Motifs are over-represented subgraphs in a complex network, and represent the building blocks of the network. There is a lack of studies that apply complex network theory in a supply chain context. In this dissertation 3-node motifs were identified and analysed in a complex network representing direct freight trips between firms in the Nelson Mandela Bay Metropolitan, South Africa. The G-Tries and ISMAGS algorithms were tested on small complex networks, and were compared according to quantitative and qualitative properties. It was found that ISMAGS is the most suitable for this dissertation. Freight activities were identified from raw GPS traces of freight vehicles, and the activities were clustered into firms using a density-based clustering algorithm. Multi-objective optimisation indicated that the clustering parameter configuration
γ = (20, 20) can be used to increase the visual accuracy of the firms, while maximising the completeness of the complex network. The freight complex network was built by identifying direct trips between firms. Using ISMAGS, it was found that three firms with two (X0X) or three (XXX) reciprocal freight trips between them are statistically overrepresented in the network. A brewery, shopping centres, distribution centres, and truck stops frequently appeared in the motifs. Motifs that contain the brewery and one of the truck stops were identified as the most central motifs based on the number of direct freight trips that occur in the motifs. Some freight trips in XXX motifs occur frequently over long distances, increasing total transport costs of the firms. Supply chain improvements can be applied to these identified firms. It was also found that there is a relationship between ranking firms according to the number of motifs they appear in and their degree centrality scores. This relationship can be studied more rigorously in future work. Another avenue for future research is to study the supply chain structures of firms in motifs, as well as the commodity flows between firms in motifs. / Dissertation (MEng)--University of Pretoria, 2016. / Industrial and Systems Engineering / Unrestricted
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/51895 |
Date | January 2016 |
Creators | Meintjes, Sumarie |
Contributors | Joubert, Johan W. |
Source Sets | South African National ETD Portal |
Language | English |
Detected Language | English |
Type | Dissertation |
Rights | © 2016 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
Page generated in 0.0024 seconds