Show simple item record

dc.contributor.authorWalliser, Michael Anthony
dc.date.accessioned2014-03-04T20:34:25Z
dc.date.available2014-03-04T20:34:25Z
dc.date.issued2012-05
dc.identifier.otherwalliser_michael_a_201205_ms
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/walliser_michael_a_201205_ms
dc.identifier.urihttp://hdl.handle.net/10724/28149
dc.description.abstractSeveral heuristic algorithms are developed to find the fastest set of routes in order for a fleet of cleanup crews to clear hurricane-related debris from the road network at Fort Stewart near Savannah, Georgia. The road network is divided into priority levels such that each level must be completely cleared before work can begin on the next level. The problem is similar to the Hierarchical Chinese Postman Problem and the Capacitated Arc Routing Problem. It presents a unique challenge, however, in that each road must be cleared before it can be used for transport. Rule-based heuristics, adaptations of local beam search, and genetic algorithms are tested in various combinations with each other to build solutions. Results are evaluated based on the time required to clear the entire network of debris. The best solutions are generated by using a combination of all three of the aforementioned heuristics.
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectArc routing
dc.subjectHierarchical Chinese Postman Problem
dc.subjectCapacitated Arc Routing Problem
dc.subjectPath finding
dc.subjectSearch heuristics
dc.subjectLocal beam search
dc.subjectGenetic algorithm
dc.titleFort Stewart shortest path analysis of debris cleanup options
dc.typeThesis
dc.description.degreeMS
dc.description.departmentArtificial Intelligence Center
dc.description.majorArtificial Intelligence
dc.description.advisorPete Bettinger
dc.description.committeePete Bettinger
dc.description.committeeWalter D. Potter
dc.description.committeeCharles Cross


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record