A stochastic linear programming modelling and solution approach for planning the supply of rewards in Loyalty Reward Programs

Abstract

In this paper, we consider the problem of planning the supply of rewards in a Loyalty Reward Program (LRP). We formulate this problem as a two-stage stochastic linear program with simple recourse, and develop a new, sampling-based stochastic optimisation heuristic procedure for solving it. The proposed heuristic is general and can be applied in other contexts as well. Our computational experiments demonstrate the viability of the modelling and solution approaches for solving realistically-sized (large-scale) problems.