Use this URL to cite or link to this record in EThOS:
Title: Vehicle routing problem with availability constraints
Author: Johar, Farhana
ISNI:       0000 0004 5916 9986
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This work is concerned with solving the vehicle routing problem (VRP) which takes into account the customer's release and due date. The problem studied can also be categorized as a non-classical VRP as the departure times of vehicles depend on the dates of orders released from the production line and become available for the distribution process. Hence, the problem is known as VRP with availability constraints (VRPAC). The VRPAC is investigated through two stages. In the first stage, vehicle routing problem with release and due date (VRPRDD) is treated. At the beginning of the planning, it is assumed that the dates where the customer orders become available are known. A mathematical formulation is developed to represent the problem studied which has been solved by several heuristics, i.e. Variable Neighborhood Search (VNS), Large Neighborhood Search (LNS) and Tabu Search (TS). The algorithms are written in C++ and run on a PC computer with an Intel PentiumCore by using 56's Solomon instances with some modification. Different kinds of vehicle routing problem has been tackled in order to see the performance of proposed heuristics. The results are then compared in order to find the best method which yields the least routing cost solution. From the outcome obtained, VNS is proved to be the best algorithm which generates the least cost solution to our problem. Further investigation has been carried out in stage two which considers the extension of VRPRDD. The coordination of production sequence and vehicle routing (PS-VRPRDD) is the main subject to our problem studied in which the best production sequence will leads to the least routing. Two proposed algorithms have been used to run the test instances. The first is classical decomposition approach; Alternate which decompose the problems into two sub-problems, i.e. production sequence and vehicle routing. This will be used as benchmark to the second approach; InOneMove which take these two decisions of the sub-problems as a whole. Decision on both sub-problems is considered simultaneously as one move. The results proved that effective coordination shows the large potential savings that attract the interest of industrial distributors in optimizing their distribution process in practice.
Supervisor: Potts, Christopher ; Bennell, Julia Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available