Implementation of the Metaheuristic Tabu Search in Route Selection for Mobility Analysis Support System

by David M. Ryer

This thesis employs a reactive tabu search heuristic implemented in the Java programming language to solve a real world variation of the vehicle routing problem with the objective of providing quality routes to Mobility Analysis Support System (MASS). MASS is a stochastic simulation model used extensively by Air Mobility Command (AMC) to analyze strategic airlift capabilities and future procurement decisions. This dynamic real world problem of strategic and tactical airlift possesses a number of side constraints such as vehicle capacities, route length and time windows in a sizeable network with multiple depots and a large fleet of heterogeneous vehicles. Finding optimal solutions to this problem is currently not practical. Currently, MASS requires all possible routes used in its simulation to be manually selected. As a result, the route selection process is a tedious and time consuming process that relies on experience and past performance of the model to obtain quality routes for the mobility system.

114 pages, Paperback

First published November 15, 2012