8D problem solving training: eight disciplines for customer complaints

Starts with a pause: think of the last 8D report you wrote or reviewed, and whether its root cause was verified with data or simply agreed in a meeting. Three recent vehicle recalls follow, including experimental ball studs from an unapproved heat-treatment process and a supplier that could not locate 40 vehicles, each raising a question about containment, failure differences and the system. You then see the eight disciplines at a glance: contain the problem, find and verify its causes, fix them and prevent them returning. You meet XYZ Wiring in Munich, which makes heated seat wiring harnesses for a car maker's seat plant. The module outlines the twelve modules, practice, final exam and certificate, and ends with a recap.

What you will be able to do

• Explain why quick fixes let customer complaints come back
• Describe the purpose of each of the eight disciplines
• Recognise the three questions that recent recalls raise

Why it matters

Seeing why recalls repeat helps you close complaints only when the cause is gone, not when the paperwork is done.

Practice: Practice: Welcome, and why 8D matters · 3 questions

Opens with two problems reaching the quality office at XYZ Wiring in the same week: a box labelled with last week's date because a printer template was not updated, and a harder problem that deserves a full 8D. You then learn to choose a method by risk, scale, knowledge of the cause and customer requirements, with a quick fix and short record for local problems. The module explains IATF 16949 clause 10.2.3, a documented problem-solving process with containment, root cause analysis, systemic action, verification and updated documents, and clause 10.2.6 on complaints and field failures. An electronics plant in Singapore shows the other side. It covers using the customer's prescribed method, form and deadlines, keeping a problem log, and ends with a recap.

What you will be able to do

• Decide when a problem needs a full 8D instead of a quick fix
• Explain what IATF 16949 clause 10.2.3 requires of problem solving
• Apply customer-prescribed 8D methods, forms and deadlines

Why it matters

Choosing the method deliberately saves effort on simple problems and puts real investigation where customers expect it.

Practice: Practice: When to use 8D, and what customers expect · 3 questions

Starts with a call to XYZ Wiring: six seats have failed the heating test at the customer and its line is waiting. Two first days are compared, one waiting for returned parts while suspect harnesses keep shipping, one acting at once. You then learn D0 as ASQ describes it, emergency response actions taken before the cause is known and checked to see that they work. The module covers the authority to stop shipment and production in IATF 16949 clause 5.3.2, the ship hold on part SH-200, immediate customer notice, and lot traceability for product-safety parts under clause 4.4.1.2. A brake hose supplier in Toronto shows an emergency response at two in the morning. It ends with preparing the 8D plan and a recap.

What you will be able to do

• Apply emergency response actions before the cause is known
• Explain who must have authority to stop shipment and production
• Prepare an 8D with the complaint, returned parts and traceability

Why it matters

Protecting the customer in the first hours stops more suspect parts reaching its line while you look for the cause.

Practice: Practice: D0: preparation and emergency response · 3 questions

Opens with two ways XYZ Wiring could staff its 8D: a quality engineer working alone at her desk, unaware that a crimping tool on line three was overhauled in March, or a team that brings the shop floor into the room. You then learn D1: a team with product and process knowledge, existing or formed for the purpose, with a named leader and a champion, here the plant manager who hears a short update every Friday. The module explains D2 and 5W2H, with each answer taken from a record, a measurement or an observation, and returned parts kept intact for D4. An office chair maker in London shows the same rule. It ends with a problem statement that says what is happening, not why, and a recap.

What you will be able to do

• Form an 8D team with product, process and shop floor knowledge
• Describe a problem with 5W2H using records and measurements
• Write a problem statement that does not assume the cause

Why it matters

The right people and a problem described in numbers point the investigation in the right direction from the start.

Practice: Practice: D1 and D2: the team and 5W2H · 3 questions

Starts with two ways the XYZ team could look for the cause: checking everything for two weeks with no pattern, or asking where and when the problem happens and where it does not. You then learn is/is-not analysis across what, where, when, who and how much, and why the procedure calls distinctions key clues to the root cause. You see the is/is-not table built for XYZ Wiring with the data the team gathered, and the questions about what is unique and what has changed. A cosmetics bottling line in Paris shows the same logic for a rare defect. The module covers the sheet itself, with the source of each fact such as date codes, records or counts, and ends with testing every idea against both columns and a recap.

What you will be able to do

• Build an is/is-not table across what, where, when, who and how much
• Identify the distinctions and changes that point to a cause
• Test each possible cause against both columns of the table

Why it matters

Asking why a problem happens here but not there narrows the search fast, instead of testing everything at once.

Practice: Practice: D2: is/is-not analysis · 3 questions

Opens with two ways of containing the XYZ problem: sorting only the harnesses in the plant, or tracing every part made in the suspect period. You then learn that containment isolates the problem from any customer across the whole pipeline, including stock in transit, cross-docks and logistics hubs, and that IATF 16949 clause 8.7.1.3 treats suspect product as nonconforming. The module shows the suspect period matched to build records for line three with tool A-17, sorting with a method shown to catch the defect, and reconciling counts until every part is accounted for. It covers customer approval and records of quantity, disposition, date and traceability, then setting a clean point with a known start, a visible mark and the customer told. It ends with a recap.

What you will be able to do

• Apply containment across the plant, transit and the customer
• Reconcile sorting counts until every suspect part is found
• Set a clean point and tell the customer how to recognise it

Why it matters

Containment verified with numbers, not confidence, is what keeps every suspect part away from the customer.

Practice: Practice: D3: containment and the clean point · 3 questions

Starts with two root cause meetings: one where the most senior person blames the new wire supplier and everyone nods, and one that lists every possible cause. You then learn the cause-and-effect diagram, with the problem statement at the head and causes by category, filtered by the is/is-not sheet, and you see the fishbone the XYZ team built, with rejected ideas recorded and the reason. The module explains 5 Whys, following one chain down to a cause in the system, with each answer checked against facts and the tool's limits in mind, including a printer template example. It covers building two chains, why the defect was made and why it was not detected, and the records clause 10.2.3 c) asks for. It ends with a recap.

What you will be able to do

• Build a fishbone diagram and filter it with is/is-not facts
• Use 5 Whys to follow a cause down to a system gap
• Document the method, analysis and results of root cause analysis

Why it matters

Listing widely and digging deeply stops the team chasing the loudest opinion while the real cause carries on.

Practice: Practice: D4: cause-and-effect and 5 Whys · 3 questions

Opens with two ways the XYZ 8D could have gone: a meeting agrees on a worn tool and closes the 8D, and a year later another worn tool produces bad crimps that ship. You then learn to separate the occurrence root cause, why the defect was made, from the escape point, the earliest control that should have caught it. The module shows the evidence: the anvil measured in the toolroom beyond the maker's wear limit, a tool-life counter that every overhaul resets, and crimp monitor limits anyone could change with no password, approval or record. An aluminium extrusion plant in Sydney shows why verification matters. It covers physical analysis, correlation and turning the problem on and off, the records a verified D4 produces, and ends with a recap.

What you will be able to do

• Tell an occurrence root cause from an escape root cause
• Find the escape point where a control should have caught the defect
• Verify a root cause by turning the problem on and off

Why it matters

Verifying both why the defect was made and why it escaped stops the same failure shipping again next year.

Practice: Practice: D4: occurrence and escape root causes · 3 questions

Starts with two action plans for the same root cause: telling maintenance staff not to reset counters when they reuse an anvil, or changing the system so it does not depend on memory. You then learn to choose permanent corrective actions for each verified root cause, including the escape point, and to consider similar processes and products as IATF 16949 clause 10.2.3 d) asks. The module explains strong actions that design the cause out or prevent it with error-proofing, and checking the error-proofing itself with controlled challenge parts and a reaction plan. You see the XYZ action plan, with the monitor setting locked by password and a press interlock owned by the process engineer. It ends with trial plans, acceptance criteria set in advance and a recap.

What you will be able to do

• Choose corrective actions that act on the system, not on memory
• Apply error-proofing and check that it works with challenge parts
• Verify each action against acceptance criteria before going live

Why it matters

Strong, tested actions remove the cause for good instead of relying on people remembering a new instruction.

Practice: Practice: D5: choosing permanent corrective actions · 3 questions

Opens with two ways to finish D6: fitting the interlock on Monday and closing the 8D on Tuesday, or waiting for data from every shift. You then learn to implement each action fully: changed, documented, trained, controlled and confirmed on every shift, with the overhaul procedure and line three work instructions updated and issued. The module explains validating effectiveness with production data over enough time and volume, as IATF 16949 clause 10.2.3 e) expects, and using capability as a second kind of evidence against any customer minimum. A published report shows a supplier adding camera inspections and pull tests since August 2025. It covers the implementation record, removing containment against exit criteria set in advance with the customer's written agreement, and a recap.

What you will be able to do

• Implement corrective actions with documents, training and shift audits
• Validate effectiveness with production data and capability
• Remove containment only against exit criteria agreed with the customer

Why it matters

Waiting for production data before closing shows the action really works, because implemented is not the same as effective.

Practice: Practice: D6: implementing and validating · 3 questions

Starts with two ways the XYZ 8D could end: fixing line three only, or reading the lessons across all forty-two crimping tools in Munich, the sister plant's lines and every process monitor with adjustable limits. You then learn D7: changing the systems, practices and procedures behind the process, here the overhaul procedure and tool-life standard rewritten so life follows the anvil, and updating the PFMEA and control plan. The module covers lessons learned under IATF 16949 clause 4.4.1.2, recorded where people will find them, and a medical device plant in Boston reading across to other products. It ends with D8: closing with the customer's acceptance, the plant manager thanking the team by name for what each person contributed, and a recap.

What you will be able to do

• Update procedures, PFMEA and control plan to prevent recurrence
• Read lessons across to similar processes, products and sites
• Close an 8D with customer acceptance and formal team recognition

Why it matters

Fixing the weakness everywhere it could exist stops the same problem appearing on another line a year later.

Practice: Practice: D7 and D8: prevent recurrence and close · 3 questions

Opens with two reports on the same work: three lines with no data that the customer rejects, and one that shows the evidence behind every discipline. You then learn what a good 8D report shows, the gaps behind rejections to check before sending, and writing for a reader who may be reading in a second language late in the day. The module covers IATF 16949 clause 10.2.6 on communicating results to the customer and clause 9.3.2.1 on bringing poor quality costs and field failures into management review. As a reviewer, you return a weak report asking for evidence at D2, D4, D5 and D6. It then walks the XYZ case from start to finish, gives a road map, recaps the course and explains the final exam.

What you will be able to do

• Write an 8D report that shows the evidence for each discipline
• Review an 8D report and return it with specific evidence requests
• Track 8D performance as an input to management review

Why it matters

Reports that show the work get accepted faster, and reviews that ask for evidence catch weak 8Ds before the customer does.

Practice: Practice: Writing and reviewing 8D reports · 3 questions