Evaluation of the Origin-Based Traffic Assignment Algorithms

Please enable javascript in your browser.

Mohammad Pure, Esmaeil | 2011

744 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 42897 (09)
University: Sharif University of Technology
Department: Civil Engineering
Advisor(s): Zokaee Aashtiani, Hedayat
Abstract:
Astransportation networksbecome larger and more congested, designers are seeking more precise, efficient and general algorithms that could use large memoryand powerfulCPUs efficiently. In thisthesis we evaluate Bar-Gera and Dial’sorigin-based algorithms thatsolvethe traffic assignment problem based on the user-equilibrium principal and compare them withthe path-based complementarity algorithm which has been improved in this study. Dial’s algorithm has been implemented in C++ programming language but for Bar-Gera’s algorithm we have usedan executable code that was implemented by him.Networks that have been used in this study are a small hypothetical network, Sioux Falls network, generated grid networks, Chicago sketch network and Chicago regional network. Based on our results and also other reported results, Dial’s algorithm is faster than Bar-Gera’s algorithm and the complementarity algorithm is faster than Dial’s origin-based algorithm
Keywords:
Traffic Assignment ; Complementary Algorithm ; Origin-Based Algorithm ; Bar-Gera Algorithm ; Dial Algorithm