Integrating Quality Function Deployment with Fuzzy Cognitive Maps for Resolving Correlation Issues in the Roof Matrix

Abstract

The roof matrix represents correlations among engineering characteristics (EC) in the house of quality (HoQ) in Quality Functions Deployment (QFD). Correlations are usually measured qualitatively and omitted in the analysis. However, ignoring them may cause duplication of effort, decreased product performance, and unsatisfied customer requirements (CR). Hence, this paper intends to propose an approach to considering the correlations quantitatively. Fuzzy Cognitive Maps (FCM) were used for this purpose. Additionally, Axiomatic Design (AD), for examining relationships between CRs and ECs, and Fuzzy Analytic Hierarchy Process (FAHP) with the Extent Analysis (EA) were used for checking the consistency of the evaluations. The proposed approach was applied in a sheet metal die-making company for ranking CRs and ECs. Results show that FCM enables analysing the quantitative roof matrix practically. The square roof matrix that supports FCM's adjacency matrix structure successfully represents asymmetric relationships among ECs. Integrating the correlations into the analysis resulted in a change in the final ranking. It also helped determine the most manageable ECs, better satisfiable CRs, and most critical/least manageable ECs.

Using Multi-Criteria Decision Making in Quality Function Deployment for Offshore Renewable Energies

“…Newer methodologies have also been used in QFD in recent case studies, including hesitant fuzzy linguistics [36,73,74], fuzzy DEMATEL [39,75], grey relational analysis (GRA) [37,42,46,47,76], interval-valued fuzzy numbers (IVFN) [65,77], fuzzy Delphi method (FDM) [37,39,40,46,47,49,75], complex proportion assessment method (COPRAS) [78], ordinal priority approach-combinative distance-based assessment (OPA-CODAS) [79], social network analysis (SNA) [80], multi-stepwise weight assessment ratio analysis (M-SWARA) [81], entropy weight method (EWM) [38], double hierarchy hesitant linguistic term sets (DHHLTSs) [36], organísation, rangement et synthèse de données relarionnelles (ORESTE) [36], best-worst method-based [39,82], combined compromise solution (Co-CoSo) [43,70], Dempster-Shafer Theory (DST) [83], fuzzy cognitive maps (FCM) [84], proportional hesitant fuzzy linguistic term sets (PHFLTSs) [85], cumulative prospect theory (CPT) [85], fuzzy measuring attractiveness by a categorically based evaluation technique (FMACBETH) [40], and fuzzy full consistency method (FUCOM) [35,86].…”

“…To rank the WHATs, methodologies such as TOPSIS [67] and the Kano model [92] are also integrated in the QFD methodology. In the roof matrix, the DEMATEL [8] and fuzzy cognitive maps [84] methodologies are taken into consideration to solve contradictions between technical requirements for the customer-oriented product design.…”

“…In the field of Manufacturing and materials, MCDM approaches in QFD have become an important tool in decision-making processes, allowing decision makers to select the best processes in various case studies, such as the improvement of welding process selection [59], the performance in metallurgy to reduce the defects in casting and lean tools [94], and manufacturing processes using CNC machines in sheet metal companies [84].…”