7: Intelligent Vehicle Routing and Scheduling

Distribution systems are dynamic providing many challenges for traditional optimisation methods. Vehicle routing and scheduling procedures need to be able to cope with uncertainty in order to provide effective decision support systems for fleet managers.

This paper presents a new method of representing the uncertainty of travel times between customers based on spare time, that is the amount of time that vehicle arrives at a customer before the end of the time windows. Spare time is incorporated within the optimisation procedures as an indirect benefit.

An example was used to estimate the benefits of using the spare-time model compared with the traditional model. Travel speeds were simulated representing a wide range of traffic conditions. Both models had similar performance of both models with good travel time conditions. However, the Spare-Time model had substantially less delay costs incurred in congested traffic conditions. The Spare-Time model was shown to be more robust with respect to travel time conditions than the conventional model. Significant overall savings in direct costs werepredicted. Increased levels of service for customers were also achieved.