Session Detail Information
Add this session to your itinerary

Cluster :  Railroad Applications

Session Information  : Monday Oct 25, 16:30 - 18:00

Title:  RASIG Student Paper Contest Presentation
Chair: Pooja Dewan,United States, Pooja.Dewan@bnsf.com

Abstract Details

Title: RASIG Student Paper Contest
 
Abstract: RASIG (Rail Applications Special Interest Group) a subdivision of INFORMS and Railway Age are sponsoring a student research paper contest on Management Science in Railroad Applications.
  
Title: Heterogeneity and Reliability of Railway Services
 Presenting Author: Michiel J.C.M. Vromans,Department of decision and Information Management, School of Management, Erasmus University Rotterdam, Room F1-76, P.O.Box 1738, 3000 DR Rotterdam, Netherlands, MVromans@fbk.eur.nl
 Co-Author: Rommert Dekker,Erasmus University Rotterdam, PO Box 1738, 3000 DR Rotterdam, Rotterdam NL, Netherlands, rdekker@few.eur.nl
 Leo Kroon,NS Reizigers, Department of Logistics, HGB IV, 9.09 P.O. Box 2025, 3500 HA Utrecht, Netherlands, L.Kroon@fbk.eur.nl
 
Abstract: Reliability is one of the key factors in transportation. This paper examines reliability in public railway systems. To reduce the propagation of delays between the trains we attempt to decrease the interdependencies by reducing the running time differences per track section and by thus creating more homogeneous timetables. We use network wide simulation for the analysis of the alternative timetables. We report on both theoretical and practical cases. Besides a comparison of different timetables, also general timetabling principles are deduced.
  
Title: New Approaches for the Train Dispatching Problem
 Presenting Author: Guvenc Sahin,Department of Industrial and Systems Engineering, University of Florida, 303 Weil Hall, Gainesville FL 32611, United States, guvencs@ufl.edu
 
Abstract: Running freight trains on time and reducing travel times are becoming increasingly important as railroads face the challenge of truck competition in an environment in which quick response and just-in-time requirements to fulfill demand are establishing as an industry standard. Most freight railroads operate over extensive lengths of single line tracks, in which a train has to be pulled over to cross or overtake another train. The train dispatching problem aims to determine detailed train movements and timetables over a rail network, under various operational constraints and restrictions, in order to minimize train deviations from the planned schedule. Our motivation is in real-time scheduling to help dispatchers that are controlling the traffic in real time make the right decisions on which train will stop in train conflicts, as new data about updated train positions becomes available. In this paper we give a new integer programming formulation of this problem based on a space-time network, propose some exact and heuristic algorithm to solve it and present some computational results. Our approach also includes some realistic constraints like maximal allowed delays and restricted stopping stations for each train individually that have not been considered before.
  
Title: A Network Flow Approach for Railroad Crew Scheduling
 Presenting Author: Balachandran Vaidyanathan,Doctoral Student, Industrial & Systems Engineering, University of Florida, Gainesville FL 32611, United States, vbala@ufl.edu
 
Abstract: In this paper, we study one of the most important railroad optimization problems, the crew scheduling problem, and develop network flow based approaches to solve this problem. The crew scheduling problem is to assign crew (train operators) to scheduled trains over a time period (generally for a week), so that the crew cost is minimum and the schedule honors many operational and contractual requirements. The cost of a crew consists of wages, accommodation cost, and deadheading cost. This problem is a very large-scale optimization problem and is highly constrained due to various labor union contracts and federal labor laws. Fortunately, the problem can be decomposed into separate problems, one for each or a set of crew districts. We have developed network flow based models which can be used to solve the crew scheduling problem in real-time. We formulate the crew scheduling problem as a minimum cost flow problem (for non-overlapping crew districts), and as a multicommodity flow problem (for overlapping crew districts) and solve them using network flow algorithms. We present preliminary computational results of our algorithms on the data provided by a major US railroad.
  
Title: An Optimization Model for Assignment of Empty Cars in Railroad Networks
 Presenting Author: Amar Kumar Narisetty,Purdue University, 110 East Columbia Street, Apt #202, West Lafayette IN 47906, United States, anariset@purdue.edu
 
Abstract: An important problem faced by large railroad companies is how to best assign supply of empty freight cars to customers demand. The assignment decision needs to be based on the location of empty cars, type of empty cars and ur- gency of demand. Given the large size of the problem (i.e., the number of customers and supply locations) manual solutions will often be diĀ±cult to obtain and very suboptimal. In this paper we present a Simple Optimization Model (SOM) for the assignment of empty freight cars with the goals of (i) reducing transportation time and miles travelled by empty cars (ii) improving on-time performance and (iii) improving customer satisfaction. The model is embedded within a simulation tool that enables the evaluation of (i) new operating policies (ii) the importance of precise demand and supply forecast, etc. Currently SOM is in an advanced development and implementation stage and will be part of a major railroad's empty car management system in which real time assignments are made. An intelligent empty freight car assignment model is an indispensable tool to keep railroads operating costs low and help them stay competitive, while providing satisfactory service to the customers.