Scheduling on-site service deliveries to minimise the risk of missing appointment times

This paper studies the stochastic service task scheduling and vehicle routing problem for a telecommunication provider where each vehicle is driven by an engineer who performs service tasks at customer premises. There is an agreed time window for starting each service task. The service times and travel times are subject to uncertainties, e.g., task taking longer or shorter than expected, traffic situation causing delays. The problem is to schedule the tasks and route the vehicles to minimise the risks of missing appointment times. Models are presented to express the risks and describe the problem. Simulated annealing and tabu search are applied for generating an initial schedule of the day and for re-optimisation during the day based on real-time information updates. The study reported is based on the work in an industrial case. The stochastic nature of the travel times and durations of different task types as well as their distribution parameters have been obtained by applying data analytics on large sets of operations data. These are used in calculating the risks and in making scheduling and routing decisions. Real-time data updates sent back from the engineers are used for re-optimisation to adjust the schedule and routes so that the risks are kept at a lower level. Simulation results show that using risk minimisation as objective and re-optimisation during the day help enhance the on-time start of tasks. With this approach organizations can achieve robust task scheduling and improved customer satisfaction, and so become more competitive in the market.