SCAN: a decision support system for railroad scheduling

Abstract

The authors present an overview of a decision-support model for the tactical scheduling of railroad traffic which is meant to support the medium-term issues facing a railroad. This initial version of the schedule analysis (SCAN) system is based on notions from both simulation and combinatorial optimization, and is designed to provide schedulers with a tool for scheduling which provides real-time response. After describing the conceptual and algorithmic underpinnings of the SCAN system and its associated user interface, the authors use examples taken from a major railroad to illustrate the capabilities and limitations of the current system. Preliminary results from the use of this system at a major railroad are also discussed.<>