Smart ATM Cash Replenishment Planning using Deep Reinforcement Learning

Access to cash for many in society is remaining
essential during the current COVID-19 lock-down around the
globe. A smart city requires the banking industry to exploit IoT
and Artificial Intelligence (AI) in order to track its ATM network
and predict outages due to cash shortages. In this paper, we study
the real-time cash replenishment planning problem under outflow
uncertainty where the fee of the security companies grows if the
replenishment ends up falling on a weekends/holidays. Our model
is based by the Double Deep Q-Network (DQN) algorithm which
combines popular Q-learning with a deep neural network. The
proposed method is used to minimize the ATM replenishment
cost where the cash demand changes dynamically at each day.
The performance analysis of the proposed method for different
amounts of replenishment cash shows that the the proposed
method can effectively work under real word conditions and
reduce the ATM operational cost compared with the other stateof-the-art cash demand prediction schemes.
Index Terms—cash replenishment planning, deep learning,
ATM, reinforcement learning, double Q-network.