This paper outlines the application of a knowledge-based expert system to assist the furnace production staff in diagnosing and correcting Electron Beam (EB) guns, which are used to melt titanium. The characteristics of a titanium melting facility, its business objectives, and the responsibilities of its furnace operators are reviewed. Alternative solutions to the problem of accurately troubleshooting EB gun failures are discussed. The chosen solution (i.e. the knowledge-based system) was in alignment with the company's on-going Continuous Improvement efforts designed to improve operational efficiencies and thus was met with support. Next, we discuss the costs and benefits associated with the system. The project has a payback of less than two years and produces a modest positive cash flow over a three year period. Other potential savings in the form of additional product revenues, however, could be in excess of $1.6 million dollars. A description of the system architecture is provided along with a review of the knowledge acquisition process. Development, testing and implementation of the system is reviewed. Finally, the impacts of the system on the company are discussed. The impacts include improved task performance by providing a more structured problem solving environment, improved training and significant reductions in employee learning curves (50-100%), improved equipment diagnostics, and support for continuous improvement initiatives.
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Artificial Intelligence