Optimizing the use of contingent labor when demand is uncertain

We develop a new model for flexible workforce management in environments with uncertainty in the demand for labor. In particular we model a workforce comprised of regular workers who have fixed schedules and may work overtime, and contingent workers whose working hours are flexible as specified by a contract over a finite planning horizon. Our model can represent a variety of contracts, including: temporary workers, on-call workers with guaranteed minimum pay, and comp-time arrangements. We formulate the model as an optimization problem that determines the regular and contingent worker pool sizes that minimize expected labor and backlog costs. Embedded within this problem is a dynamic programming problem of making optimal operational staffing decisions with respect to the utilization of contingent and overtime resources. Numerical examples demonstrate the effect that the timing of information has on the benefits of flexibility. We also derive structural results that may be exploited to reduce the computational effort required to use the model.