Fuzzy methodology appears to be one of the most promising tools for managing the ambiguity and indetermination that characterize the evaluation of suppliers' contribution to product development. This kind of evaluation, in fact, is particularly complex, given the difficulty in identifying, circumscribing and quantifying the supplier's co-design performance. Moreover, this evaluation is mainly perceptive and strongly influenced by subjective judgements, being linked to the experience and common sense of decision-makers (purchasers, designers, etc). In this paper we present a fuzzy expert system prototype able to manage the evaluation process and to offer a reliable measurement of the contribution of the suppliers in new product development (NPD). The objectives of this study are the following: (a) to develop a vendor rating tool based on fuzzy logic; (b) to refine it using a neural application; and (c) to compare it with an ordinary least squares (OLS) regression, which reproduces the traditional method of evaluation. The evaluation tool was tested in a company where co-design is a critical factor for success and the tool's potentials and limits are pointed out.
Evaluation of supplier contribution to product development: fuzzy and neuro-fuzzy based approaches
NASSIMBENI, Guido;
2003-01-01
Abstract
Fuzzy methodology appears to be one of the most promising tools for managing the ambiguity and indetermination that characterize the evaluation of suppliers' contribution to product development. This kind of evaluation, in fact, is particularly complex, given the difficulty in identifying, circumscribing and quantifying the supplier's co-design performance. Moreover, this evaluation is mainly perceptive and strongly influenced by subjective judgements, being linked to the experience and common sense of decision-makers (purchasers, designers, etc). In this paper we present a fuzzy expert system prototype able to manage the evaluation process and to offer a reliable measurement of the contribution of the suppliers in new product development (NPD). The objectives of this study are the following: (a) to develop a vendor rating tool based on fuzzy logic; (b) to refine it using a neural application; and (c) to compare it with an ordinary least squares (OLS) regression, which reproduces the traditional method of evaluation. The evaluation tool was tested in a company where co-design is a critical factor for success and the tool's potentials and limits are pointed out.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
