Genetic algorithm based path optimization methodology for spatial decision

In this paper, we proposed a method based on GA to solve the path-optimization problem. Unlike the traditional methods, it considers many other factors besides the road length including the task assignment and its balance, which are beyond the capability of path analysis and make this problem a Combinatorial Optimization problem. It can't be solved by a traditional graph-based algorithm. This paper proposes a new algorithm that integrates the Graph Algorithm and Genetic Algorithm together to solve this problem. The traditional Graph-Algorithm is responsible for preprocessing data and GA is responsible for the global optimization. The goal is to find the best combination of paths to meet the requirement of time, cost and the reasonable task assiginment. The prototype of this problem is named the TSP (Traveling Salesman Problem) problem and known as NP-Hard Problem. However, we demonstrate how these problems are resolved by the GA without complicated programming, the result proves it's effective. The technique presented in this paper is helpful to those GIS developer working on an intelligent system to provide more effective decision-making.