Testing and classifying vehicle dispatching rules in three real-world settings
Guided vehicles (GVs) are commonly used for the internal transportation of loads in warehouses, production plants and terminals. These guided vehicles can be routed with a variety of vehicle dispatching rules in an attempt to meet performance criteria such as minimizing the average load waiting times. In this research, we use simulation models of three companies to evaluate the performance of several real-time vehicle dispatching rules, in part described in the literature. It appears that there is a clear difference in average load waiting time between the different dispatching rules in the different environments. Simple rules, based on load and vehicle proximity (distance-based) perform best for all cases. The penalty for this is a relatively high maximum load waiting time. A distance-based rule with time truncation, giving more priority to loads that have to wait longer than a time threshold, appears to yield the best possible overall performance. A rule that particularly considers load-waiting time performs poor overall. We also show that using little pre-arrival information of loads leads to a significant improvement in the performance of the dispatching rules without changing their performance ranking.
|Keywords||case study research, dispatching, simulation, vehicle scheduling|
|Persistent URL||dx.doi.org/10.1016/j.jom.2004.05.006, hdl.handle.net/1765/11871|
de Koster, M.B.M., Le-Anh, T., & van der Meer, J.R.. (2004). Testing and classifying vehicle dispatching rules in three real-world settings. Journal of Operations Management, 22(4 SPEC. ISS.), 369–386. doi:10.1016/j.jom.2004.05.006