Optimizing the efficiency, vulnerability and robustness of road-based para-transit networks using genetic algorithm

College

College of Computer Studies

Department/Unit

Software Technology

Document Type

Conference Proceeding

Source Title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume

10860 LNCS

First Page

3

Last Page

14

Publication Date

1-1-2018

Abstract

In the developing world, majority of people usually take para-transit services for their everyday commutes. However, their informal and demand-driven operation, like making arbitrary stops to pick up and drop off passengers, has been inefficient and poses challenges to efforts in integrating such services to more organized train and bus networks. In this study, we devised a methodology to design and optimize a road-based para-transit network using a genetic algorithm to optimize efficiency, robustness, and invulnerability. We first generated stops following certain geospatial distributions and connected them to build networks of routes. From them, we selected an initial population to be optimized and applied the genetic algorithm. Overall, our modified genetic algorithm with 20 evolutions optimized the 20% worst performing networks by 84% on average. For one network, we were able to significantly increase its fitness score by 223%. The highest fitness score the algorithm was able to produce through optimization was 0.532 from a score of 0.303. © Springer International Publishing AG, part of Springer Nature 2018.

html

Digitial Object Identifier (DOI)

10.1007/978-3-319-93698-7_1

Disciplines

Computer Sciences | Software Engineering

Keywords

Transportation--Developing countries; Bus travel--Developing countries; Genetic algorithms

Upload File

wf_no

This document is currently not available here.

Share

COinS