This thesis consist of two parts. The first deals with bike-sharing systems which are now ubiquitous across the U.S.A. We have worked with Motivate, the operator of the systems in, for example, New York City, Chicago, and San Francisco, to innovate a data-driven approach to managing both their day-to-day operations and to provide insight on several central issues in the design of their systems. This work required the development of a number of new optimization models, characterizing their mathematical structure, and using this insight in designing algorithms to solve them. Many of these projects have been fully implemented to improve the design, rebalancing, and maintenance of Motivate’s systems across the country. In the second part, we study a queueing-theoretic model of on-demand transportation systems (e.g., Uber/Lyft, Scoot, etc.) to derive approximately optimal pricing, dispatch, and rebalancing policies. Though the resulting problems are high-dimensional and non-convex, we develop a general approximation framework, based on a novel convex relaxation. Our approach provides efficient algorithms with rigorous approximation guarantees for a wide range of objectives and controls.