Client-driven, capacitated systems have special characteristics that pose a unique challenge for those attempting to schedule and control such systems. The variable nature of client requirements, the prioritization of clients, limited resource availability, parallel tracks for projects, and stochastic nature of project activities all contribute to the unique nature of these types of systems. Using information gathered from representative systems, we first show how these stochastic, client-driven, capacitated systems can be modeled. Furthermore, these models can be used for real-time scheduling of project activities using a Bayesian approach and an objective function that targets project revenue. We conclude with an illustrative application demonstrating model components and capabilities.