|Title:||Agile bus route-timetable simulation using artificial intelligence||Other Titles:||Agile Simulation von Bus-Routen und Fahrplänen unter Zuhilfenahme künstlicher Intelligenz||Language:||English||Authors:||Nazari, Iman||Qualification level:||Diploma||Advisor:||Emberger, Günter||Assisting Advisor:||Brezina, Tadej||Issue Date:||2021||Citation:||
Nazari, I. (2021). Agile bus route-timetable simulation using artificial intelligence [Diploma Thesis, Technische Universität Wien]. reposiTUm. https://doi.org/10.34726/hss.2021.78002
|Number of Pages:||120||Qualification level:||Diploma||Abstract:||
This thesis assesses and explores a practical transportation model known as the Open Vehicle Routing Problem (OVRP) for urban and regional bus operation systems. The aims of the OVRP are generally to minimize the total number of busses and hence the total travel distance. OVRP can be seen as a Hamiltonian path in which a vehicle returns to the depot (starting point) after servicing the last stop on a route but in a reverse order. We implement and evaluate the combination of Ant Colony Optimization (ACO) and Tabu Search (TS) as a hybrid metaheuristic optimization algorithm in order to solve undirected symmetric open travelling salesman problem to find the shortest path in subgraphs. The benefits of my approach are related to exploring and examining a wide range of possible solutions to the problem within reasonable computational effort. My case study is based on an operational system in a city of the Lower Austria region. The results show, approx. 16% improvement in terms of total number of bus lines and approx. 4% for total operational length. In addition, the system was improved by approx. 12% in terms of the total number of courses, total number of vehicles required to operate in the system, and total travel time per day. To generalize the improvement by considering all the terms defined above, the system was improved by approx. 2% for total travel distance per day. This work shows the potential of applying artificial intelligence metaheuristics algorithms in transportation systems and their benefits, including greatly reduced time and cost of fleet management and bus operation systems.
|Keywords:||Bus system; Timetable; Artificial intelligence||URI:||https://doi.org/10.34726/hss.2021.78002
|DOI:||10.34726/hss.2021.78002||Library ID:||AC16200219||Organisation:||E230 - Institut für Verkehrswissenschaften||Publication Type:||Thesis
|Appears in Collections:||Thesis|
Files in this item:
checked on Jan 12, 2022
checked on Jan 12, 2022
Items in reposiTUm are protected by copyright, with all rights reserved, unless otherwise indicated.