A Heuristic Approach for On-line Signal Repair Problems

Abstract

Due to the advancement of information and communication, city logistics operations could be designed more efficiently, especially for on-line requests. This research proposes an on-line algorithm to consider new requests and real-time traffic conditions for signal repair problems. The research framework includes two parts, off-line route planning and on-line route updating. The off-line problem is solved through a time-dependent mathematical formulation and the CPLEX solver. The on-line algorithm is constructed based on two heuristic approaches, the INTERVAL strategy and the Simulated Annealing (SA) algorithm. The INTERVAL strategy considers new demands at appropriate intervals, while the SA algorithm is applied to generate the best service route. Numerical experiments through the simulation-assignment model, DynaTAIWAN, are conducted to illustrate the algorithm. The results describe that positive benefit, about 13.4% of reduction in the total completion time, is achieved when compared with empirical data in a real city network.