Metaheuristics and Cloud Computing: A Case Study on the Probabilistic Traveling Salesman Problem with Deadlines

摘要

In this work we study the potential of cloud computing in the context of optimization. For this purpose we perform a case study on the Probabilistic Traveling Salesman Problem with Deadlines, an extremely difficult stochastic combinatorial optimization problem. By using a large amount of computational resources in parallel, it is possible to obtain significantly better solutions for the Probabilistic Traveling Salesman Problem with Deadlines in the same amount of time. For many different applications this advantage clearly outweighs the requirement for additional computational resources, in particular considering the convenience of modern cloud infrastructures.