A simplified neuro-fuzzy inference system (S-NFIS) tool for criteria matching

  • Authors:

  • Alex T. H. Sim;Vincent C. S. Lee

  • Affiliations:

  • School of Business Systems, Monash Univ., PO Box 63B, Clayton, Wellington Road, Victoria 3800, Australia and Info. Sys. Grp., Faculty of Comp. Sci. and Info. Sys., Univ. of Tech. Malaysia, 81300 S ...;School of Business Systems, Monash University, PO Box 63B, Clayton, Wellington Road, Victoria 3800, Australia

  • Venue:
  • Design and application of hybrid intelligent systems
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

A Simplified Neuro-Fuzzy Inference System (S-NFIS) tool for evalnating counter-party (e.g. a customer and a vendor) criteria is proposed. The proposed S-NFIS, which derived and made unique from Jang's [3] ANFIS structure, allows one party (e.g. a vendor) to define a set of rules with specific weight derived from its business objectives. The rules represent this party's criteria on targeted market segment. Similarly, a set of criteria from the counterparty (e.g. a customer), representing the desired purchasing goals in rules from, is also incorporated. The proposed S-NFIS utilises the degree of membership and the concept of weight to compute the summation of all possible first-order equations generated from the SNFIS structure. The output from S-NFIS, expressed as a percentage offers an overall indication of the customer's demand and vendor's supply criteria mismatches. Based on the size of the gap, either of the party can therefore make own appropriate action towards each other. An illustrative example is presented to demonstrate the computational procedures of the proposed S-NFIS in a typical automotive transactional process.