8.1 Affinity, Tree, Matrix & Process Decision Program Charts

8.1 Affinity, Tree, Matrix & Process Decision Program Charts

While the Seven Basic Quality Tools (Flowcharts, Check Sheets, Histograms, Pareto Charts, Cause-and-Effect Diagrams, Scatter Plots, and Control Charts) excel at collecting, analyzing, and monitoring quantitative, numerical process data, quality professionals frequently confront unstructured, qualitative, and complex organizational challenges. In 1976, a committee of the Japanese Union of Scientists and Engineers (JUSE), led by Shigeru Mizuno and colleagues, codified the Seven Management and Planning (7 MP) Tools (also known as the New 7 Quality Tools).

These tools enable cross-functional teams to organize verbal data, clarify ambiguous relationships, establish strategic priorities, anticipate implementation risks, and execute multi-phase action plans. On the ASQ Certified Quality Improvement Associate (CQIA) examination, candidates must understand the purpose, construction methodology, and practical application of each management tool.

+-------------------------------------------------------------------------+
|          THE 7 BASIC QUALITY TOOLS vs. THE 7 MANAGEMENT & PLANNING TOOLS |
+-------------------------------------------------------------------------+
|  7 BASIC QUALITY TOOLS (Quantitative)  |  7 MANAGEMENT TOOLS (Qualitative)      |
|  * Focus: Data, Variation & Control   |  * Focus: Strategy, Planning & Logic   |
|  * Check Sheets & Histograms          |  * Affinity Diagram (KJ Method)        |
|  * Pareto Charts & Scatter Diagrams   |  * Tree Diagram (Systematic Diagram)   |
|  * Cause-and-Effect Diagrams (Fishbone)|  * Matrix Diagrams (L, T, Y, X)        |
|  * Flowcharts & Run / Control Charts  |  * Process Decision Program Chart (PDPC)|
|  * Defect counts, cycle times, gages  |  * Interrelationship Digraph           |
|                                       |  * Arrow Diagram (CPM / PERT)          |
|                                       |  * Matrix Data Analysis Diagram (MDAD) |
+-------------------------------------------------------------------------+

---

1. The Affinity Diagram (The KJ Method)

Developed in the 1960s by Japanese anthropologist Jiro Kawakita (from whose initials the term KJ Method derives), the Affinity Diagram is an organizational brainstorming tool designed to synthesize large volumes of chaotic, unstructured verbal information (ideas, opinions, customer voice statements, operational challenges) into natural, meaningful thematic groupings based on inherent affinities.

+-------------------------------------------------------------------------+
|                    AFFINITY DIAGRAM (KJ METHOD) WORKFLOW                |
+-------------------------------------------------------------------------+
|                                                                         |
|  STEP 1: PHRASE ISSUE   ──► "What factors cause delivery delays?"       |
|                                                                         |
|  STEP 2: SILENT         ──► Write 1 idea per sticky note / card         |
|          BRAINSTORMING      (Generate 30 - 100 concrete ideas)          |
|                                                                         |
|  STEP 3: SILENT         ──► Team silently moves cards into natural      |
|          SORTING            affinity clusters (NO TALKING ALLOWED)      |
|                                                                         |
|  STEP 4: CREATE         ──► Group assigns a header card capturing the   |
|          HEADERS            underlying theme of each cluster            |
|                                                                         |
|  STEP 5: DRAW & ACT     ──► Connect related clusters with borders and   |
|                             arrows; feed into Tree or Matrix diagrams   |
|                                                                         |
+-------------------------------------------------------------------------+

Step-by-Step Construction Rules

1. Define the Problem Statement: State the focus question broadly and neutrally (e.g., "What barriers prevent on-time patient discharge from surgical recovery?").
2. Generate Unstructured Ideas (Cards/Sticky Notes): Participants write distinct ideas on individual sticky notes using concise noun-verb phrases (e.g., "Pharmacy delays pain medication delivery" rather than vague single words like "Pharmacy"). Aim for 30 to 80 cards.
3. Perform Silent Sorting: Team members physically group cards that share natural connections or underlying root themes. Crucial Rule: Sorting must be conducted in total silence. Silence prevents dominant team members from arguing their perspectives, neutralizes organizational hierarchy, encourages intuitive right-brain pattern recognition, and allows quiet contributors equal influence.
4. Create Consensus Header Cards: Once sorting stabilizes, the team breaks silence to discuss the groupings and author a descriptive Header Card for each cluster (e.g., "Pharmacy Fulfillment Bottlenecks," "Transportation Gaps," "Electronic Health Record Lag").
5. Identify Sub-Clusters (Super-Headers): If a cluster contains more than 8 to 10 cards, break it down into secondary sub-clusters under the parent header.

+-------------------------------------------------------------------------+
|                    AFFINITY DIAGRAM CLUSTER OUTPUT                      |
+-------------------------------------------------------------------------+
|                                                                         |
|  [ LOGISTICS & TRANSIT ]     [ DATA & RECORDS ]    [ STAFF SCHEDULING ] |
|  ┌─────────────────────┐    ┌──────────────────┐  ┌───────────────────┐ |
|  │ Wheelchair shortage │    │ EHR login latency│  │ Shift change gap  │ |
|  │ Elevator maintenance│    │ Missing physician│  │ Nurse ratio low   │ |
|  │ Escort staff delays │    │ signature on Rx  │  │ Discharge clerk   │ |
|  │ Hallway congestion  │    │ Billing clearance│  │ on break at peak  │ |
|  └─────────────────────┘    │ hold in system   │  └───────────────────┘ |
|                             └──────────────────┘                        |
+-------------------------------------------------------------------------+

When to Use an Affinity Diagram

• The team faces a broad, ill-defined, or complex problem with no obvious mathematical model.
• An extensive set of qualitative customer feedback (Voice of the Customer survey text, focus group comments) requires structuring.
• Cross-functional team consensus is needed on controversial or emotionally charged topics.
• As an input to subsequent tools (e.g., feeding Affinity clusters into a Cause-and-Effect diagram or Tree diagram).

Exam Tip: If an exam question describes a team sorting sticky notes in total silence to organize brainstormed qualitative data into natural categories, the correct answer is always the Affinity Diagram (KJ Method).

---

2. The Tree Diagram (Systematic Diagram)

A Tree Diagram (also called a Systematic Diagram or Hierarchy Diagram) is a graphical technique used to break down broad strategic goals, complex systems, or general concepts into progressively increasing levels of operational detail. It transforms abstract objectives into concrete, executable tasks using rigorous logical branching.

+-------------------------------------------------------------------------+
|                        STRUCTURE OF A TREE DIAGRAM                      |
+-------------------------------------------------------------------------+
|                                                                         |
|  [ Primary Goal ] ──┬──► [ Level 1 Strategy A ] ──┬──► [ Action Step A1 ]|
|                     │                             └──► [ Action Step A2 ]|
|                     │                                                   |
|                     └──► [ Level 1 Strategy B ] ──┬──► [ Action Step B1 ]|
|                                                   └──► [ Action Step B2 ]|
|                                                                         |
|  ◄── "WHY?" (Purpose)                               "HOW?" (Action) ──► |
+-------------------------------------------------------------------------+

Types and Logics of Tree Diagrams

1. "How-How" Tree (Action / Means-End Tree): Used for project planning and solution development. Moving from left to right answers the question "How will this be accomplished?" Moving from right to left answers "Why are we doing this step?"
2. "Why-Why" Tree (Root Cause Tree): Used for diagnosing problem causes. Moving from left to right repeatedly asks "Why did this failure occur?", functioning as a structured graphical branching of the 5 Whys.
3. Component Tree (Functional Breakdown / WBS): Used in systems engineering, value engineering, and project management (Work Breakdown Structure) to decompose a physical product or project into assemblies, sub-assemblies, and discrete parts.
4. Fault Tree Analysis (FTA): A specialized top-down deductive tree diagram that uses Boolean logic gates (AND gates, OR gates) to calculate system failure probabilities based on component-level breakdown rates.

+-------------------------------------------------------------------------+
|             EXAMPLE: "HOW-HOW" TREE DIAGRAM FOR REDUCING SCRAP          |
+-------------------------------------------------------------------------+
|                                                                         |
|                    ┌──► Upgrade Tooling ────┬──► Install carbide dies   |
|                    │                        └──► Add auto-lubrication   |
|                    │                                                    |
|  Reduce Machining  ┼──► Standardize Setup ──┬──► Create visual SOPs     |
|  Scrap by 40%      │                        └──► Implement torque limit |
|                    │                                                    |
|                    └──► Improve Operator ───┬──► Cross-train on setup   |
|                         Competence          └──► Weekly gage audits     |
|                                                                         |
+-------------------------------------------------------------------------+

Construction Guidelines

• Maintain consistent logical depth across all parallel branches.
• Verify that the sub-elements at each level are collectively exhaustive (completing all lower-level steps guarantees accomplishment of the parent branch).
• Stop branching when an actionable work package with a single owner, defined deliverable, and verifiable completion metric is achieved.

---

3. Matrix Diagrams

A Matrix Diagram is a systematic quality tool that displays the presence, direction, and relative strength of relationships between two, three, or four sets of factors arranged in intersecting rows and columns. Matrix diagrams provide a structured visual matrix for evaluating correlations, such as customer requirements versus technical design parameters in Quality Function Deployment (QFD).

+-------------------------------------------------------------------------+
|                     THE FOUR MAJOR MATRIX CONFIGURATIONS                |
+-------------------------------------------------------------------------+
|                                                                         |
|     [ L-MATRIX ]              [ T-MATRIX ]              [ Y-MATRIX ]    |
|      Factors A                 Factors B                 Factors B      |
|     ┌─────────┐               ┌─────────┐               ┌─────────┐     |
|   F │         │             F │         │             F │  (B x A)│     |
|   a │  (A x B)│             a │  (A x B)│             a ├─────────┤     |
|   c ├─────────┤             c ├─────────┼─────────┐   c │         │     |
|   t │Factors B│             t │Factors A│(A x C)  │   t │  (A x C)│     |
|   o └─────────┘             o └─────────┴─────────┘   o ├─────────┤     |
|   r                         r             Factors C   r │  (C x B)│     |
|   s                         s                         s └─────────┘     |
|                                                         Factors C       |
|                                                                         |
|                         [ X-MATRIX (Hoshin Kanri) ]                     |
|                                Factors B (Metrics)                      |
|                                    ┌─────────┐                          |
|                     Factors A      │ (A x B) │ Factors C                |
|                    (Strategies) ───┼─────────┼─── (Projects)            |
|                                    │ (D x C) │                          |
|                                    └─────────┘                          |
|                                 Factors D (Owners)                      |
+-------------------------------------------------------------------------+

Major Matrix Types and Their Configurations

1. L-Matrix (Two Dimensions):
   • Structure: A standard two-dimensional grid shaped like the letter "L" that compares two groups of items against each other ($A \times B$).
   • Common Application: Mapping customer requirements against product design specifications, or matching process steps to responsible departments (RACI matrix).
2. T-Matrix (Three Dimensions / Twin Relationships):
   • Structure: Formed by joining two L-matrices sharing a single common factor list ($A$), creating a "T" shape. Set $A$ is compared simultaneously to Set $B$ and Set $C$ ($A \times B$ and $A \times C$), while $B$ and $C$ are not directly compared to each other.
   • Common Application: Comparing employee job roles ($A$) against required technical skills ($B$) on one side, and against regulatory training modules ($C$) on the other.
3. Y-Matrix (Three Dimensions / Circular Closed Loop):
   • Structure: Connects three sets of factors in a circular 3D loop where each set relates to the other two ($A \times B$, $B \times C$, and $C \times A$), resembling an unfolded cube or a "Y".
   • Common Application: Relating customer requirements ($A$) to engineering specifications ($B$), engineering specifications to manufacturing process variables ($C$), and process variables back to customer requirements ($A$).
4. X-Matrix (Four Dimensions / Hoshin Kanri):
   • Structure: Connects four sets of factors arranged around a central cross ($A \times B$, $B \times C$, $C \times D$, and $D \times A$).
   • Common Application: Hoshin Kanri (Policy Deployment), linking 3-5 year corporate strategic breakthrough goals ($A$) to annual tactical objectives ($B$), cross-functional improvement projects ($C$), and resource/department ownership metrics ($D$).
5. C-Matrix (3D Roof / Cubic):
   • Structure: A three-dimensional cube comparing three factors simultaneously at intersection points ($A \times B \times C$). Rarely used due to visual complexity.

Standard Relationship Symbols and Numerical Weights

In quality matrices (especially the House of Quality in QFD), relationship strengths are marked with standardized symbols and weighted numerical scores:

| Symbol | Visual Notation | Relationship Strength | Typical Numerical Weight | | :--- | :---: | :--- | :---: | :---: | | Double Circle / Filled Circle | $\bullet$ or $\odot$ | Strong Relationship | 9 (or 5) | | Open Circle | $\circ$ | Moderate Relationship | 3 | | Triangle | $\triangle$ | Weak Relationship | 1 | | Blank Cell | (empty) | No Relationship | 0 |

+-------------------------------------------------------------------------+
|               EXAMPLE: L-MATRIX (CUSTOMER NEEDS vs. DESIGN)             |
+-------------------------------------------------------------------------+
|                               │ Low Weight │ High Battery │ Waterproof  |
|  CUSTOMER REQUIREMENTS        │ Spec (kg)  │ Life (Hours) │ Rating (IP) |
+───────────────────────────────┼────────────┼──────────────┼─────────────+
|  Easy to carry all day        │     ● (9)  │      ○ (3)   │      - (0)  |
|  Usable in heavy rain         │     - (0)  │      - (0)   │      ● (9)  |
|  Operates across 12-hr shift  │     △ (1)  │      ● (9)   │      - (0)  |
+-------------------------------------------------------------------------+

---

4. Process Decision Program Chart (PDPC)

A Process Decision Program Chart (PDPC) is a dynamic planning tool designed to systematically identify potential risks, deviations, failure modes, and roadblocks in an implementation plan, and to map out effective contingency countermeasures before the plan is rolled out.

+-------------------------------------------------------------------------+
|             PROCESS DECISION PROGRAM CHART (PDPC) STRUCTURE             |
+-------------------------------------------------------------------------+
|                                                                         |
|  [ GOAL ] ──► [ TASK ] ──► [ STEP ] ──► [ POTENTIAL FAILURE ]           |
|                                                  │                      |
|                                                  ├──► [ COUNTERMEASURE 1] (Selected [O])
|                                                  │                      |
|                                                  └──► [ COUNTERMEASURE 2] (Rejected [X])
|                                                                         |
+-------------------------------------------------------------------------+

Operational Workflow of PDPC

1. Level 1 (Goal): State the project or operational objective (e.g., "Implement automated optical inspection on Assembly Line 2").
2. Level 2 (Phases / Major Activities): Outline the chronological sequence of required phases (e.g., "Procure hardware," "Install software," "Train operators").
3. Level 3 (Operational Steps): Break each phase into discrete tasks.
4. Level 4 (Potential Failure Points / "What Ifs"): Brainstorm what could go wrong at each step (e.g., "Vendor software version incompatible with legacy PLC").
5. Level 5 (Contingency Countermeasures): Develop specific actions to prevent or recover from each identified failure.
6. Evaluation of Feasibility: Evaluate each countermeasure and designate it visually:
   • Selected Countermeasure ($[\text{O}]$ or Box): Feasible, cost-effective, and officially incorporated into the contingency protocol.
   • Rejected Countermeasure ($[\text{X}]$): Infeasible, cost-prohibitive, or technically impractical.

+-------------------------------------------------------------------------+
|                   PDPC EXAMPLE: HOSPITAL EHR CUTOVER                    |
+-------------------------------------------------------------------------+
|                                                                         |
|  Phase: Database Migration to Cloud                                     |
|  Step: Live Server Data Transfer at Midnight                            |
|    │                                                                    |
|    ├──► Potential Problem: Network Bandwidth Throttling / Data Drop     |
|    │      │                                                             |
|    │      ├──► Countermeasure A: Provision redundant dedicated fiber [O]|
|    │      └──► Countermeasure B: Defer cutover to next month [X]        |
|    │                                                                    |
|    └──► Potential Problem: Patient Record Format Corruption             |
|           │                                                             |
|           ├──► Countermeasure A: Automated checksum script rollback [O] |
|           └──► Countermeasure B: Manual chart reconciliation [X]        |
|                                                                         |
+-------------------------------------------------------------------------+

PDPC vs. FMEA Comparison

• PDPC: A graphical, tree-structured contingency planning tool used during project design to map "what-if" decision trees and establish active branching countermeasures.
• FMEA (Failure Mode and Effects Analysis): A tabular, quantitative risk assessment methodology that calculates a numerical Risk Priority Number ($\text{RPN} = \text{Severity} \times \text{Occurrence} \times \text{Detection}$) to rank failure modes for mitigation.

---

5. Other 7 Management & Planning Tools Overview

To round out the complete 7 MP toolkit tested on the CQIA examination, candidates should recognize the remaining three planning tools:

+-------------------------------------------------------------------------+
|               SUMMARY OF THE 7 MANAGEMENT & PLANNING TOOLS              |
+-------------------------------------------------------------------------+
|  Tool Name                | Primary Function and Output                 |
+---------------------------+---------------------------------------------+
|  1. Affinity Diagram      | Groups large volumes of chaotic qualitative |
|     (KJ Method)           | ideas into natural thematic clusters.       |
|  2. Interrelationship     | Identifies multi-directional cause-effect   |
|     Digraph (Relations)   | drivers (root causes) vs. key outcomes.     |
|  3. Tree Diagram          | Systematically breaks down goals or systems |
|     (Systematic Diagram)  | into granular "How-How" or "Why-Why" steps.  |
|  4. Matrix Diagram        | Displays relationships and strengths between|
|     (L, T, Y, X Shapes)   | two, three, or four sets of variables.      |
|  5. Matrix Data Analysis  | Uses multivariate statistics (e.g., PCA) to |
|     Diagram (MDAD)        | numerically plot and group matrix factors.  |
|  6. Process Decision      | Anticipates implementation risks and maps   |
|     Program Chart (PDPC)  | contingency countermeasures before launch.  |
|  7. Arrow Diagram         | Sequences interdependent tasks and defines  |
|     (Activity Network/CPM)| the Critical Path with float/slack times.   |
+-------------------------------------------------------------------------+

Interrelationship Digraph (Relations Diagram)

• Purpose: Analyzes complex, tangled cause-and-effect relationships where multiple variables interact non-linearly.
• Method: Brainstormed factors are arranged in a circle. Arrows are drawn from the cause variable to the effect variable ($A \to B$).
• Root Cause vs. Key Outcome:
  • Nodes with the highest Outgoing Arrows (Out-arrows $\gg$ In-arrows) are fundamental Root Causes / Drivers.
  • Nodes with the highest Incoming Arrows (In-arrows $\gg$ Out-arrows) are Key Symptoms / Key Indicator Outcomes.

Arrow Diagram (Activity Network / CPM / PERT)

• Purpose: Graphically plans the chronological sequence of interdependent project tasks, showing early/late start dates, early/late finish dates, float (slack) time, and the Critical Path (the longest sequence of dependent activities having zero slack, defining the shortest possible project completion duration).