Practical Manual of Quality Function Deployment

In this chapter the basic concepts of Quality Function Deployment are presented together with the context in which the method can be applied. In more detail, following a 5Ws 1H scheme, Where, What, Who, When, Why, How (5W1H), there are some basic questions to be answered:

In this chapter the first two phases of the QFD framework will be discussed:In the QFD “strategy” area, we will calculate:Priority is a numerical value we obtain as an output of a matrix such as:Displaying customer segment priorities and their deployment with a graph will allow for focused planning of the first activity around the customer: the open answer “gemba” interviews with a small group of selected customers. After collecting several raw data, phrases and expressions from the customers interviewed, the QFD team will start brainstorming sessions making use of the Jiro Kawakita (KJ) method, to display, organise and select ideas and data. The output of KJ is a set of sentences that form the so called Demanded Quality Deployment Chart. A QFD questionnaire will then be prepared and sent to a large group of customers, contacts and potential customers. Following this questionnaire competitor benchmarking starts. The average scores obtained from the questionnaire will be inserted in a matrix called “Preplan” the output of which will be an assessment of the customer’s requests (Demanded Qualities) through an index called Demanded Quality Weight. An educational case study (the development of “Royal” classic bicycle) is used to describe each step of the method. Exercises for the reader are proposed at each stage.

In this chapter the rest of the QFD framework phases will be discussed:The reader can continue developing her/his QFD project, using the exercises as a guide. The output from Preplan is the customer requests priority, also called Demanded Quality Weight. This index is used in creating the Quality Function Deployment House of Quality in order to obtain the priority of the product Quality Characteristics. Characteristics help to build the product Functions (for a bicycle, “It lights up the road ahead” or “It brakes”). Functions are weighted from both the customer and the technician’s point of view. In the QFD model, Functions then lead the team to the Mechanisms, macro-systems which carry out the product functions, for example, the “Steering mechanism” or the “Transmission mechanism” or the “Dirt protection mechanism”. After selecting either existing or new Technology that can be adopted for the new product, we establish their priority using a mechanism-technology matrix. Now Parts can be deployed with the input from technology selected and product characteristics. The team can set the Costs of the parts. Parts for the product are either made or purchased from suppliers and the final product must be assembled: Process phases can be deployed and their priority calculated. At the end an analysis of Faults is implemented, in order to find out, from the customer’s point of view, what the most serious faults that can occur are.

Fuzziness concept, i.e. vagueness, is an essential part of the human mind and philosophy, like nature and the environment, which are guided by some mathematical rules but also by vagueness. A simplified approach to fuzzy mathematics applied to QFD is proposed in this chapter. The aim is to give the reader some basic tools to be used if the team wants to manage fuzziness in QFD projects. The reader can consider, for example, how much fuzziness exists when she/he fills in a questionnaire or puts a symbol of correlation inside a QFD matrix. In other words, interviewees, QFD team people and specialists from the departments involved in a QFD project, all are humans and so, they can sometimes change their opinion or are unable to express what they really feel. Starting from “fuzzy sets” and operations with fuzzy sets, we will focus on the so called “fuzzy numbers”, that are particular fuzzy sets. Then, the mathematical operations between fuzzy numbers, like addition, subtraction, product and division are presented with examples. At a later stage we will return to the traditional, or “crisp”, mathematics and workflow, which the previous chapters are founded on. We will review, from a fuzzy point of view, some QFD process phases, like fuzzy questionnaire design, fuzzy Preplan matrix and fuzzy QFD matrices. We will learn how to calculate the final ranking of sentences, examined in the QFD matrices, and how to set a ranking order through “defuzzification” of fuzzy numbers.

Reading Quality Function Deployment case studies can be useful for understanding how professionals applied QFD and what the specific findings were. In this chapter, we take a look at eight real QFD projects, four in detail and four others briefly. However, imaginary names have been used for the companies involved. The Geotherm case study focuses on a small company that sells geothermal heat pumps, it is a longitudinal study, and in fact QFD was applied on two occasions. Wepartner is a large association of small and artisan companies, that applied QFD to study a territory and its potentialities. Citymove is a public transport company and the aim in that case was to understand who the customer was and how the service offered could be redefined. Elight shows the case of a lighting company which was going to plan the new product line. Paint is a paints producer that deployed QFD to design its production process better. Insure is a telephone and web based insurance company and the case study concerns the organisational planning of the Insure call centre. Mobile 1 is a project our team managed some time ago concerning new product planning for a mobile phone. Stone provides services to marble workers. Our target was to study the needs of registered companies and increase service performance through QFD. Each case has its own peculiarities that will be discussed. The reader needs a certain amount of knowledge about QFD, so I recommend reading Chaps. 2, 3 and 4 beforehand.