Modelos de generación de viajes motivo estudio: caso Universidad de la Costa

In the present work, generation models were estimated for travel study reasons of the University of the coast. The data used was obtained from a source-destination survey carried out for university students. Generation models were estimated using at least three methods: Multiple Linear Regression (R...

Full description

Autores:
Gómez Gómez, Aixa Liliana
Medina Romero, Jorge Alberto
Tipo de recurso:
Trabajo de grado de pregrado
Fecha de publicación:
2018
Institución:
Corporación Universidad de la Costa
Repositorio:
REDICUC - Repositorio CUC
Idioma:
spa
OAI Identifier:
oai:repositorio.cuc.edu.co:11323/1975
Acceso en línea:
https://hdl.handle.net/11323/1975
https://repositorio.cuc.edu.co/
Palabra clave:
Generación de viajes
Análisis por categorías
Análisis de clasificación lineal múltiple
Regresión lineal múltiple
Logit ordinal
Trip generation
Analysis by categories
Multiple linear classification analysis
Multiple linear regression
Ordinal logit
Rights
openAccess
License
Atribución – No comercial – Compartir igual
Description
Summary:In the present work, generation models were estimated for travel study reasons of the University of the coast. The data used was obtained from a source-destination survey carried out for university students. Generation models were estimated using at least three methods: Multiple Linear Regression (RLM), category analysis (AC) and Ordinal Logit (LO). According to the results, the models (LO) showed greater econometric consistency and better indicators of goodness of fit. The models used are key for strategic planning in terms of mobility for the university of the coast. The generation of trips is a process that allows to relate the activities of the population with the trips that are made, the latter are intimately linked to the socioeconomic characteristics of the population, this relationship can be estimated through generation models used in the present draft. The average travel rates obtained with the travel models by categories and Ordinal Logit were analyzed, as well as models were generated by the Multiple Linear Regression method and the models obtained through the different statistical tests were evaluated to select the most reliable.