Written by Muqi Wulan.
Let’s assume that you have had knowledge of gathering and understanding the customer’s needs or user’s demands for a particular product or service. Therefore, after customer requirements are clearly identified through certain methods, the next task that engineers will confront with is, from the perspective of engineering, to translate these customer requirements into engineering characteristics or functional requirements which can be recorded as product design specifications (PDS). PDS is “the basic control and reference document for the design and manufacture of the product” (Dieter and Schmidt 2009: 109). Quality Function Deployment (QFD) is the graphical and structured method for this purpose and its applications have been beyond product design (Akao 1994: 339). QFD was developed by Yoji Akao in Japan in 1966 and rapidly adopted by the Japanese automobile industry. By the mid-1980s many US companies were using QFD in various areas of automotive, defence and electronics.
“Quality Function Deployment” is a direct translation of Japanese characters Hin Shitsu, Ki No, Ten Kai. In Japanese, the phrase means something like the strategic arrangement (deployment) throughout all aspects of a product (function) of appropriate characteristics (quality) according to customer needs (Cross 2008: 122). The core structure in QFD is the House of Quality (HoQ). Its most complete configuration is shown in Figure 1. There are 8 “Rooms” or sections to be filled in. Each labelled “Rooms”, from 1 to 8, is a structured, systematic expression of a product development team’s understanding of an aspect in the overall planning process for a new product or service (Cohen 1995: 12). Correspondingly 8 steps are required to input information into these “rooms” until the whole HoQ is populated.
Figure 1 – The HoQ converts customer requirements into engineering characteristics
(Dieter and Schmidt 2009: 101)
At first Figure 2 gives an overview on a fully defined HoQ based on an example. It shows the design process of making chocolate chip cookies, and mapping customer requirements to functional or engineering characteristics. Later the procedural method of establishing the HoQ will be described step by step.
Figure 2 – An example of the HoQ (QFD Online 2011)
Step 1: Identify customer requirements (CRs) in Room 1.
The information from identifying the needs of the customer or end user is compiled as CRs in Room 1 in the form of customer requirements and their importance ratings. These are the initial input of a HoQ. Room 1 is also called the section “Whats”. In this example, five customer requirements are identified as shown in Figure 3. They are: good texture, generous portions, tastes good, low price, and appetising appearance.
Importance ratings on CRs follow the simple method below.
– Customers prioritise CR categories (1 to 5);
– Customers prioritise CRs within a category (1 to 5);
– The CR global importance is the product of the category and the within category priority (1 to 5) X (1 to 5). The value of importance for each CR is filled in the column “Weight/Importance”.
Figure 3 – Identified CRs in Room 1 (QFD Online 2011)
Step 2: Identify engineering characteristics (ECs) in Room 2.
ECs describe the product’s performance as a whole and its functional features to meet CRs. The alternative term of the ECs is functional requirements. At this step, the CRs identified in Room 1 will be translated into the language of ECs which are expressed in parameters, design variables and constraints. These are the input of the Room 2, also called the section “Hows”. In the case of chocolate chip cookies, 9 ECs are defined to put in Room 2. They are colour, tensile yield strength, tensile ultimate strength, weight, size (diameter), thickness, average Hedonic scale rating, cost per cookie, and density of chocolate chips (Figure 4).
Figure 4 – Identified ERs in Room 2 (QFD Online 2011)
An additional row is included to illustrate the improvement direction in each EC which is considered to result in an improvement in product performance (Figure 4). Symbols indicating preferred improvement direction of each EC are placed at the top of Room 2 (Table 1). Usually the row “units of ECs” is combined with the “target values” in Room 8.
Table 1: Improvement direction codes
Step 3: Build a correlation matrix of the ECs in Room 3.
The triangular roof of HoQ (“Room3”) is utilised to establish the correlation matrix on how the engineering characteristics (ECs) support or impede one another. When pairing any two of ECs, you need to decide if improving one EC causes a deterioration or improvement in the other EC. Where there is a conflict between two ECs, an engineering trade-off and negative relation exist. If improving one EC automatically leads to an improvement in other ECs, the interaction of these ECs is positive. Three or four strength levels are commonly symbolised for representing the correlations. Their coding is shown in Table 2. The example (Figure 5) uses four levels: strong positive, positive, negative, and strong negative.
Table 2 – Correlation strength codes
Figure 5 – The correlation matrix in Room 3 (QFD Online 2011)
Step 4: Build a relationship matrix between the CRs and the ECs in Room 4.
Having input the required information of CRs and ECs into Room1, Room 2 and Room 3, Step 4 will go to the main body of HoQ – relationship matrix in Room 4. Its structure is the standard two-dimensional matrix with cells that relate individual CRs to the ECs. Each cell is marked with a symbol that indicates the strength of the combination between the CR of its row and the EC of the column. It means how significant the EC is in satisfying the CR. The coding scheme for each cell is given as a set of symbols that represent an exponential range of numbers (e.g., 9, 3, 1, and 0) (Table 3). The example is shown in Figure 6.
Table 3 – Relationship strength codes
Figure 6 – The relationship matrix in Room 4 (QFD Online 2011)
You may have noticed that terms and codes in HoQ are allowed to vary a little in different applications of QFD. However, throughout the same project, the terminology and coding scheme should be consistent in order to share common understanding.
Summary
Quality Function Deployment (QFD) is a comprehensive method of matching customer requirements to engineering characteristics. In QFD, the core structure is the House of Quality (HoQ). The information required for the HoQ is constructed in a step-by-step process, which consists of 8 steps. The tasks in the former 4 steps are briefly presented as follows.
Step 1: Identify customer requirements (CRs) in Room 1.
Step 2: Identify engineering characteristics (ECs) or functional requirements in Room 2.
Step 3: Build a correlation matrix of the ECs in Room 3.
Step 4: Build a relationship matrix between the CRs and the ECs in Room 4.
The rest 4 steps for HoQ will be described in the material for Level 2.
Back to QFD
Back to Planning and Clarification
Back to MAE Design Model
Selected References
Akao, Y. (1994) ‘Development history of Quality Function Deployment’. in QFD: the customer-driven approach to quality planning and deployment. ed. By Mizuno, S. and Akao, Y. Tokyo: Asian Productivity Organization
Cohen, L. (1995) Quality Function Deployment; How to make QFD work for you. Reading: Addison-Wesley
QFD Online (2011) House of Quality (QFD) example [online] available from <http://www.qfdonline.com/qfd-tutorials/house-of-quality-qfd-example/> [26 July 2011]