ATM Cash Flow Prediction Using Local and Global Model Approaches in Cash Management Optimization

Cash management optimization is one of the most essential tasks for any bank, because it helps save a significant amount money by reducing cost ATMs funding and encashment. This paper focuses on forecasting customer cash demand, which key components system. Furthermore, first time, our research touches problem nonstationarity, typical real-world ATM data, proposes data preprocessing pipeline to tackle it. We proposed new methods in paradigms local global models, proved their superiority over classical approaches time series used specifically demand problem.