2024 : 6 : 13
Amirali Zarrinmehr

Amirali Zarrinmehr

Academic rank: Assistant Professor
ORCID:
Education: PhD.
ScopusId:
Faculty: Faculty of Technology and Engineering
Address: Uiversity of Mazandaran
Phone: 011-35302903

Research

Title
Application of a hill-climbing algorithm to public transportation routes design in grid networks
Type
JournalPaper
Keywords
Transit Routes Design, Grid Network, Coverage, Hill Climbing
Year
2021
Journal international journal of transportation engineering
DOI
Researchers Amirali Zarrinmehr ، Hanie Moloukzede

Abstract

Transit Routes Design (TRD) problem deals with optimizing the configuration of transit routes to satisfy a given objective, such as maximizing network coverage, while holding the budget constraint. In its discrete form, TRD is recognized as a computationally interactive problem. A review of the literature reveals that, despite extensive research on this problem, the number of studies on specific urban network configurations has remained limited. Among these studies, many have applied simplifying assumptions such as continuous design variables which may not be applicable to real-world settings. The present study focuses on a discrete version of the TRD problem for an urban “grid” network and aims to maximize the service coverage through the network. To this end, a local search Hill-Climbing (HC) heuristic algorithm is proposed and evaluated. The proposed HC algorithm performs several replications in which it starts with a combination of randomly selected routes and iteratively improves them by moving to the “best neighbor” until it reaches a local optimum solution. Our results for a 6×10 grid network for three budget levels (i.e. low, medium, and high budget levels) indicate that, in much shorter run-times than the exact algorithm, the proposed HC algorithm can produce high-quality solutions with 0.12%, 4.16%, and 2.22% difference from global optimums.