REPOMAN: AN OVERALL COMPUTER-AIDED DECISION SUPPORT SYSTEM FOR PLANNING RAIL REPLACEMENTS

The North American railroad industry spends hundreds of millions of dollars annually on installation of rails. Decisions made by different rail experts are likely to be inconsistent and result in an inefficient utilization of resources. This paper describes a computer-aided decision support system that uses expert systems, knowledge systems, and network optimization methods to efficiently arrive at a solution to the rail replacement scheduling problem. The scheduling system efficiently utilizes the automatically collected rail profile data obtained from the electronic rail profile measurement technology to produce a five-year rail replacement plan. This system has been successfully used to analyze about 20 percent of Burlington Northern Railroad's mainline track.