Computational Intelligence Algorithms for Optimized Vehicle Routing Applications in GIS

This project seeks to explore the application of two developing algorithmic paradigms from the field of computational intelligence towards optimized vehicle routing within geographic information systems (GIS). Ant Colony Optimization (ACO) is a type of multi-agent, swarm-based algorithm designed to mimic the emergent problem-solving behavior of real ants within a colony. Genetic Algorithms (GA) are another natureinspired type of algorithm designed for evolving optimal or near-optimal solutions to a problem through the use of techniques based on natural selection, crossover, and mutation. The goal of this project is to implement and test a hybrid version of these two algorithms, aimed at evolving agents (ants) for optimized routing within the vehicle routing program. Introduction Within the last few decades of the 20 century, and continuing into the new millennium, there has been exponential growth in the research, development, and utilization of geographic information systems (GIS). However, while GIS have been designed to handle most types of common spatial analysis problems, many of the more complex spatial problems are still beyond their current capabilities to solve. These types of problems often involve extremely large search spaces with correspondingly large numbers of potential solutions. In such cases, standard analytical techniques typically fall short of finding optimal solutions to the problem within practical temporal and/or computational limits. One such problem within the realm of spatial analysis is that of the vehicle routing problem (VRP). This project will be focused on the development of algorithmic solutions for the VRP, designed specifically for solving such complex problems. Vehicle routing has many practical applications within the fields of operations research, logistics, distribution, supply chain management, and transportation, to name a few. In general, vehicle routing involves finding efficient routes for vehicles along transportation networks, in order to minimize route length, service cost, travel time, number of vehicles, etc. As will be demonstrated in the next section, in which a more detailed and formal definition of the VRP is given, this is a combinatorial optimization problem for which no simple solutions exist. As an alternative, solution techniques from the field of computational intelligence shall be implemented and tested for solving instances of the VRP. According to Engelbrecht (2002), computational intelligence may be defined as a sub-field of artificial intelligence (AI) that entails “the study of adaptive mechanisms to enable or facilitate intelligent behavior in complex and changing environments. These mechanisms include those AI paradigms that exhibit an ability to learn or adapt to new situations, to generalize, abstract, discover, and associate” (p.4). Computational intelligence is itself divided among many sub-fields of study, including (but not limited