Sterle, Claudio (2009) Location-Routing models and methods for Freight Distribution and Infomobility in City Logistics. [Tesi di dottorato] (Unpublished)


Download (3MB) | Preview
Item Type: Tesi di dottorato
Uncontrolled Keywords: Location-routing, flow interception, facility location, tabu search
Date Deposited: 24 May 2010 08:49
Last Modified: 30 Apr 2014 19:39


In this thesis two City Logistics problems have been tackled. The first is a design problem for a two-echelon freight distribution system. The aim is to define the structure of a system optimizing the location and the number of two different kinds of facilities, the size of two different vehicle fleets and the related routes on each echelon. The problem has been modeled as a two-echelon location-routing problem (2E-LRP). It is NP-hard and the results obtained with the proposed models, in terms of quality of solution and computation time, confirmed the need of approaching it with a heuristic method. To this aim a Tabu Search heuristic has been proposed and implemented. It is based on the decomposition of the whole problem in four subproblems, one FLP and one VRP for each echelon. The four sub-problems are sequentially and iteratively solved and their solutions are opportunely combined in order to determine a good global solution. Tabu Search has been experienced on three set instances. Experimental results prove that the proposed TS is effective in terms of quality of solutions and computation times in the most of the solved instances. The second problem concerns the location of flow intercepting facilities (FIFLP). Differently to what happens in classical location problems, in this case facilities do not generate and/or attract flows, but they intercept flows traveling on the network. The aim is to define the locations of the facilities which optimize a performance criterion related to the flow values on each path. This is a path-covering problem which finds many applications in City Logistics. Four problems of flow intercepting facility location have been treated with several models and methods, some of them present in literature and other proposed by us. An extensive computational experimentation of them has been carried out in terms of quality of solutions and computation times. FIFLP models have been successfully applied to transportation and communication networks.

Actions (login required)

View Item View Item