English

Abstract

In this paper we perform a computational study of 112 generated instances for the workforce scheduling and routing problem (WSRP). WSRP has applications in many service provider industries which require employees to visit customers to perform a diverse range of activities. We use a mathematical programming model from the literature of vehicle routing problem with time windows (VRPTW) to tackle WSRP due to their similarity. The computational study has three objectives: (1) To apply a VRPTW model to tackle medium size WSRP instances containing different distribution of visits, teaming and connected activities constraints. These constraints are an important because they allow the modelling of scenarios in which is necessary to link two or more activities over time. (2) To compare the difficulty between WSRP and VRPTW. And, (3) to investigate which time limit is adequate for the mathematical programming solver to find feasible solutions for our adapted instances. Based on our study results, we conclude that although the solver can achieve feasible solutions for some instances using the model, other solutions methods need to be considered due to the number of instances in which solutions could not be found. It is also found that our adaptations to Solomon's VRPTW data set to generate WSRP instances, increase their difficulty. Finally, when using a mathematical programming solver 2 hours time limit is recommended to get feasible solutions in medium size scenarios.