A multiple criteria decision making (MCDM) tool is being developed to solve urban form optimization problems (UFOP). The UFOP involves identification of urban layouts that result in the lowest pollutant concentrations in a simulated urban domain. At present, the decision variables optimized include spatial locations and physical dimensions of the buildings, and the wind speed and wind direction over the domain. The solution of UFOP is comprised of three principal components - the forward prediction model, the objective function, and the optimization method. The Quick Urban and Industrial Complex (QUIC) dispersion modeling system adapted to graphics processing units (GPU) is adopted as the forward model in the MCDM tool because of its ability to produce reasonably accurate simulations extremely rapidly. For the UFOP, examples of objective functions include the maximum and the average pollutant concentration at the street level, the mortality rate, etc. In the MCDM tool being developed, several objectives can be taken into account simultaneously. This is accomplished by using single and multi-objective optimization techniques. In multi-objective optimization, the underlying assumption is that there is no single solution that is optimum with respect to all the objectives. Consequently, the notion of a best solution is replaced by Pareto optimal, or non-inferior, or effective solutions. A solution is said to be Pareto optimal if the reallocation of resources cannot improve one objective without deteriorating another. The goal of multi-objective optimization is to find as many Pareto solutions as possible. If all the Pareto solutions of a multi-objective problem are found, then the problem is said to have been solved to optimality. In the absence of any additional information about the objective functions, all the Pareto optimal solutions are equally satisfactory. In the MCDM tool, a non-dominated sorting genetic algorithm (NSGA) is used to identify families of urban designs that are Pareto optimal. The MCDM tool is also equipped with several well-established and accepted decision-maker preference structures. A simple example of a decision-maker preference structure is the weighting of the objective functions. If a decision-maker knows the weighting coefficients for each objective function a priori, then the weighted sum of the objectives can be optimized. In the MCDM tool, whenever the multi-objective problem is formulated as a single objective problem through the suitable selection of a decision-maker preference structure, single objective optimization techniques such as simulated annealing and genetic algorithms are used. At the conference, the various features of the MCDM tool will be demonstrated by considering a simple 2x2 building array problem.