A PC-based expert system in engineering

  • Authors:

  • G. L. Lovegrove;G. J. Curtis;R. A. Farrar

  • Affiliations:

  • Department of Electronics and Computer Science, University of Southampton, UK;Department of Mechanical Engineering, University of Southampton, UK;Department of Mechanical Engineering, University of Southampton, UK

  • Venue:
  • IEA/AIE '90 Proceedings of the 3rd international conference on Industrial and engineering applications of artificial intelligence and expert systems - Volume 2
  • Year:
  • 1990

Quantified Score

Hi-index 0.00

Visualization

Abstract

Over the last decade expert systems have been applied in many different domains with varying degrees of success. Much has been learned about techniques for knowledge representation, inference engines and user interfaces. Another important factor in the success of an expert system is to take into consideration at the design stage the person for whom the final system is intended: expert, informed user or beginner.Southampton University and Marchwood Laboratories of CEGB (Central Electricity Generating Board) have collaborated in recent years to produce an expert advisory system for selecting processes for welding. Apart from knowledge about the processes themselves, other factors have been taken into consideration such as equipment cost and availability, depreciation, running costs, the site, reject rate and quality level. This PC-based advisory system was designed for use by an informed but not necessarily expert welding engineer, for use in a laboratory, with an emphasis on the friendliness of the interface and flexibility of the system.Recent research work has concentrated on designing the software in an object-oriented way using an object-oriented language. This emphasis on encapsulation of data and relevant procedures has improved not only the efficiency of the system but also its overall design, meaning that future modifications or extensions can be incorporated with far greater ease. Current work includes the prototyping of tools, with the future aim of allowing a new user in a related area of engineering to use the tools to produce a similar expert system.