Modelos integrados de optimización de transporte público: modelos de programación lineal entera para la calendarización de viajes y enrutamiento de pasajeros.

Abstract

Public transportation plays a vital role in urban mobility, and its efficient planning is essential for reducing travel time and enhancing the quality of life of commuters. The planning of public transportation systems involves several challenging problems, such as line planning, timetabling, and vehicle scheduling. Among these, timetabling is the first step that interacts directly with passengers and can be impacted by human behaviour. The primary objective of timetabling is to schedule arrival and departure times for the trips of different lines at each station, with the aim of minimizing passenger travel time. Numerous approaches have been proposed in the literature to solve the timetabling problem, but one of the most significant challenges is the feedback interaction with passenger routing. If the timetabling model does not consider passenger routes or the routes are fixed beforehand, it may lead to suboptimal timetables for passengers. This has led to academic interest in the study of integrated models that address both timetabling and passenger routing. This work considers two models for timetabling: a classic model that assumes fixed routes for passengers and an integrated model that considers both timetabling and passenger routing. The models were implemented and tested on five instances of public transport systems. Additionally, a heuristic was proposed for the solution of the largest instance. The computational results showed that the integrated model outperforms the classic model in terms of passenger travel time. Overall, the study highlights the importance of considering passenger routing in the timetabling problem. It also emphasizes the potential benefits of using integrated models that optimize both timetabling and passenger routing.