“We Missed You!”: A joint optimization strategy of appointment window and reminder sending

Abstract

Customized service, specifically designed to meet individual customer preferences, is a rising consumption pattern along with the progress of socialization and urbanization. The customization nature requires service lead time, during which providers reserve required resources and prepare for individualized service delivery. As a result, when customers fail to show up (customer no-shows), it is impossible to use that time slot right away to serve another customer who can have a different set of service requirements. The vacant time slot and preparations will be forfeited, leading to significant indirect costs and opportunity loss. Meanwhile, service lead time makes it unrealistic for customized service providers to adopt some of the extant optimal scheduling strategies designed for traditional services, such as overbooking, open-access scheduling, and allowing walk-ins. In this study, we recognize the difference between customized services and traditional services in our analytical model formulation and jointly optimize the utilization of customized services from two aspects: adjusting the appointment windows and subsequently the reminder sending time. Optimal reminder sending time depends on when customers reserve their spots during the optimal appointment window. The earlier they reserve, the earlier they should be reminded. We further adopt a dynamic programming algorithm to calculate the optimal appointment window and the optimal reminder sending time in the closed form. Our proposed joint optimization strategy performs significantly better than two extant strategies: optimal appointment window only, optimal appointment window and sending the customer a reminder one day before the appointment window is closed. Finally, we conduct the sensitivity analysis and comparative analysis. We find that our joint optimization strategy is most beneficial to these customized services that have relatively short service lead time yet face the challenge of high customer no-show profiles and a low customer demand.