3 Presentations

• 
  04:00 PM - 04:30 PM

  Optimization and Analysis of Railway Timetables

  • Eva Barrena Algara, presenter, CIRRELT - HEC Montréal
  • David Canca Ortiz, Universidad de Sevilla
  • Leandro C. Coelho, Université Laval
  • Gilbert Laporte, HEC Montréal
  • Alejandro Zarzo Altarejos, Universidad Politécnica de Madrid

  In this work we design railway timetables having into account the users and the service provider points of view. So, we first propose a method to characterize the demand in an approximate way by adjusting aggregate demand functions along the whole planning horizon and secondly, we present a non-linear integer programming model to suit the timetable to a dynamic behavior of demand. The model enables the analysis of measures of the quality of service, such as average waiting times and the number of passengers waiting, in contrast to measures of the network performance and profitability, such as the number of services or the train capacity.

• 
  04:30 PM - 05:00 PM

  Efficient Ways of Designing Optimal Vertical Alignment of Roads

  • Shahadat Hossain, presenter, UBC

  During road construction, determining an optimal vertical alignment maintaining design constraints is an important task. An optimal vertical alignment finds a smooth road profile that minimizes the earthwork operations and overall construction costs from a given ground profile. In this talk, we propose new models for automating the design of vertical alignment. The models include features for modeling side-slopes and physical blocks in the terrain. In order to work towards an interactive design tool, we also focus on techniques to improve the solution time.

• 
  05:00 PM - 05:30 PM

  Metaheuristic for the Integrated Approach to the Freight Train Routing and Block-to-Train Assignment

  • João Telhada, presenter, Operations Research Center, University of Lisbon
  • Martim Moniz, Graphes et Optimisation Mathématique, Université Libre de Bruxelles
  • Diogo Silva, Department of Statistics and Operations Research, Faculty of Science, University of Lisbon

  Freight railroad transportation is a critical issue in an economic setting with growing concern about pollution and about road congestion. With an efficient management, the rail option for freight transportation may lead to more competitive costs. In this article, we address one part of the complex issue of freight railroad transportation, which deals with the definition of train routes and, simultaneously, with the way freight are carried from their origins to their destinations. The integration of these two problems is called Train Design. A new genetic algorithm is proposed to solve this problem, which tries to find a freight processing order so that the iterative dispatch subproblems leads to a lower global cost solution. Some results show the effectiveness of the proposed algorithm.