Selection policies for a multifunctional workforce

Abstract

This paper investigates the selection and assignment of workers to tasks based on individual learning and forgetting characteristics in order to improve system throughput. We examine the performance of five policies that rank and select workers using a learning and forgetting-based assignment to address the task allocation problem. We are interested in the interaction between the implemented policies and various factors such as the ratio of generalists to specialists in the workforce, the level of multifunctionality and the level of workforce heterogeneity. Results demonstrate that, when implementing cross-training, selecting workers based on a greedy prediction of system output is outperformed by other simpler policies based on prior expertise.