Article ID: | iaor19941657 |
Country: | Netherlands |
Volume: | 56 |
Issue: | 1 |
Start Page Number: | 15 |
End Page Number: | 29 |
Publication Date: | Jan 1992 |
Journal: | European Journal of Operational Research |
Authors: | King Malcolm, Phythian Gary J. |
Keywords: | artificial intelligence: expert systems |
Assessing customers’ tender enquiries to decide whether to bid or not is important for companies where the cost of tendering is high. In the case considered the expertise of two managers involved in assessing enquiries was developed into an expert support system. Knowledge was elicited from these managers by asking them to consider previous tenders and specify the factors used in discriminating between them. Their responses were represented in repertory grids. Analyses of these grids led to maps which showed the hierarchical nature of the relationships between the key factors used by the managers. The maps were confirmed by the managers and then rules were developed using statistical techniques on the data in the grids. The resulting expert support system improved the consistency of the enquiry review group, revealed unexpected aspects of the managers reasoning and was generally welcomed by the company.