Discussion on Papers 29-30
-
Published:1985
1985. "Discussion on Papers 29-30", COMPUTER TECHNOLOGY IN CONSTRUCTION: Proceedings of a conference organized by The Institution of Civil Engineers and held in London on 25-27 September 1984
Download citation file:
MR M.A. SLINN, Mott, Hay & Anderson
With reference to Paper 30, it needs a structural analyst to make sensible use of a structural analysis program and to check that its output is relevant and reasonable. These latter aspects are important because nobody has yet written any significant software that is 100 percent error proof. Will it require an expert to make sensible use of an expert system?
MR TAFFS
It will require an expert to tell whether the expert system has made a mistake, but there is nothing to stop anyone using such a facility. The beauty of the technique is that it can learn along with its users. When a mistake is made in a conventional program, locating it, then correcting, testing and reissuing the program, is a laborious process; whereas the expert system will readily accept a new statement or rule and absorb it into its knowledge base without the user having to bother about the structure of the program.
