Change search
ReferencesLink to record
Permanent link

Direct link
A Tree Neighborhood Topology Particle Swarm Optimization Algorithm for the Vehicle Routing Problem with Stochastic Travel and Service Times
School of Production Engineering and Management, Technical University of Crete, Decision Support Systems Laboratory, Department of Production Engineering and Management, Technical University of Crete.
School of Production Engineering and Management, Technical University of Crete.
Luleå University of Technology, Department of Business Administration, Technology and Social Sciences, Business Administration and Industrial Engineering.
2015 (English)Conference paper, Presentation (Refereed)
Abstract [en]

In recent years, a number of different Vehicle Routing Problems (VRP) with stochastic variableshave been studied and an excessive number of results have been published. Typically, stochasticvariables are associated with the customers or/and their demands, and/or time windows. In thistalk, the stochasticity concerns the travel times between two customers and the service times neededupon arrival to the customer. This variant of the problem is inspired from real life circumstances, as inmany cases the travel time between two customers can not be determined exactly due to fluctuationsin the conditions of the traffic network. The problem is solved using a Tree Neighborhood TopologyParticle Swarm Optimization (TNTPSO) algorithm. In this algorithm, a new neighborhood topologyfor the particles is used, where instead of a global topology, a local one is utilized based on a treestructure of the swarm. This structure provides a layered swarm where each member reacts onlywith its successors in the tree. The root at the top of the tree corresponds to the global best particle.Depending on the quality of the associated solutions, the particles move up or down in the tree.The particles on top levels of the tree have more influence on the swarm as they react with moreparticles from lower levels. We evaluate several different tree structures, and compare the best ofthem to previously published PSO implementations with global and local neighborhood topologiesshown efficient for other types of VRP. The reported computational comparisons are performed on anumber of benchmark instances.

Place, publisher, year, edition, pages
2015.
Keyword [en]
Tree Neighborhood Topology, Particle Swarm Optimization, Vehicle Routing Problem, Stochastic parameters, Local Search, Mathematics
Keyword [sv]
Tree Neighborhood Topology, Particle Swarm Optimization, Vehicle Routing Problem, Stochastic parameters, Local Search, Matematik
Research subject
Industrial Logistics
Identifiers
URN: urn:nbn:se:ltu:diva-39161Local ID: dccf24dc-f9ab-400d-beb0-db53ab240f88OAI: oai:DiVA.org:ltu-39161DiVA: diva2:1012670
Conference
World Congress on Global Optimization : 22/02/2015 - 25/03/2015
Note
Godkänd; 2015; 20150306 (athmig)Available from: 2016-10-03 Created: 2016-10-03Bibliographically approved

Open Access in DiVA

No full text

Other links

http://www.caopt.com/WCGO/

Search in DiVA

By author/editor
Migdalas, Athanasios
By organisation
Business Administration and Industrial Engineering

Search outside of DiVA

GoogleGoogle Scholar

Total: 28 hits
ReferencesLink to record
Permanent link

Direct link