Vous êtes sur la page 1sur 12

What is QFD ?

"Time was when a man could order a pair of shoes directly from the cobbler. By measuring the foot himself and personally handling all aspects of manufacturing, the cobbler could assure the customer would be satisfied," lamented Dr. Yoji Akao, one of the founders of QFD, in his private lectures. Quality Function Deployment (QFD) was developed to bring this personal interface to modern manufacturing and business. In today's industrial society, where the growing distance between producers and users is a concern, QFD links the needs of the customer (end user) with design, development, engineering, manufacturing, and service functions. QFD is: 1. Understanding Customer Requirements 2. Quality Systems Thinking + Psychology + Knowledge/Epistemology 3. Maximizing Positive Quality That Adds Value 4. Comprehensive Quality System for Customer Satisfaction 5. Strategy to Stay Ahead of The Game As a quality system that implements elements of Systems Thinking with elements of Psychology and Epistemology (knowledge), QFD provides a system of comprehensive development process for:

Understanding 'true' customer needs from the customer's perspective What 'value' means to the customer, from the customer's perspective Understanding how customers or end users become interested, choose, and are satisfied Analyzing how do we know the needs of the customer Deciding what features to include Determining what level of performance to deliver Intelligently linking the needs of the customer with design, development, engineering, manufacturing, and service functions Intelligently linking Design for Six Sigma (DFSS) with the front end Voice of Customer analysis and the entire design system

QFD is a comprehensive quality system that systematically links the needs of the customer with various business functions and organizational processes, such as

marketing, design, quality, production, manufacturing, sales, etc., aligning the entire company toward achieving a common goal. It does so by seeking both spoken and unspoken needs, identifying positive quality and business opportunities, and translating these into actions and designs by using transparent analytic and prioritization methods, empowering organizations to exceed normal expectations and provide a level of unanticipated excitement that generates value. The QFD methodology can be used for both tangible products and non-tangible services, including manufactured goods, service industry, software products, IT projects, business process development, government, healthcare, environmental initiatives, and many other applications.

Why Need QFD ?

QFD is mainly useful for those industries that need to make decisions and please customers. QFD is all about communication and decision making, and its tools can be used in any industry. The House of Quality matrix is a universal tool that can be implemented to prioritize anything from a family budget to the complex tasks of an automobile manufacturer. The House of Quality matrix is one of the best tools available for clarifying the voice of the customer. The customer in question can vary from being a product or service consumer, a corporate executive, or even another department within the same company. In fact, the customer in question may even be ones selffor the HOQ tool is an excellent way to evaluate a complex decision and prioritize ones own requirements. Many companies have discovered the importance of asking what their customers requirements are. Unfortunately, most companies stop their dialog with their customers at simple requirements gathering. They fail to ask the clarifying questions necessary to truly understand the wants and needs of their customer. The House of Quality tool, on the other hand, helps teams to gather information such as the following:

Are any of the customers requirements more important/critical than their other requirements? How much more/less important is any given requirement when compared to the others? What are the measurable goals for fulfilling the requested requirements? Do any of the goals conflict with each other? Does satisfying any particular goal help to satisfy another? How difficult will it be to accomplish any given goal?

The House of Quality tool can assist with competitive analysis as well. In short, the HOQ matrix not only helps to capture the voice of the customer, it helps to map an effective and efficient path to satisfying that voice. In order for the House of Quality to be a valuable tool for any team or

individual, the people utilizing the tool must be at liberty to influence the decision-making process. Furthermore, teams and/or companies that care little about satisfying the wants and needs of their customers will find little value in the HOQ matrix or the Quality Function Deployment methodology. However, environments that foster micro-management and ignore their customers are generally the product of individual company cultures rather than the attributes of entire industries or disciplines. In Short, the QFD process and the House of Quality tool can be applied in most any industry to increase quality and better meet the needs of its customers. The HOQ matrix can assist in prioritizing the initiatives of anything ranging from computer processor design to hotel management. It facilitates decision making for engineers and sales people, alike. It helps to document the voice of the customer just as easily for healthcare services as for fast food services. In truth, the House of Quality is universally applicable, and probably the only building big enough to accommodate all the industries in the world.

The Logic for Notation

There has been a lot of debate over which representation is the best to use when filling in the ratings for a House of Quality: symbolic notation or numeric notation. The premise behind the debate is that there is somehow a greater inherent value in using circles, filled circles, and triangles or in using 1s, 3s, and 9s. In actuality, neither format is universally superior to the other. On the contrary, they both serve different purposes and are uniquely suited for working with different groups. Thus, just as it is important to tailor a speech or written argument to the needs of the intended audience, it is important to choose the notation for a QFD that is best suited for its target audience. The choice of notation formats depends largely on how much attention one wants to draw to the relative magnitudes between rating values. If an engineer or product manager wishes to encourage more careful analysis of the relationships, then numeric notation is often a better choice. Viewing the exponential scale while one is rating often helps stake holders to be more judicious in assigning low and strong relationship values. By the same token, if a QFD practitioner is worried about inducing analysis paralysis among the targeted stake holders, then symbolic representation is probably the preferred notation. Symbolic notation often shortens the perceived chasm between moderate and strong ratings, and helps to reduce the reluctance felt by some stake holders to provide decisive ratings. Another factor to consider when choosing a QFD notation style is the technical and/or mathematical aptitude of the target audience. Numerical notation is often more appealing to technical audiences such as engineers and scientists who like to understand the algorithms involved in the House of Quality. These audiences frequently prefer numeric notation because it replaces magical ranking behavior with a comprehendible scientific process. Symbolic notation, on the other hand, is often more appealing to business, marketing, and sales individuals. These stake holders are often distracted by numeric notation and begin to view a

Quality Function Deployment as a numbers game when presented with numeric notation. Using symbolic notation often helps these target audiences to focus on the simple relationships between requirements, rather than on the numbers and calculations involved in a House of Quality. Symbolic notation helps these stake holders view a Quality Function Deployment as the communication tool that it is rather than as the mechanized dictator that it isnt. The choice of notation when constructing a Quality Function Deployment can have a major impact on the audience involved. Numeric notation is better suited for teams that are technicallyoriented and/or teams that need to focus more carefully on their ratings. Symbolic notation is more appropriate for less-technical audiences and/or audiences that are prone to endless debate over individual ratings. In truth, there are many characteristics of a QFD that can and should be modified to meet the needs of the audience involved. Different audiences require different communication mechanisms, different meeting frequencies, different meeting durations, different leadership styles, and even different tools. However, when one is tailoring his or her Quality Function Deployment to meet the needs of its stake holders, notation style should not be overlooked. For when it comes to making team members comfortable with a House of Quality, using 1s, 3s, and 9s has a substantially different affect than do circles and triangles. Indeed, when representing a moderate relationship in an HOQ matrix, one will likely find that to three or not to three that is the question.

Youve probably heard the old adage, If you fail to plan, then youre planning to fail. That sentiment is certainly echoed in the basic principles of the Quality Function Deployment (QFD) methodology. There is another old adage concerning failure that, although not quite as recognized, is just as true: Fail to mitigate failure and you will succeed in minimizing success. (Okay, so it isnt really an old adage. I just made it up. However, you have to admit, it does sound rather catchy, and it does convey the underlying precept fairly well.) This maxim (regardless of how it is worded) is similarly echoed by the tenets of Quality Function Deployment. So then, the question ariseswhat is the best tool for prioritizing steps to mitigate potential failures: Failure Mode and Effects Analysis (FMEA) or the House of Quality (HOQ) tool? Coming from a Quality Function Deployment enthusiast, my answer may surprise you The House of Quality is a fantastic tool. It helps to capture, quantify, categorize, and prioritize the Voice of the Customer. That being said, it does have its place and its purpose, and that purpose is primarily for planning the activities required to augment and improve a particular product or service. In other words, it is designed to assist in planning how a team should affect or modify the product or service in question. It really wasnt created for the intent of analyzing the effects of failures or breakdowns and prioritizing the need to address those different failure modes.

I have heard of teams trying to use a House of Quality for the purpose of mitigating failure. There are two major problems with taking such a path: First the mathematical structure of an HOQdoesnt lend itself well to prioritizing activities for mitigating failure. Secondly, the HOQ doesnt have fields for identifying and analyzing the causes, effects, and controls related to different potential failure modes. Failure Mode and Effects Analysis on the other hand was created specifically for the purpose of analyzing potential failure modes and prioritizing activities to mitigate these failures. Thus, its mathematical structure is better suited for such a task. Additionally, it contains fields that guide its users through the analysis of different potential defects and their impact on ones customers. As far as the math goes, one should note that the House of Quality is setup on an additive scale. If a particular functional requirement satisfies two separate customer requirements, then the weightings of those two customer requirements are addedtogether (after multiplying them by their respective rating factor). Note that those weightings are not multiplied by each other. This use of addition rather than multiplication makes sensea characteristic that satisfies two requirements is probably only as important as the sum of the importance of the individual requirements. It does not become exponentially more important simply because it satisfies multiple requirements. Consider the following example: Requirement A is weighted by a customer to be just as important as Requirement B. If a particular characteristic satisfies both A and B, and satisfies each requirement to the same degree, then it is reasonable to say that is twice as important as a characteristic that just satisfies either requirement. However, it would be unreasonable to say that it is five or ten times more important than a characteristic that satisfies only A or B. The math for an FMEA, on the other hand, is essentially multiplicative in nature. The probability of occurrence rating is multiplied by the severity rating, which is multiplied by the probability of escaped detection rating. This too makes sense when considered properlyan error that will happen almost universally with catastrophic consequences is far more than twice as disconcerting as a catastrophic error that will rarely if ever happen. Similarly, such a potential failure is far more than twice as dangerous as a defect that happens universally with minimal effects. Thus, in an FMEA, the Risk Priority Number (RPN) is rightfully a calculation of the different ratings multiplied times each other. As mentioned earlier, an HOQ does not contain any fields specifically designed to walk its users through an analysis of the potential failures for a given product or service. An FMEA, on the other hand, does contain such fields, and it is in the filling out of such fields that discovery and realization often takes place for an engineering or management team. An FMEA forces its participants to not only consider a given failure mode, but to consider the effects and causes of those failures. Additionally, an FMEAencourages its users to consider any controls that might already be in place for preventing or mitigating such a failure. This step-by-step analysis helps teams to properly consider the different aspects of a potential failure so that they can properly rate it. Without such analysis and/or information gathering, team members might be prone to rashly rate the priority of addressing a potential breakdown.

In summary, although the House of Quality is a fantastic tool for capturing and quantifying the Voice of the Customer, it is not the tool of choice for prioritizing efforts to prevent failures, defects, and/or breakdowns. Failure Mode and Effects Analysis, on the other hand, is an excellent tool for prioritizing efforts to mitigate failure. In short, HOQs are excellent for mapping Critical to Customer requirements (CTCs) to Critical to Quality parameters (CTQs). However, FMEA is a better tool for quantifying the effects of failing to meet those CTQs. Because the term Quality Function Deployment has come to mean two largely different things to the quality community, I should clarify somethingAt the beginning of this article I mentioned that my recommendations on the use of FMEA vs. HOQsmight surprise people because I am a proponent of QFD. In truth, you may or may not be surprised depending on what the term QFD means to you. Depending on where you first became exposed to the term you may think that it is either synonymous with the House of Quality tool, or you may think that the House of Quality tool is simply a minor subset of the QFD methodology. In actuality, QFD was originally (and still is) a full methodology that simply utilized the House of Quality matrix as one of its tools. However, other methodologies (such as Design for Six Sigma) have come to label the series of sequential House of Quality matrices as a QFD. (For more information on this topic, take a look at the article Where Did QFD Get Its Terrible Name?.) If you are used to referring to QFD as a tool, then my opinions recommendation to use FMEA for failure mitigation may surprise you. However, if you are acquainted with the original Quality Function Deployment methodology, then my comments are probably quite unremarkable since the QFD methodology generally incorporates FMEA as an integral step.

Steps Needed to build QFD

The House of Quality (HOQ) is the first matrix that a product development team uses to initiate a Quality Function Deployment (QFD) process. This matrix is especially powerful because of the amount of information that can be documented and analyzed. QFD methodology requires that the team ask specific questions about customer needs, competitors, and how their organization will meet the challenges of providing products that delight the customer. Below are the step needed to create a House of Quality.

Step 1
Customer Requirements - The first step in a QFD project is to determine what market segments will be analyzed during the process and to identify who the customers are. The team then gathers information from customers on the requirements they have for the product or service. In order to organize and evaluate this data, the team uses simple quality tools like Affinity Diagrams or Tree Diagrams.

Interrelationship between Technical Descriptors Technical Descriptors (Voice of the organization)

Customer Requirements (Voice of the Customer)

Relationship between Requirements and Descriptors

Prioritized Technical Descriptors

Step 2.
Regulatory Requirements - Not all product or service requirements are known to the customer, so the team must document requirements that are dictated by management or regulatory standards that the product must adhere to.

Step 3.
Customer Importance - On a scale from 1 - 5, customers then rate the importance of each requirement. This number will be used later in the relationship matrix.

Step 4.
Customer Rating of the Competition - Understanding how customers rate the competition can be a tremendous competitive advantage. In this step of the QFD process, it is also a good idea to ask customers how your product or service rates in relation to the competition. There is remodeling that can take place in this part of the House of Quality. Additional rooms that identify sales opportunities, goals for continuous improvement, customer complaints, etc., can be added.

Step 5.
Technical Descriptors/"Voice of the Engineer" - The technical descriptors are attributes about the product or service that can be measured and benchmarked against the competition. Technical descriptors may exist that your organization is already using to determine product specification,

Prioritized Customer Requirements

however new measurements can be created to ensure that your product is meeting customer needs.

Step 6.
Direction of Improvement - As the team defines the technical descriptors, a determination must be made as to the direction of movement for each descriptor.

Step 7.
The Relationship Matrix - The relationship matrix is where the team determines the relationship between customer needs and the company's ability to meet those needs. The team asks the question, "if I control this technical descriptor, will it have a positive impact on the customer's need?" Relationships can either be weak, moderate, or strong and carry a numeric value of 1, 3 or 9.

Step 8.
Technical Analysis of Competitor Products - To better understand the competition, engineering then conducts a comparison of competitor technical descriptors. This process involves reverse engineering competitor products to determine specific values for competitor technical descriptors.

Step 9.
Technical Values for Technical Descriptors - At this stage in the process, the QFD team begins to establish target values for each technical descriptor. Target values represent "how much" for the technical descriptors.

Step 10.
Correlation Matrix - This room in the matrix is where the term House of Quality comes from because it makes the matrix look like a house with a roof. The correlation matrix is probably the least used room in the House of Quality; however, this room is a big help to the design engineers in the next phase of a comprehensive QFD project. Team members must examine how each of the technical descriptors impact each other. The team should document strong negative relationships between technical descriptors and work to eliminate physical contradictions.

Step 11.
Absolute Importance - Finally, the team calculates the absolute importance for each technical descriptor. This numerical calculation is the product of the cell value and the customer importance rating. Numbers are then added up in their respective columns to determine the importance for each technical descriptor. Now you know which technical aspects of your product matters the most to your customer!

Interrelationship between Technical Descriptors (correlation matrix) HOWs vs. HOWs +9 +3 -3 -9 Strong Positive Positive Negative Strong Negative

Technical Descriptors (HOWs)

Relationship between Customer Requirements and Technical Descriptors WHATs vs. HOWs +9 +3 +1 Strong Medium Weak

Primary Secondary

Customer Requirements (WHATs)


Our Product

Bs Product Importance to Customer

As Product

As Product

Bs Product Degree of Technical Difficulty Target Value Absolute Weight and Percent Relative Weight and Percent

Prioritized Technical Descriptors

How QFD helps in making process Agile

Imagine for a moment that you are the president of a successful software development company. Your company is doing reasonably well from a sales perspective, but you have been dealing with some sizable challenges in terms of your development team hitting their scheduled release dates on time. (The past 2 releases have been late by more than six months a piece.) Then one day your development manager comes into your office droning on about the success of something called Agile development methodologies. He goes on to tell you that he knows how to eliminate the slippages that he and his team have experienced in relation to your two year development plan: simply do away with the two year development plan. Needless to say, the conversation would probably not go well. However, there is a sweetener that can assist executive management in swallowing the sometimes bitter pill of Agile developmentand that sweetener bears the name QFD.

Scale-up Factor

Sales Point Absolute Weight and Percent

Technical Competitive Assessment

Our Product

Customer Competitive Assessment

Target Value

Prioritized Customer Requirements

Swallowing the Bitter Pill Despite the documented performance improvements that companies have experienced by adopting Agile methodologies, many corporate executives and stake holders have been slow to approve such a change due to their consternation about certain key principles in Agile development. One of the most challenging precepts for executives is the fact that under Agile methodologies (such as Scrum, XP, or Feature Driven Development) there are no two or five year development plans. Indeed, Agile methodologies gained their name from the idea of being able to quickly respond to ever-changing customer requirements. This means that Agile teams only plan out their development a few weeks or months in advance so that they can change course easily when needed. This principle of iterative development helps Agile teams to respond to the voice of the customer before that voice starts singing a different tune. The proponents of these methodologies have had tremendous success in remaining on the leading edge of competitive functionality. However, many corporate executives, product managers, and other stake holders frequently fear that development will proceed without direction under Agile development and that critical features and functionality will be omitted due to lack of proper planning. The issue of prioritizing the pool of features to be developed under an Agile methodology is indeed a vexing one. In truth, many Agile teams do wander aimlessly due to the sudden disappearance of a long range plan. These teams typically do a fantastic job of delivering working software every two weeks that has numerous new features. The new features are typically coded well and are very stable and maintainable. Unfortunately, these teams also find that in their rush to begin solving their customers problems correctly, they have ceased to solve the correct problems. The results are stable applications with long laundry lists of features, but that arent competitive in the market because they lack theright features.

Benefits of QFD
Quality function deployment was initially implemented to bring down start-up costs. Organizations using QFD have shown in their reports a reduced product development time. For example, U.S. car manufacturers of the late 1980s to early 1990s need an average of 5 years to place a product in the market, from drawing board to showroom, whereas Honda can place a new product in the market in 2 And 1/2 years and Toyota does it in 3 years. Both organizations credit this reduced time to the use of QFD. Product quality and, consequently, customer satisfaction improves with QFD due to various factors depicted in Figure.

Customer Driven
Quality function deployment looks beyond the usual customer response and tries to define the requirements in a set of basic needs, which are compared to all competitive information. All competitors are judged equally from customer and technical perspectives. This information can then be assessed using a Pareto diagram. Management can then place resources where they will

be the most beneficial in improving quality. Also, QFD uses the experience and information that are available within an organization and brings them together as a structured format that is easy to assimilate. This is important when an employee leaves a particular job and a new employee is hired for that job.

Reduces Implementation Time

Fewer engineering changes are required when using QFD, and, when used properly, all conflicting design needs can be identified and addressed prior to production. This results in reduction in retooling, operator training, and changes in traditional quality control measures. By using QFD, critical parts are identified and can be checked from product inception to production. Toyota reports that the quality of their product has improved by 1/3rd since the implementation of QFD.


Creates focus on customer requirements Uses competitive information effectively Prioritizes resources Identifies items that can be acted upon Structures resident experience/information Decreases midstream design change Limits post introduction problems Avoids future development redundancies Identifies future application opportunities Surfaces missing assumptions



Based on concensus Creates communication at interfaces Identifies actions at interfaces Creates global view out of details


Documents rationale for design Is easy to assimilate Adds structure to the information Adapts to changes (a living document) Provides framework for sensitivity analysis

Promotes Teamwork
Quality function deployment pushes a horizontal deployment of communication channels. Inputs are taken from all facets of an organization from marketing to production to sales, thus ensuring

that the voice of the customer is being met and that each department knows what the other is doing. This activity mitigates misinterpretation, opinions, and miscues. So, the left hand always knows what the right hand is doing. Efficiency and productivity always becomes better with enhanced teamwork.

Provides Documentation
A data base for future design or process improvements is created. Data that was historically scattered within operations, often lost or referenced out of context, are now saved in an orderly manner to serve future needs. This data base also serves as a training tool for new engineers. QFD is also very flexible when new information surfaces or things have to be changed on the QFD matrix.

The House of Quality works as a living document and acts as a source of ready reference for related products and future upgrades. While HOQ is a great communication tool at each step in the process, the matrices are the means and not the end. Its purpose is to serve as a mean for dialogue to strengthen vertical and horizontal communications. Through customer needs and competitive analysis, the House of Quality helps to find out the critical technical components that require change. Issues are solved that may never have surfaced before. The critical issues are then driven through the other matrices to identify the critical processes, manufacturing operations, and quality control measures needed to produce a product that fulfills both customer and producer needs within a shorter development cycle time. The total effect of all of this is that the parts that drive the companys actions are driven by the customers requirements. There is an better focus on the customer and a better awareness of their wants. Because of all these , the process leads to improved customer understanding and the ultimate outcome a satisfied customer.