Quality management system audits routinely stall at Clause 7.2. Auditors request evidence of competence, and organisations present a binder of training records. Each record features a completion date, a trainer signature, and an employee initials box. The auditor ticks the clause, the certificate is issued, and the plant continues producing the same defect rates. The training record has proven attendance, not capability.
This gap between documented training and actual process knowledge is the most common structural weakness I encounter during audits. When a supervisor signs off on a training matrix, they confirm the operator sat through a session. They do not confirm that the operator learned the correct torque sequence, the critical-to-quality tolerances, or the specific failure modes listed in the PFMEA. The QMS requires competence, but the audit trail ends at exposure.
Training Within Industry (TWI), specifically the Job Instruction module, offers a rigorous mechanism for closing this gap. Originally developed by the US War Department in the 1940s, TWI demands that supervisors verify operator knowledge against a documented standard before releasing them to production. For an auditor, TWI transforms training from a subjective personnel function into an objective, verifiable quality assurance step with traceable evidence.
What Auditors Ask For vs What Proves Competence
During an IATF 16949 or AS9100 audit, the evaluator samples a few operators from the training matrix and asks to see their records. They check the dates and move on. This superficial check misses the actual failure mode. An operator can possess a perfectly completed training file and still execute the process incorrectly because the informal instruction they received contradicted the engineering standard.
I have audited plants where the control plan specified a 45-degree fastener angle, but the training matrix and informal buddy-system instruction advocated a 35-degree shortcut. The 10-degree deviation reduced cycle time by a few seconds but also reduced fastener penetration depth by 15%, creating intermittent cold welds. Every training record was perfectly compliant. The defect was designed into the training, not the process.
True competence evidence requires a cross-reference. The auditor must trace the control plan characteristic to the training document, and then trace the training document to the operator's actual knowledge. If the control plan specifies a torque sequence, the training material must explicitly teach that sequence. If the operator cannot explain why the sequence matters, the training has failed, regardless of what the paperwork says.
Training Record Audit Approaches
Compliance check
- Signed training matrix with completion dates
- Generic standard operating procedure on file
- Trainer signature authorising production release
- Computer-based training certificate printed and stored
Competence verification
- Job Breakdown Sheet mapped to control plan CTQs
- Operator verbally explains key points and reasons
- Supervisor follow-up logs showing declining frequency
- Defect data tracked against training deployment date
Job Instruction as Audit Evidence
Job Instruction (JI) provides the exact documentation an auditor needs to prove competence. The core artifact is the Job Breakdown Sheet (JBS). Unlike a standard operating procedure, which describes the whole process, the JBS itemises the major steps, the key points (how to do each step), and the reasons for the key points (why it matters for quality or safety). It is a precision training tool engineered to eliminate variation.
When integrated into the QMS, the JBS becomes auditable evidence. Each control plan characteristic maps directly to a key point on the JBS. If a PFMEA identifies a specific torque specification as a control measure, the JBS must list that torque as a key point and explain that incorrect torque causes joint failure. This closed loop ensures the operator's training matches the engineering specification precisely.
The PPAP process offers a natural integration point. When a new process or revision is launched, the JBS should be submitted alongside the control plan and flow diagrams. The JBS demonstrates that a structured training method exists for the revised process before the first production run. It transforms the launch from a risk into a controlled event, and it gives the auditor a concrete artifact to review.
The discipline of JI requires supervisors to follow a rigid four-step sequence. They prepare the worker, present the operation step-by-step, have the worker perform the task while explaining the key points, and follow up on a declining schedule. This follow-up schedule generates a documented audit trail. The auditor can verify not just that training occurred, but that supervisory verification continued for days or weeks afterwards.
Auditing Job Methods for Continuous Improvement
Clause 10.3 requires continual improvement, but auditors rarely find objective evidence of supervisor-led process changes. Organisations present Six Sigma project charters and kaizen event summaries led by specialists. The frontline supervisor's role in improving their own process is invisible. TWI's Job Methods (JM) module generates the missing audit evidence by structuring how supervisors analyse and optimise their operations daily.
JM relies on the 5W1H method to question every detail of an existing operation. The supervisor documents the job, questions every step, eliminates unnecessary elements, and rearranges the sequence for efficiency. The new method is then standardised using a revised Job Breakdown Sheet. When this cycle operates on the shop floor, it produces tangible evidence: before-and-after process documentation linked to measurable defect reduction or cycle time improvement.
From an audit perspective, JM provides traceable improvement data. When a supervisor eliminates a step from an operation, they remove a potential failure mode from the PFMEA. The internal auditor can trace that process change back to a specific JM review, verify the resulting defect reduction, and confirm the updated JBS was deployed to the operators. This traces the improvement from the gemba to the management review.
Corrective actions provide another integration point. When an 8D investigation identifies process layout or operator movement as a contributing factor, the corrective action should mandate a JM review. The resulting documentation becomes objective evidence that the root cause was addressed at the process level, not just documented on a form. The auditor sees a closed loop: defect identified, process improved, standard updated, training verified.
Measuring the Human Variable in Quality Audits
Job Relations (JR) addresses the friction and interpersonal conflict that destroy process stability. Disengaged operators do not report near-misses; resentful operators hide nonconformances. While these behaviours resist traditional audit measurement, their consequences are visible in the data. An auditor investigating erratic scrap rates or sudden spikes in customer complaints should look for evidence of supervisory inconsistency and poor conflict resolution.
JR teaches supervisors to gather facts, weigh options, take action, and check results. It demands consistency. A supervisor who ignores a safety violation on Monday but issues a formal warning for the same act on Friday destroys the psychological safety required for transparent quality reporting. The auditor can detect this failure mode by examining the correlation between disciplinary actions and near-miss reporting.
Audit Indicators for Training Effectiveness
JR also builds habits that generate measurable quality outcomes: providing regular feedback, giving credit, communicating changes in advance, and utilising individual abilities. These are not soft skills. They are the mechanisms that ensure operators feel empowered to stop the line when they spot a defect. An organisation with a healthy JR culture will show higher near-miss reporting rates and lower escaped-defect rates.
Building the TWI Audit Trail
Deploying TWI to satisfy audit requirements starts with Job Instruction. Organisations should not attempt a simultaneous rollout of all three modules. JI provides the most immediate, quantifiable return and the clearest audit evidence. Plants typically see reductions in first-pass defect rates and faster time-to-competency for new hires within the first month of disciplined implementation.
Internal auditors must be trained to look beyond the training matrix. Audit criteria should include operator knowledge verification: asking the operator to explain the key points and reasons from the JBS while performing the task. If the operator can perform the task but cannot explain the key points, the Job Instruction has failed. The standard is not the document; the standard is what the operator actually knows and does.
A training record proves attendance. A Job Breakdown Sheet proves intent. Only the operator's verbal explanation proves competence.
Control plans should reference specific Job Breakdown Sheet numbers, creating a cross-reference that auditors can trace. Management review meetings should track OEE, scrap rates, and Cpk stability on lines where supervisors actively apply TWI versus lines where they do not. The data will confirm what Toyota discovered decades ago: structured instruction and systematic improvement outperform sophisticated systems that lack foundational shop-floor discipline.
The Auditor's Checklist for Competence
An effective audit of operator competence requires three layers of evidence. The first is the engineering layer: the control plan and PFMEA defining what must be controlled. The second is the training layer: the Job Breakdown Sheet prescribing how the operator is taught. The third is the verification layer: documented follow-up showing the supervisor confirmed the operator retained the knowledge over time.
Most audits stop at the first layer and sample the second. They verify that a training document exists and that it was signed. They almost never reach the third layer. This is why organisations pass their surveillance audits while continuing to produce defects categorised as operator error. The audit methodology itself becomes part of the problem, because it validates the paperwork instead of the process.
To close this gap, audit checklists must be revised. Instead of asking to see training records for a sampled employee, auditors must ask the supervisor to explain the key points for that employee's operation. Then they must ask the employee the same question. If the answers do not match, or if neither can articulate the reasons behind the key points, the finding is clear: the training system is not functioning, and the QMS is vulnerable at the point of production.
This approach shifts the audit from a documentary exercise to an operational verification. It tests the system where it matters: at the gemba, where the operator, the supervisor, and the process converge. TWI provides the structure that makes this level of auditing possible. Without Job Breakdown Sheets, there is no standard to audit against. Without the four-step JI sequence, there is no follow-up to verify. The framework and the audit methodology are inseparable.