Prescriptive Analytics for a Multi-Shift Staffing Problem

Motivated by the work with a leading maintenance service provider in the aviationindustry, this paper examines novel data-driven approaches to solving a certaintype of capacity-sizing problem—the multi-shift staffing problem—with uncertain,time varying arrival rates and patient "customers" that do not abandon the queuewhile waiting for a service, but who must be served by a pre-defined time. Drawingon established methods in both capacity management and prescriptive analytics, wepropose to use fluid and stationary approximations to apply tailored prescriptiveanalytics approaches to determine staffing levels for multiple interrelated shifts. Theprescriptive analytics approaches rely on machine learning techniques that incorporatea detailed representation of the non-stationary structure of arrivals and leverageextensive auxiliary data that may be predictive of demand. In particular, we adaptestablished prescriptive analytics approaches—weighted sample average approximationand kernelized empirical risk minimization—and propose a new optimizationprediction approach to solving the multi-shift staffing problem. Using a case studythat is based on extensive data from our project partner, the maintenance serviceprovider, we demonstrate the applicability of these approaches, highlight their benefitsover traditional "estimate then optimize" approaches, and shed light on theirstructural properties and performance drivers. In the context of our real-world application,we derive a clear recommendation for the choice of method with which tosolve the multi-shift staffing problem