Multiconstrained Timetable Optimization and Performance Evaluation in the Presence of Travel Time Noise

Abstract

This work addresses the problem of bus timetabling under multiple operational constraints with the objective of minimizing the waiting time variations of passengers at stops. It places specific emphasis on practical operational constraints such as the required meal and resting time breaks, the required headway ranges between successive bus trips, and many others that are largely ignored in the literature but impose significant problems on timetable planners. Given the computational intractability of the multiconstrained timetabling problem, the authors introduce an exterior point penalty scheme for approximating the constrained optimization problem by an unconstrained one. In this way, it becomes easier to explore the effect of multiple timetabling options with the use of evolutionary optimization. The proposed approach is tested at a major Asia Pacific (APAC) bus operator, demonstrating that a favorable trade-off between improving the service reliability and satisfying most of the operational constraints can be achieved. Finally, additional extensive simulation tests demonstrate that optimized timetables perform well for travel time noise levels of less than 3%.