This capstone project is sponsored by a water technology company and particularly covers its industrial pump rental business across the United States. With millions of dollars of annual spending for pump mobilization, the company looks for ways to improve the overall asset utilization rate. At its current practice, the company has not regularly used any statistical method or algorithm for demand prediction. Moreover, decisions for asset movement between branches are largely arranged between individual branch managers on an as-needed basis. We propose an improvement for the company’s asset management practice by modeling an integrated decision tool which involves evaluation of several machine learning algorithms for demand prediction and mathematical optimization for a centrally-planned asset allocation. We find that a feed-forward neural network (FNN) model with single hidden layer is the best performing predictor for the company’s intermittent product demand and the optimization model is proven to prescribe the most efficient asset allocation given the demand prediction from FNN model.