Abstract— Many banks maintain as much as 40 % more cash than necessary at their different ATMs which implies this cash is non-circulating. The aim of this paper is to decide the optimum amount of money that needs to be placed in the ATM so that the surplus amount can be used in other bank products while tending to the customer’s uncertain demand. Stocking cash in ATM entails costs like interest rates, out-of-service, risk of robbery, transportation costs, cash upload cost. The most effective way of understanding the withdrawal pattern is by gaining insights from the available historical data in order to predict demand for the future.
Index Terms— ATM, neural network, regression, simulation.
The authors are with the Institute of Systems Science, National University of Singapore, Singapore 119615 (e-mail: dandekarpushkar@nus.edu.sg, a0119950@nus.edu.sg).
[PDF]
Cite: Pushkar V. Dandekar and Ketki M. Ranade, " ATM Cash Flow Management," International Journal of Innovation, Management and Technology vol. 6, no. 5, pp. 343-347, 2015.