Development of an innovative neuro-fuzzy assessment system for the european foundation for quality management

In growingly competitive business environment, numerous organizations adopt the total quality management (TQM) approach to achieve business excellence. To monitor the progress towards business excellence, thousands of organizations across the world use self-assessment on a regular basis. The European Foundation for Quality Management (EFQM) is among the most popular ones. However, the current selfassessment methods in EFQM model have some drawbacks and problems. Critical review of self-assessment models from literature showed that the majority of the assessment and self-assessment models developed were ambiguous, assumed assessment is only limited to certain and precise data and inability to consider the empirical investigation and expert knowledge in scoring and lack of a non-linear methodology for assessment system. Besides, none of the examined models developed consider simultaneous knowledge and experience of experts and historical behavior of variables. Therefore, this research aims to develop a comprehensive intelligent assessment system using Neuro-Fuzzy system (Hybrid System) to overcome the uncertainties and complexities in the EFQM model and design fuzzy decision making model for best selection and ranking of Area for Improvement (AFI) in EFQM model. A new assessment system based on Neuro-fuzzy is introduced and developed incrementally to address the deficiency in the existing models. Three different models have been introduced in this work: The first model is an assessment system based on the fuzzy inference system (FIS) in EFQM business excellence framework under conditions of imprecise (uncertain) data and nonlinear relations. The second model considers simultaneous knowledge and experience of experts and historical behavior of variables in EFQM and this model is a hybrid assessment system (Neuro-Fuzzy) which includes fuzzy inference system and adaptive neuro-fuzzy inference system (ANFIS). The third model is based on fuzzy multi criteria decision making (FMCDM) for selecting AFI in EFQM. The models were tested and verified under real condition and were implemented in Rahyab Rayaneh Alborz Company. The case had been assessed by assessors and experts of an EFQM business excellence organization and internal assessors of the companies. Then the models were analyzed using the MATLAB software. Also by comparison of classic and new model, assessors and experts agreed with outputs of the developed (new) models. The contribution of this research is modeling a new comprehensive assessment system in EFQM considering simultaneous knowledge and experience of experts and historical behavior of variables (EFQM Criteria) using ANFIS. Moreover, organizational assessment and extraction of final Score for EFQM model under conditions of imprecise (uncertain) data and in nonlinear relations using FIS and employing FMCDM for priority of AFI in EFQM model would be considered as contributions for this study. The performances of the innovative assessment system proposed in this research include 1) considering the relation between variables as a nonlinear function,2) ability to be implemented for any number of inputs and outputs, 3) providing more informative and reliable analytical results, 4) facilitating rapid assessment and decision making for managers, experts and assessors of organizations, 5) improving the FIS model efficiency by considering historical data and knowledge and experiences of experts through using hybrid assessment system (ANFIS), 6) being valid based on the hybrid system (ANFIS), due to the mean error between assessment of assessors and the output of model which was 0.000981517, 7) using FMCDM model for ranking and selecting AFIs in EFQM in practice and 8) being verified under real conditions and implemented in Rahyab Rayaneh Alborz Co. By comparison of classic and new model,assessors and experts agreed with the outputs of the developed (new) models.

Preview PDF FK 2013 12R.pdf Download (742kB) | Preview

Actions (login required)

View Item