The current process for the design of an urban master plan typically involves a team of architects and urban planners that conceive a handful of schemes based on zoning requirements with the help of CAD software. They may intend for the plan to achieve a set of performance goals (economic, environmental, etc.), but quantitative analysis is rarely conducted early and consistently through the design process. This makes it difficult to understand the full range of approaches that are possible on a site, and the relative performance of each scheme. In order to best accommodate rapid urbanization while making cities more sustainable, livable, and equitable, designers must utilize quantitative tools to make informed decisions about their designs. Computational design techniques have been successfully used at the building scale to test numerous designs and quantify their performance, but are challenging to apply at the urban scale due to increased computational expense, difficulty in limiting inputs, and more stakeholders involved in the process. This paper outlines a methodology developed in practice for applying computational design at the urban scale through four steps: 1) Simplified Input Definition 2) Procedural Geometry Generation, 3) Performance Evaluation and 4) Analysis & Communication to generate and test thousands of master planning scenarios.