Real-time vehicle scheduling in bus rapid transit systems using distributed artificial intelligence

This project is focused on Bus Rapid Transits Systems (BRTS), which is a modem transportation System important for urban traffic development. In fact, BRT Systems have become popular in many countries, reaching 160 cities around the world. BRT systems have two components, design and operational plan...

Full description

Autores:
Galindo Acuña, David Esteban
Plazas Bernat, Sylvia Paola
Tipo de recurso:
Trabajo de grado de pregrado
Fecha de publicación:
2016
Institución:
Pontificia Universidad Javeriana
Repositorio:
Repositorio Universidad Javeriana
Idioma:
spa
OAI Identifier:
oai:repository.javeriana.edu.co:10554/38377
Acceso en línea:
http://hdl.handle.net/10554/38377
Palabra clave:
Programación
BRT
Inteligencia artificial
Control el tiempo real
Scheduling
Bus rapid transit
Artificial intelligence
Real time control
Ingeniería industrial - Tesis y disertaciones académicas
Líneas de autobuses - Bogotá (Colombia)
Programación en tiempo real
Rights
openAccess
License
Atribución-NoComercial-SinDerivadas 4.0 Internacional
Description
Summary:This project is focused on Bus Rapid Transits Systems (BRTS), which is a modem transportation System important for urban traffic development. In fact, BRT Systems have become popular in many countries, reaching 160 cities around the world. BRT systems have two components, design and operational planning. One key problem of operational planning is the Vehicle Scheduling Problem (VSP), which aims to plan vehicle trips. If the VSP is solved in an effective way, it can also have a positive impact on operation costs such as fuel consumption, the carbon foot-print, driver’s wages, number of purchased vehicles, and by all means, the quality of service. In this project, the VSP will be analyzed through a decentralized architecture using Distributed Artificial Intelligence (DAI). The main purpose is to allow the VSP to work in real time, making each part of the system capable to adapt and evolve, thus favoring reactivity to perturbations and constant changes. The propose approach will be validated using a case study based on a Colombian BRT.