8.2 Quality Function Deployment (QFD) & House of Quality

Key Takeaways

  • Quality Function Deployment (QFD) was developed by Dr. Yoji Akao and Dr. Shigeru Mizuno at Mitsubishi Kobe Shipyard in 1972 to translate VOC into engineering specs.
  • The House of Quality (HoQ) contains 6 rooms: Customer Requirements (WHATs), Engineering Characteristics (HOWs), Relationship Matrix, Technical Matrix, Correlation Roof Matrix, and Competitive Assessment.
  • Standard numerical weights in the relationship matrix use non-linear values: 9 for Strong, 3 for Medium, 1 for Weak, and 0 for No relationship.
  • The Correlation Roof Matrix maps interrelationships among engineering characteristics to identify positive synergies and resolve negative engineering trade-offs.
  • QFD cascades customer requirements across four waterfall matrices: Product Planning, Part Deployment, Process Planning, and Production Planning.
Last updated: July 2026

8.2 Quality Function Deployment (QFD) & House of Quality

Exam Tip: Expect ASQ CMQ/OE exam questions regarding the specific numerical scale used in the HoQ Relationship Matrix (9-3-1-0), the primary purpose of the Correlation Roof Matrix (identifying engineering trade-offs and synergies), and the sequence of the four QFD phases from customer input to shop-floor quality control.

Foundations of Quality Function Deployment (QFD)

Quality Function Deployment (QFD) is a comprehensive planning tool used to translate customer requirements—known as the Voice of the Customer (VOC)—into specific, quantifiable technical requirements, engineering specifications, and manufacturing control parameters. Rather than relying on assumptions made by isolated engineering departments, QFD ensures that the customer's priorities guide every phase of product design, engineering, and manufacturing.

QFD was created in Japan in 1972 by Dr. Yoji Akao and Dr. Shigeru Mizuno at Mitsubishi's Kobe Shipyard. They sought a system that would embed quality assurance directly into the design process before initial production. Toyota subsequently refined QFD, reporting a 60% reduction in pre-launch development costs and a 33% reduction in product development cycle time between 1977 and 1984.


Translating the Voice of the Customer (VOC) into Technical Requirements

The fundamental challenge in product development is that customers express needs in qualitative, imprecise language (e.g., "I want a car door that shuts with a solid sound" or "I need a laptop battery that lasts all day"). Engineers cannot design directly to vague adjectives. QFD provides a framework for mapping qualitative WHATs into measurable technical HOWs.

The Kano Model Integration

Before building the QFD matrix, customer requirements are often categorized using Noriaki Kano's model of customer satisfaction:

  1. Must-Be / Basic Requirements: Expected features that customers take for granted (e.g., leak-free automotive fuel tank). Their absence causes severe dissatisfaction, but their presence does not increase satisfaction.
  2. One-Dimensional / Performance Requirements: Spoken customer requests where satisfaction is directly proportional to performance (e.g., fuel efficiency, CPU processing speed).
  3. Attractive / Delighter Requirements: Unexpected innovative features that delight the customer (e.g., built-in wireless charging pad in a vehicle console).

Once collected via surveys, focus groups, and warranty analysis, customer requirements are prioritized using an Importance Rating (Ii), typically on a 1 to 5 scale (where 5 represents critical importance to the customer).


The Six Rooms of the House of Quality

The core analytical engine of QFD is the House of Quality (HoQ), a matrix structure named for its house-like shape formed by a triangular roof matrix sitting atop a central grid.

                    +-------------------+
                    |    ROOM 5: ROOF   |
                    | Correlation Matrix|
                    +-------------------+
                    |  ROOM 2: TECHNICAL|
                    |  CHARACTERISTICS  |
+-------------------+-------------------+-------------------+
|  ROOM 1: CUSTOMER |  ROOM 3: CENTRAL  | ROOM 6: CUSTOMER  |
|   REQUIREMENTS    | RELATIONSHIP GRID |    COMPETITIVE    |
|      (WHATs)      |   (WHATs vs HOWs) |    EVALUATION     |
+-------------------+-------------------+-------------------+
                    |  ROOM 4: TECHNICAL|
                    |   PRIORITIES &    |
                    |   TARGET VALUES   |
                    +-------------------+

Room 1: Customer Requirements (WHATs)

Located on the left horizontal axis, Room 1 lists customer demands organized hierarchically into primary, secondary, and tertiary needs, alongside their corresponding Customer Importance Ratings (Ii).

Room 2: Engineering Characteristics (HOWs)

Located on the top vertical columns, Room 2 lists the quantifiable technical parameters established by design engineers to satisfy customer needs. Each technical characteristic includes a direction of movement indicator:

  • Maximize: Higher numeric value is desirable (e.g., tensile strength, thermal resistance).
  • Minimize: Lower numeric value is desirable (e.g., mass, noise decibels, electrical resistance).
  • Target: Specific nominal target value is required (e.g., length = 150 mm +/- 0.1 mm).

Room 3: Relationship Matrix (Intersection Grid)

Room 3 occupies the main interior grid, evaluating the impact of each technical characteristic (HOW) on each customer requirement (WHAT). Standard QFD practice uses non-linear weighting values to accentuate strong interactions:

  • Strong Relationship (Solid Circle): Assigned a numerical value of 9.
  • Medium Relationship (Open Circle): Assigned a numerical value of 3.
  • Weak Relationship (Triangle): Assigned a numerical value of 1.
  • No Relationship (Blank): Assigned a numerical value of 0.

Room 4: Technical Matrix (Priorities, Benchmarks, and Targets)

Located at the bottom of the matrix, Room 4 computes the Absolute Importance Weight (Wj) for each technical requirement j by summing the product of customer importance ratings and relationship values across all m customer requirements:

Wj=i=1m(Ii×Rij)W_j = \sum_{i=1}^{m} (I_i \times R_{ij})

Where:

  • Ii is the Customer Importance Rating for requirement i.
  • Rij is the relationship weight (9, 3, 1, or 0) between customer requirement i and technical characteristic j.

Relative importance percentages are then calculated to identify the top technical drivers that engineers must focus on during detailed design.

Room 5: Correlation Roof Matrix

Sitting above Room 2, the triangular roof matrix evaluates how engineering characteristics interact with one another. Symbols represent degrees of correlation:

  • Strong Positive (++): Improving HOW 1 automatically improves HOW 2.
  • Positive (+): Improving HOW 1 slightly supports HOW 2.
  • Negative (-): Improving HOW 1 degrades HOW 2.
  • Strong Negative (--): Improving HOW 1 severely compromises HOW 2.

Room 6: Customer Competitive Assessment

Located on the far right, Room 6 compares the organization's current product against key competitors on a 1 to 5 scale across each customer requirement (WHAT). This pinpoints competitive gaps and marketing advantages.


The Correlation Roof Matrix: Resolving Engineering Trade-Offs

The roof matrix is arguably the most valuable strategic component of QFD. It exposes engineering conflicts early in design before physical prototypes are fabricated.

Example of an Engineering Trade-Off

Consider designing an electric vehicle battery enclosure:

  • HOW 1: Increase casing wall thickness (Direction: Maximize crash safety).
  • HOW 2: Reduce total vehicle curb weight (Direction: Minimize energy consumption).

The intersection of HOW 1 and HOW 2 in the roof matrix reveals a Strong Negative (--) correlation. Increasing wall thickness improves crash safety but increases mass, directly conflicting with vehicle weight reduction targets. Recognizing this conflict through QFD prompts engineers to investigate advanced lightweight composite materials or structural rib reinforcements rather than discovering the weight penalty during vehicle testing.


Cascading QFD: The Four-Phase Model

QFD is not a single matrix; it is a sequential four-phase cascade that translates customer desires all the way down to shop-floor quality control procedures.

+-------------------------------------------------------------------------+
| Phase 1: Product Planning (House of Quality)                           |
| Customer Requirements (WHATs) ---> Technical Characteristics (HOWs)    |
+------------------------------------+------------------------------------+
                                     |
                                     v
+-------------------------------------------------------------------------+
| Phase 2: Parts Deployment                                               |
| Technical Characteristics (WHATs) ---> Part & Subsystem Specs (HOWs)   |
+------------------------------------+------------------------------------+
                                     |
                                     v
+-------------------------------------------------------------------------+
| Phase 3: Process Planning                                               |
| Part Specs (WHATs) ---> Manufacturing Process Operations (HOWs)       |
+------------------------------------+------------------------------------+
                                     |
                                     v
+-------------------------------------------------------------------------+
| Phase 4: Production Planning                                            |
| Process Operations (WHATs) ---> Production Controls & SOPs (HOWs)      |
+-------------------------------------------------------------------------+
  1. Phase 1: Product Planning (House of Quality): Translates VOC into critical engineering technical characteristics.
  2. Phase 2: Parts Deployment: The prioritized technical characteristics from Phase 1 become the inputs (WHATs) to define component and part specifications (HOWs).
  3. Phase 3: Process Planning: Part specifications become the inputs (WHATs) to determine manufacturing process steps and equipment parameters (HOWs).
  4. Phase 4: Production Planning: Process operations become the inputs (WHATs) to formulate operator work instructions, statistical process control (SPC) charts, and inspection control plans (HOWs).
Loading diagram...
The Six Rooms of the House of Quality (HoQ)
Test Your Knowledge

What is the primary purpose of the triangular roof matrix at the top of the House of Quality (HoQ)?

A
B
C
D
Test Your Knowledge

In the traditional House of Quality relationship matrix, what numerical values are standardly assigned to Strong, Medium, and Weak relationships between customer requirements and engineering characteristics?

A
B
C
D
Test Your Knowledge

A cross-functional engineering team completes the initial House of Quality matrix (Phase 1: Product Planning). In Phase 2 (Parts Deployment), what becomes the input column (WHATs) for the new matrix?

A
B
C
D