Empowering Cash Managers Through Compromise Programming

Typically, the cash management literature focuses on optimizing cost, hence neglecting risk analysis. In this chapter, we address the cash management problem from a multiobjective perspective by considering not only the cost but also the risk of cash policies. We propose novel measures to incorporate risk analysis as an additional goal in cash management. Next, we rely on compromise programming as a method to minimize the sum of weighted distances to an ideal point where both cost and risk are minimum. These weights reflect the particular preferences of cash managers when selecting the best policies that solve the multiobjective cash management problem. As a result, we suggest three alternative solvers to cover a wide range of possible situations: Monte Carlo methods, linear programming, and quadratic programming. We also provide a Python software library with an implementation of the proposed solvers ready to be embedded in cash management decision support systems. We finally describe a framework to assess the utility of cash management models when considering multiple objectives.