EXPERT SYSTEMS FOR OPTIMAL REAL TIME MANAGEMENT OF RAILROAD ROLLING STOCK.

Research output: Contribution to journalArticlepeer-review

Abstract

The paper describes the basic theoretical three-dimensional spatial-temporal network construct that forms the basis for the multicommodity network optimization models that are at the heart of the expert systems. The real-time management of a heterogeneous fleet of freight cars (boxcars) is used as the example in formulating the approach. A second application is a locomotive distribution system (LDS) which is being developed for a major U. S. railroad. This system allocates all road locomotives to trains or maintenance facilities so as to minimize locomotive costs while satisfying a large number of real operational constraints. The system anticipates the supply, demand and maintenance requirements for the locomotive fleet for a seven day planning horizon, optimizes the flow of horsepower for that planning horizon, then assigns actual locomotive units to trains for the next 16 hours. The paper describes the various elements of the LDS.

Original languageEnglish (US)
Pages (from-to)23-32
Number of pages10
JournalRail International
Volume19
Issue number1-2
StatePublished - Jan 1 1988

All Science Journal Classification (ASJC) codes

  • Mechanical Engineering
  • Transportation

Fingerprint Dive into the research topics of 'EXPERT SYSTEMS FOR OPTIMAL REAL TIME MANAGEMENT OF RAILROAD ROLLING STOCK.'. Together they form a unique fingerprint.

Cite this