flow. The example problem has demonstrated the applicability of this hybrid methodology in pavement maintenance scheduling.