Padella, A., Mannhardt, F., Vinci, F., De Leoni, M., & Vanderfeesten, I. (2024). Experience-Based Resource Allocation for Remaining Time Optimization. In A. Marrella, M. Resinas, M. Jans, & M. Rosemann (Eds.), Business Process Management: 22nd International Conference, BPM 2024, Krakow, Poland, September 1–6, 2024, Proceedings (pp. 345-362). Article Chapter 20 (Lecture Notes in Computer Science (LNCS); Vol. 14940). Springer. https://doi.org/10.1007/978-3-031-70396-6_20
Abstract
Prescriptive process analytics aims to suggest interventions for those process instances that are predicted to not achieve a satisfactory outcome. Typical interventions are recommending a task to be performed by a specific resource. State-of-the-art prescriptive resource allocation techniques typically propose interventions that allocate the best-fitting resources at a given time. This may result in those resources to become more skilled at the task over time whereas other less experienced resource are rarely allocated. In the long run, such system may result in a unbalanced situation in which some expert resources are overloaded with very high workload and the less experienced resource are assigned fewer tasks and fail to improve. This paper proposes an approach for resource allocation to process instances that aims at a more balanced workload distribution among the resources, even if this means slightly lower process improvements in the short term. Experiments on event logs related to two real processes show that we indeed achieve a more balanced workload distribution, which often yields an overall higher improvement of the whole set of running process instances.