New approaches for determining greenest paths and efficient vehicle routes on transportation networks

Official URL: http://risc01.sabanciuniv.edu/record=b1589540 (Table of Contents)

Road transportation has hazardous and threatening impacts on the environment. However, the traditional logistics models and approaches used in transportation planning have mainly focused on minimizing the internal costs and lack the environmental aspect. Therefore, new planning techniques and approaches are needed in road transport by explicitly accounting for these negative impacts. In this thesis, we address these issues by first concentrating on solution methods for the Greenest Path Problem (GPP) where fuel consumption and GHG emission objectives are incorporated to find the least GHG generating path, namely the greenest path, and propose a fast and effective heuristic. Taking the strong relation between the speed and the GHG emission into account, we also address the speed embedded minimum cost path problem in the most general case where the speed is also a decision variable as well as the departure time Within this context, we develop a new networkconsistent (which implies spatially and temporally consistent speeds) time-dependent speed and travel time layer generation scheme since real data is difficult to acquire. In the second part, we mainly focus on Vehicle Routing Problems (VRP). First, we propose an Ant Colony Optimization (ACO) approach for solving the Vehicle Routing Problem with Time Windows (VRPTW). Then, we adapt this method to solve the environment friendly VRP, namely the Green VRP, where the greenest paths between all customer pairs are used as input. Finally, we extend the ACO algorithm to a parallel matheuristic approach for solving a class of VRP variants.