Evaluating the value of information on the revenue performance of mobility on-demand system

Abstract

This paper presents an approach to incorporate demand prediction in flexible mobility ondemand revenue management system. This system has to deal with real-time management of ride services as well as customers demand in order to maximize the profit. We formulate the problem as a mixed-integer linear optimization model and present a math-based rolling horizon algorithm to solve it. We first compare our algorithm against the case in which no forecasted demand is considered. Thereafter, we present various scenarios with different demand forecasting error. Our results indicate that taking into account the future demand can dramatically increase the revenue performance of the system; however, this improvement does not linearly increase with forecast accuracy.