Doctor of Philosophy
The public transit system in Sanandaj has been under review and modification for the last several years. The goal is to reduce the traffic congestion and the share of private car usage in the city and increase the very low share of the public transit. The bus routes in Sanandaj are not connected. There is no connected transit network with the ability to transfer between the routes in locations outside of the downtown terminal. The routes mostly connect the downtown core directly to the peripheries without providing travel options for passengers between peripheries. Although there has been some improvement in the transit system, but lack of service in many populated districts of Sanandaj and town nearby makes the transit system unpopular and unreliable.
This research is an attempt to provide solutions for the transit network design (TND) problem in Sanandaj using the capabilities of GIS and artificial intelligence methods. GIS offers several tools that enables the decision-makers to investigate the spatial correlations between different features. One of the contributions of this research is developing a transit network design with utilizing a spectrum of GIS software modeling functionalities. The visual ability of GIS is used to generate TNDs. Many studies focus on artificial intelligence as the main method to generate the TNDs, but the focus of this research is to combine GIS and artificial intelligence capabilities in order to generate a multi-objective GIS-based procedure to construct different bus network designs and explore and evaluate them to find the suitable transit network alternative.
The GIS-based procedure results will be assessed and compared with the results of metaheuristic approaches. Metaheuristic methods are partial search procedures that may provide sufficiently good solutions to an optimization problem characterized by incomplete information or limited computation capacity (Talbi, 2009). Yang, Cui, Xiao, Gandomi, and Karamanoglu (2013) classified metaheuristic methods into two groups: single-agent procedures (e.g., simulated annealing algorithm involves one agent navigating in the environment), and multiple agents (e.g., population-based genetic algorithm, and swarm intelligence methods). This study focuses on swarm intelligence methods, such as ant colony optimization and honeybee algorithm. These methods provide a multi-objective assessment of the TND scenarios generated by GIS applications. The outcome of this study will help us to find the optimal solutions for the TND in Sanandaj.
Summary for Lay Audience
Public transit systems such as bus play a crucial role in reducing the traffic and transport people with low-income to their destinations in the cities and suburbs. In this research, we propose and evaluate several transit network designs in order to improve the access to buses in the city of Sanandaj and reduce the usage of the private vehicles to travel to downtown or other towns and districts in the outskirts of Sanandaj. Several maps for the future bus network have been proposed and analyzed to find the suitable transit network layout for Sanandaj, based on the objectives such as better access to the bus system, minimum travel distance and daily cost of operation of the public transportation.
Ahmadi, Armin, "Transit Network Design using GIS and Metaheuristics in Sanandaj, IRAN" (2020). Electronic Thesis and Dissertation Repository. 6886.