Semantic-web Technology Based Freight Forwarding Matching System

No-load running on backhaul transportation is one of the urgent problems thatmake the cost of transportation maintain at a high level. In recent years, manylogistics information systems focusing on freight matching have appeared. However,the searching and matching engines of most of the traditional logistics informationsystems are developed basing on simple text search technology, which results inlow matching accuracy and efficiency. Hence, improving the matching accuracyand efficiency between freight information and truck information in the logisticsinformation system is the main purpose of the thesis.Firstly, semantic web technology is introduced, following by the literaturereview on its application in logistics field. This part includes the concept ofsemantic web, its main theories and technologies, several useful softwares, as wellas the methodology of ontology. We then reviewed the previous research andapplication of semantic web in logistics area. The result shows that applyingsemantic web technology to freight matching field is really feasible and innovative.In the second part of the thesis, the ontology of the freight matching realm isdeveloped. Protégé4.1is used to build the ontology. And the ontology is saved as aRDF file for further research.The third part is the core of the thesis. A freight matching information systembased on semantic web technology is built with Java. Jena is used to read RDFontology files and user-defined reference rule files in the system. The systemintegrates the ontology successfully. SPARQL is used for querying vehicle andgoods in the inferred ontology. An algorithm is designed to calculate thevehicles-goods matching rate. The vehicles and goods which are found by semanticsearching engine are sorted by the matching rate and the final results are presentedto the system users.In the last part of the thesis, the actual data is used to simulate two commoncases. Experiments are implemented on the two cases. After analyzing the cases andresults, the result shows the great potential application value of the system.