Retention Optimization in Maintenance Training Programs

Abstract

Training is an important business in the service sector. Although performance-enhancement training programs receive considerable attention from operations management researchers, the literature lacks guidance on designing maintenance training programs. We close this gap by investigating the modeling and optimization of a maintenance training program, considering the behavior that a customer may abandon the program if the training experience is too stressful. We formulate the maintenance training program design problem (MTDP), which maximizes overall service retention across all training sessions through activity scheduling. Customers make their participation decisions about their next training program based on the remembered holistic utility of past training activities. By our analysis, MTDP is a 0-1 constrained exponential sum problem (0-1 CESP), which is proven to be NP-hard. To resolve MTDP, we introduce a novel geometric branch and bound (GB&B) algorithm that searches for the optimal solution by resolving a series of subproblems. Our GB&B algorithm is proven to be efficient through our computational studies, and it has the potential to resolve other 0-1 CESPs that cannot be efficiently solved by traditional branch and bound algorithms due to difficulty in computing the bounds directly. We also investigate the joint training retention-performance optimization problem with a modified GB&B algorithm. We contribute to the literature by discussing a mathematical model, a solution method, and managerial insights for maintenance training program design.  

Keywords: maintenance training; service scheduling; customer retention management; geometric branch and bound algorithm; 0-1 constrained exponential sum problem