Multiobjective Urban Planning Using Genetic Algorithm

Abstract

A genetic algorithm was used to search for optimal future land-use and transportation plans for a high-growth city. Millions of plans were considered. Constraints were imposed to ensure affordable housing for future residents. Objectives included the minimization of traffic congestion, the minimization of costs, and the minimization of change from the status quo. The genetic algorithm provides planners and decision makers with a set of optimal plans known as the Pareto set. The value of each plan in the Pareto set depends on the relative importance that decision makers place on the various objectives.