A distance-based methodology for increased extraction of information from the roof matrices in QFD studies
提出一种基于曼哈顿距离的方法，将屋顶矩阵中的相关性数据整合到技术特性的最终权重中，帮助设计者更高效地选择技术特性，避免重复工作或解决矛盾。
Quality Function Deployment (QFD) is a process in which customer needs are operationalised into deliverable Technical Characteristics (TCs) at the design stage. A system of matrices known as the House of Quality (HOQ) works collectively to produce Final Weightings (FWs) for TCs, enabling prioritisation and focusing design activity. In prioritising TCs, QFD practitioners often fail to fully integrate the diverse information within the HOQ. In this article, we address the inclusion of ‘Roof Matrix Correlations’ (RMCs). We show that, while other heuristics have been developed to integrate RMCs, they each have limitations and only result in changes to the FW values. We present a methodology based on the Manhattan Distance Measure (MDM) that integrates RMC data into the FWs, but also measures the overall nature and level of intercorrelation within the matrix. This facilitates a more efficient selection of TCs because the MDM provides a consistent informational basis for substituting negatively correlated TCs with better alternatives, and reducing duplication of effort in cases of highly positively correlated TCs. Application of the method is illustrated through re-analysis of a well-known, published QFD example. Our approach can help practitioners to avoiding duplicating effort or to address contradictions between TCs in a timely fashion.