7.1 Qualitative Root Cause Tools: 5 Whys & Cause-and-Effect (Fishbone) Diagrams

7.1 Qualitative Root Cause Tools: 5 Whys & Cause-and-Effect (Fishbone) Diagrams

Core Principle: In the Analyze phase of DMAIC, the continuous improvement team pivots from observing, measuring, and baselining process output performance ($Y$) to diagnosing, isolating, and validating the critical input variables and process parameters ($X$'s). Six Sigma expresses this governing relationship through the transfer function $Y = f(X_1, X_2, \dots, X_k)$. Qualitative root-cause analysis leverages structured brainstorming, categorical mapping, and iterative inquiry to cut through superficial symptoms. The 5 Whys method interrogates causal symptom chains to uncover systemic policy or procedural failures while strictly prohibiting practitioners from stopping at "human error." Cause-and-Effect (Ishikawa or Fishbone) diagrams systematically categorize potential causes across industry-tailored frameworks—the 6Ms in manufacturing, the 4Ss in healthcare and service, and the 8Ps in transactional settings. Fusing fishbone mapping with 5 Whys interrogation provides both horizontal breadth and vertical diagnostic depth.

---

The Strategic Purpose of the Analyze Phase: Shifting from $Y$ to $X$

Throughout the Define and Measure phases, the project team maintained an external, customer-facing orientation. Define scoped the business case, chartered the team, and identified the primary Critical to Quality (CTQ) characteristic. Measure validated the measurement system (via Gage R&R and Attribute Agreement Analysis) and established baseline capability indices ($C_p, C_{pk}, P_p, P_{pk}$) and defect rates (DPMO, First Time Yield).

However, a foundational axiom of Lean Six Sigma quality engineering dictates that you cannot directly control or adjust $Y$; you can only control the inputs, operating setpoints, raw materials, and process variables ($X$'s) that determine $Y$:

$Y = f(X_1, X_2, X_3, \dots, X_k)$

Where:

• $Y$ (The Effect): The dependent process output, symptom, customer requirement, or CTQ characteristic.
• $X_1, \dots, X_k$ (The Causes): The independent input variables, operational parameters, environmental factors, and procedural methods.

The primary mission of the Analyze phase is to identify all candidate $X$'s, filter out non-influential variables, and empirically verify the "vital few" critical root causes that drive unwanted variation in $Y$.

                     The DMAIC Analytical Funnel

     DEFINE & MEASURE                      ANALYZE PHASE
   ┌──────────────────┐               ┌───────────────────────┐
   │ Baseline Output  │               │ 30 - 50 Potential     │  Qualitative Tools
   │   Performance    │               │ Causes Identified     │  (Brainstorming, Affinity,
   │      ( Y )       │               └──────────┬────────────┘   Fishbone, 5 Whys)
   └────────┬─────────┘                          │
            │                                    ▼
            │                         ┌───────────────────────┐
            │                         │ 8 - 12 Likely Causes  │  Graphical & Exploratory
            │                         │ Prioritized           │  (Pareto, Histograms,
            │                         └──────────┬────────────┘   Multi-Vari, Boxplots)
            │                                    │
            ▼                                    ▼
   ┌──────────────────┐               ┌───────────────────────┐
   │ Problem Scoped   │               │ 2 - 4 Verified Root   │  Quantitative Validation
   │   & Quantified   │               │ Causes (Critical X's) │  (Hypothesis Testing,
   └──────────────────┘               └───────────────────────┘   ANOVA, Regression)

Root Cause vs. Superficial Symptom

A root cause is the most fundamental, underlying operational or systemic condition that creates a non-conformance or defect. If eliminated, the defect cannot recur. In contrast, a symptom is merely the observable manifestation or downstream consequence of that root cause.

Treating symptoms provides only temporary relief while guaranteeing that defects will return. For example, frequently replacing a worn cutting tool, mopping up an oil leak, or retraining an operator who made a data-entry error treats only symptoms. A genuine root-cause countermeasure investigates why the tool wore prematurely (incorrect coolant flow), why the seal leaked (unapproved gasket material), or why the software permitted invalid keystrokes (lack of input validation mistake-proofing).

---

The 5 Whys Technique: Interrogating Causal Chains

Originally conceived by Sakichi Toyoda (the founder of Toyota Industries) and refined by Taiichi Ohno within the Toyota Production System (TPS), the 5 Whys is an iterative interrogative technique used to explore the cause-and-effect relationships underlying an observed failure. By asking "Why?" five successive times, the investigator peels away superficial symptom layers to reach the core breakdown.

                       The 5 Whys Diagnostic Chain

  PROBLEM     ▶ The automatic palletizer jammed during final packing.
    │
  Why? (1)    ▶ A shear pin on the primary transfer arm snapped.
    │
  Why? (2)    ▶ The arm encountered severe mechanical resistance from dried adhesive.
    │
  Why? (3)    ▶ The hot-melt glue applicator nozzle was leaking onto the drive track.
    │
  Why? (4)    ▶ The nozzle seal degraded due to operating above its thermal rating.
    │
  Why? (5)    ▶ Purchasing substituted an unrated silicone seal without engineering review.
    │
  ROOT CAUSE  ▶ Procurement lacks a formal Management of Change (MOC) sign-off policy.

The Four Cardinal Rules of the 5 Whys

1. Never Stop at Human Error or "Retraining": On CSSC Green Belt examinations and in professional practice, stopping at "operator inattention," "clerical negligence," or "lack of employee training" is classified as an analytical failure. Humans operate within systems designed by leadership. When an operator makes a mistake, the practitioner must ask: Why did the system permit the error to occur? Why was there no mistake-proofing (Poka-Yoke) mechanism? Why were standard operating instructions ambiguous or inaccessible?
2. Ground Every Step in Empirical Evidence: Speculation, intuition, and consensus voting have no place in a 5 Whys chain. Each answer to "Why?" must be confirmed through direct physical inspection, maintenance logs, sensor telemetry, or observational data.
3. Validate with the Reverse "Therefore" Test: To verify that the causal logic is robust, read the chain backward from root cause to initial problem using the connector word "therefore". If the logic fails to hold in reverse, a logical leap or missing causal link exists:
   • Reverse Syllogism: "Procurement lacked a Management of Change policy; therefore, an unrated seal was purchased; therefore, the seal degraded under heat; therefore, adhesive leaked onto the track; therefore, the transfer arm jammed; therefore, the shear pin snapped."
4. Branch Causal Pathways for Multi-Factor Failures: In complex technical and administrative processes, a failure mode rarely possesses a single linear cause. Practitioners should allow the 5 Whys chain to branch into multiple parallel trees when a step exhibits multiple necessary conditions.

---

Cause-and-Effect (Ishikawa / Fishbone) Diagrams

Developed in 1943 by Dr. Kaoru Ishikawa at the University of Tokyo to assist engineers at Kawasaki Steel, the Cause-and-Effect Diagram (commonly called an Ishikawa diagram or fishbone diagram due to its skeletal appearance) is a graphical structuring tool that organizes brainstormed potential causes of a problem into logical categories.

                        Classic 6M Fishbone Architecture

      METHODS               MACHINES               MATERIALS
         │                     │                      │
   SOP Outdated          Spindle Runout         Impurity in Resin
         │     Lack of         │     Bearing          │     Supplier
         └───▶ Tooling         └───▶ Vibration        └───▶ Lot Variance
                 │                     │                      │
  ──────────────────────────────────────────────────────────────────▶ [ DEFECT /
                 │                     │                      │         EFFECT ]
         ┌───▶ Calibration     ┌───▶ Ambient          ┌───▶ Shift Fatigue
         │     Drift           │     Humidity         │     Turnover
         │                     │                      │
    MEASUREMENT          MOTHER NATURE            MANPOWER
                        (ENVIRONMENT)            (PEOPLE)

Anatomical Structure of the Fishbone

• The Head (The Effect): Located in a box at the far right. It must state the precise, quantified problem or undesirable effect ($Y$) established during Define and Measure (e.g., "Coating Thickness Exceeds 45 Microns on Line 2"). It must never state a presumed cause.
• The Spine: A heavy horizontal backbone pointing directly to the effect box.
• The Major Bones (Primary Categories): Large diagonal ribs branching off the spine. These represent the standardized organizational categories used to structure the investigation.
• The Minor Bones (Secondary & Tertiary Causes): Horizontal and diagonal branches extending from the primary ribs, representing specific candidate input variables ($X$'s) identified during brainstorming.

Categorization Frameworks Across Industries

To ensure exhaustive brainstorming and prevent blind spots, Six Sigma utilizes standardized category frameworks tailored to specific operational environments:

---

Affinity Diagrams & Structured Brainstorming

Before populating a fishbone diagram, continuous improvement teams frequently generate dozens of fragmented, unstructured ideas. Attempting to force raw thoughts immediately onto fishbone ribs leads to premature debates over categorization and suppresses unconventional insights.

The Affinity Diagram (also called the KJ Method, developed by Japanese anthropologist Jiro Kawakita) resolves this by organizing high volumes of subjective ideas into natural, cognitive clusters based on underlying relationships.

                       Affinity Diagramming Workflow

   STEP 1: SILENT DIVERGENCE             STEP 2: CONVERGENT CLUSTERING
  ┌─────────────────────────────┐       ┌─────────────────────────────┐
  │ Team members write one      │       │ Cards sorted silently into  │
  │ distinct potential cause per│ ────▶ │ organic thematic clusters   │
  │ sticky card without debate. │       │ without pre-set categories. │
  └─────────────────────────────┘       └──────────────┬──────────────┘
                                                       │
                                                       ▼
   STEP 4: FISHBONE POPULATION           STEP 3: HEADER FORMULATION
  ┌─────────────────────────────┐       ┌─────────────────────────────┐
  │ Verified clusters transfer  │       │ Create consensus header     │
  │ directly to the ribs of an  │ ◀──── │ cards defining the core     │
  │ Ishikawa fishbone diagram.  │       │ failure mechanism of each.  │
  └─────────────────────────────┘       └─────────────────────────────┘

Operational Rules for Effective Brainstorming

• Defer Judgment: Criticism, ridicule, or immediate dismissal of ideas is strictly prohibited during ideation. Premature evaluation shuts down creative exploration.
• Encourage Quantity: The objective is sheer volume. From 50 or 60 raw suggestions, the 2 or 3 genuine breakthrough causes will emerge.
• Piggyback on Ideas: Participants should build upon, modify, or recombine concepts proposed by colleagues.
• Equalize Participation: Silent brainstorming via sticky notes or digital collaboration boards ensures that introverted frontline operators possess the same analytical voice as assertive managers or technical specialists.

---

The Diagnostic Fusion: Combining Fishbone with 5 Whys

A critical failure mode in Six Sigma deployment is executing quality tools as disjointed check-the-box exercises. High-performing Green Belts unite the fishbone diagram and the 5 Whys into an integrated root-cause diagnostic engine:

1. Horizontal Breadth via Fishbone: The team builds an Ishikawa diagram using the 6Ms, 4Ss, or 8Ps to map the complete landscape of suspected causes, ensuring no operational category is overlooked.
2. Prioritization via Multi-Voting: Using multi-voting (e.g., dot voting where each team member receives a fixed number of votes) or preliminary exploratory data, the team narrows the candidate causes down to 3 to 5 high-probability branches.
3. Vertical Depth via 5 Whys: The team takes each prioritized branch and conducts a rigorous 5 Whys interrogation directly on the fishbone rib, drawing sub-bones outward until the systemic root cause is isolated.

                 Combining Fishbone Breadth with 5 Whys Depth

     [ PRIMARY RIB: MACHINE ]
         │
         ├──▶ Why? Spindle overheating
         │       │
         │       └──▶ Why? Lubrication failure
         │               │
         │               └──▶ Why? Oil pump seized
         │                       │
         │                       └──▶ Why? Filter clogged with swarf
         │                               │
         │                               └──▶ ROOT CAUSE: Preventative
         │                                    maintenance interval was
         │                                    extended without engineering sign-off

---

Real-World Case Study: Aircraft Hydraulic Actuator Failure

An aerospace maintenance and overhaul facility experienced unexpected pressure loss in newly overhauled commercial aircraft hydraulic actuators during pre-flight functional tests ($Y = \text{Actuator Internal Pressure Drop} > 15\text{ psi/min}$).

Step 1: Horizontal Brainstorming (6Ms Fishbone)

The cross-functional team constructed an Ishikawa diagram. Under the Materials rib, technicians noted that O-ring seals arrived from a secondary distributor. Under Methods, technicians noted that seal installation lubrication procedures were revised last month. Under Measurement, pressure gages were calibrated and certified.

Step 2: Prioritizing the Critical Branch

Teardown inspections revealed that internal nitrile O-rings exhibited premature micro-cracking and hardening after exposure to Skydrol hydraulic fluid. The team selected the "O-Ring Material Degradation" branch for vertical interrogation.

Step 3: Vertical 5 Whys Interrogation

1. Why did the actuator lose hydraulic pressure? The secondary piston O-ring experienced circumferential micro-cracking, allowing fluid bypass.
2. Why did the O-ring crack? The elastomer hardened rapidly when exposed to high-pressure synthetic phosphate-ester hydraulic fluid (Skydrol).
3. Why did the elastomer harden? The installed O-ring was composed of standard Buna-N nitrile instead of the specified fluorocarbon (Viton / FKM) elastomer.
4. Why was a Buna-N nitrile O-ring installed? The parts bin contained a mixed lot of seals with identical physical dimensions and color codes.
5. Why were unapproved seals present in the assembly bin? The procurement department sourced an unvetted commercial surplus lot without requiring material test reports (MTRs) to meet a quarter-end cost reduction target.

Systemic Root Cause: Procurement lacks an automated supplier quality gate requiring mandatory Material Test Report validation prior to inventory stocking for flight-critical components.

Step 4: Verification via the Reverse "Therefore" Test

"Procurement lacked an automated MTR gate; therefore, unvetted commercial surplus seals were stocked; therefore, Buna-N seals were mixed into flight-critical bins; therefore, technicians installed non-compatible elastomers; therefore, the O-ring hardened upon exposure to Skydrol; therefore, the actuator lost pressure."

The reverse logic holds perfectly, verifying that the team identified an actionable, systemic root cause rather than blaming technician assembly error.

---

Critical Exam Traps to Avoid

• Trap 1: Stopping at "Operator Error" or "Retraining Required" — On CSSC Green Belt exams, any option attributing root cause to human carelessness or prescribing employee retraining as the primary countermeasure is almost universally an incorrect distractor. Look for systemic, procedural, or mistake-proofing (Poka-Yoke) root causes.
• Trap 2: Confusing Cause with Effect in the Fishbone Head — Ensure the problem statement at the fish's head represents an observed undesirable effect ($Y$), not a presumed cause ($X$). If the head states "Worn Cutting Tools," the team has plotted an input variable rather than the output defect (e.g., "Hole Diameter Out of Specification").
• Trap 3: Treating Brainstormed Causes as Proven Facts — Placing an item on an Ishikawa diagram does not prove it causes the defect; it merely records a hypothesis. In the Analyze phase, candidate $X$'s must be verified through empirical data, stratification, scatter plots, or hypothesis testing before moving to Improve.
• Trap 4: Forcing Non-Manufacturing Workflows into the 6Ms — Attempting to force an insurance claims process or hospital discharge workflow into "Machines" and "Mother Nature" causes confusion. Use the 4Ss for service/healthcare and the 8Ps for transactional processes.