Applying Genetic Algorithms and RIA Technologies to the Development of Complex-VRP Tools in Real-World Distribution of Petroleum Products

Distribution problems had held a large body of research and development covering the VRP problem and its multiple characteristics, but few investigations examine it as an Information System, and far fewer as how it should be addressed from a development and implementation point of view. This paper describes the characteristics of a real information system for fuel distribution problems at country scale, joining the VRP research and development work using Genetic Algorithms, with the design of a Web based Information System. In this paper a view of traditional workflow in this area is shown, with a new approximation in which is based proposed system. Taking into account all constraint in the field, the authors have developed a VRPWeb-based solution using Genetic Algorithms with multiple web frameworks for each architecture layer, focusing on functionality and usability, in order to minimizing human error and maximizing productivity. To achieve these goals, authors have use SmartGWT as a powerful Web based RIA SPA framework with java integration, and multiple server frameworks and OSS based solutions, applied to development of a very complex VRP system for a logistics operator of petroleum products.