Cycle Time Optimization of Processes Using an Entropy-Based Learning for Task Allocation

Abstract

Cycle time optimization could be one of the great challenges in business process management. Although there is much research on this subject, task similarities have been paid little attention. In this paper, a new approach is proposed to optimize cycle time by minimizing entropy of work lists in resource allocation while keeping workloads balanced. The idea of the entropy of work lists comes from the fact that the time it takes for a resource to do similar tasks in a rather consecutive order is less than the time it takes to do the same tasks separately. To this end, an entropy measurement is defined, which represents task similarities on some given work lists. Furthermore, workload balancing is also regarded as an objective because not only is cycle time optimization important, but also workload fairness should also be met. Experimental results on a real-life event log of BPI challenge 2012 showed that the proposed method leads to 32% reduction in cycle time, compared with a reinforcement learning resource allocation without involving the entropy.