LAUSR

Purpose This paper aims to propose an intelligent system that serves as a cost estimator when new part orders are received from customers. Design/methodology/approach The methodologies applied in this study were case-based reasoning (CBR), analytic hierarchy process, rule-based reasoning and fuzzy set theory for case retrieval. The retrieved cases were revised using parametric and feature-based cost estimation techniques. Cases were represented using an object-oriented (OO) approach to characterize them in n-dimensional Euclidean vector space. Findings The proposed cost estimator retrieves historical cases that have the most similar cost estimates to the current new orders. Further, it revises the retrieved cost estimates based on attribute differences between new and retrieved cases using parametric and feature-based cost estimation techniques. Research limitations/implications The proposed system was illustrated using a numerical example by considering different lathe machine operations in a computer-based laboratory environment; however, its applicability was not validated in industrial situations. Originality/value Different intelligent methods were proposed in the past; however, the combination of fuzzy CBR, parametric and feature-oriented methods was not addressed in product cost estimation problems.