Integration of adaptive machine learning and knowledge-based systems for routing and scheduling applications

摘要

The combination of good mathematical models, knowledge-based systems, artificial neural networks, and adaptive genetic searches are shown to be synergistic. Practical applications of this combination produces near-optimal results, which none of the individual methods can produce on its own. We have developed XROUTE, a software system that demonstrates an integrated framework for this synergism, in the domain of computer-aided vehicle routing and scheduling problems. The purpose of this system is to assist researchers and decision makers who are applying the mathematical models to a specific routing problem instance by “tuning” the models to the problem description. The neural network modules store knowledge of previously solved problems and their solutions which facilitates the process of arriving at solutions to new problems. The knowledge-based system stores partial solutions from various knowledge sources, like the neural network and genetic algorithm modules, in the working memory and closely supervises the solution process in heuristic mathematical models. XROUTE provides an experimental, exploratory framework that allows many variations, and compares the alternatives on problems with different characteristics. The resultant system is dynamic, expandable, and adaptive and typically outperforms alternative methods in computer-aided vehicle routing.