In this talk, we consider the design of car-sharing services for a major car-sharing service providor. The problem is formulated as a high-dimensional discrete simulation-based optimization (DSO) problem. We propose a method that combines disaggregate car-sharing reservation data, analytical mathematical programming (MP) models, and simulation-based optimization algorithms. We present various ways in which the MP formulations can be used to enhance both the computational efficiency of DSO algorithms, as well as their ability to tackle high-dimensional problems. We present numerical results on a Boston case study.

Carolina Osorio is an Associate Professor in the Department of Civil and Environmental Engineering (CEE), and in the Operations Research Center (ORC) at the Massachusetts Institute of Technology (MIT). Her work develops operations research techniques to inform the design and operations of urban mobility systems.