Exploring the Solution Space of Semi‐structured Geographical Problems Using Genetic Algorithms

Abstract

Semi‐structured geographical problems are often addressed by groups of decision‐makers. Each group member is likely to have a specific set of objectives that they wish to address and a unique perspective on the way in which the problem should be solved. The solution to such problems often requires consensus building and compromise among decision‐makers as they attempt to optimize their own criteria. The set of criteria adopted by a particular decision‐maker constrains the set of solutions he/she will deem acceptable. Compromise among multiple decision‐makers can occur at the intersection of these constrained solution sets. Knowledge about the criteria space, the solution space, and the relation between the two is often incomplete for semi‐structured problems. New tools are needed to explore, analyze, and visualize the solution space of a problem with respect to multiple analytical models and criteria. In this research we explore the utility of genetic algorithms as an effective means to: (1) search the solution space of geographical problems; (2) visualize the spatial ramifications of alternative criteria spaces; and (3) identify compromise solutions.