Cash inventory management at automated teller machines under incomplete information

Abstract

Automated Teller Machines are one of the most important cash distribution channels for the banks. Since branches have more information about the ATM cash demand compared to headquarters, cash inventory at ATMs is usually managed by the branches. Inventory holding and stock-out costs, however, are incurred by the headquarters. We examine the headquarters’ cash inventory management problem for out of working hours under a Newsboy type setting. In this setting, inventory replenishment is not possible during the period and the headquarters cannot fully observe the amount of stock-outs. The headquarters seeks an inventory policy that would lead the branch to make ordering decisions to minimize the headquarters’ cost. We examine three policies: (M) policy with a lump-sum stock-out cost charged at the end of period, (t, M) policy with a lump-sum stock-out cost charged at time t within the period and (t, L, p) policy with a unit stock-out cost charged at time t within the period for inventory below L. Numerical analysis show that checking the inventory during the period - (t, M) and (t, L, p) policies - can lead to important reductions in cost for the headquarters.