Lean and TOC: complementary questions, distinct controls

Lean and TOC: complementary questions, distinct controls

Start with the outcome and the actual problem, not a contest between methods. Lean questions can reveal customer value, waiting, unevenness, overburden and barriers to flow. Constraints thinking focuses attention on what limits the defined system goal and how other work should support it. A setup or reliability improvement may serve both perspectives when it addresses a relevant limitation. Keep the control mechanisms distinct: customer takt, a scheduled pacemaker, kanban authorization and a constraint-based release rule are not interchangeable labels. Choose a bounded intervention, verify the whole-system effect and update the method as the evidence changes.

Follow the method

1. Shared starting point
2. Lean questions
3. TOC questions
4. One measured trial

Choose an intervention from a shared system problem

Fictional repair route: preparation can supply 14 eligible units per hour, test can accept eight, and dispatch can handle 12. Demand is 10. Test loses time to incomplete kits, while a proposal promises to raise preparation from 14 to 18. Team lead Farah must choose a useful next experiment.

Follow the method

1. System gap
2. Lean inquiry
3. TOC inquiry
4. Preparation speed
5. Trial

The next project needs a reason beyond its tool name

Select a bounded improvement from system evidence using Lean and TOC questions, preserve distinct release controls and evaluate the effect at the customer boundary.

Fictional repair route: preparation can supply 14 eligible units per hour, test can accept eight, and dispatch can handle 12. Demand is 10. Test loses time to incomplete kits, while a proposal promises to raise preparation from 14 to 18. Team lead Farah must choose a useful next experiment.

Role: Improvement lead with preparation, test, dispatch and planning owners.

Normal condition

The team improves reliable accepted delivery while maintaining quality, safety and manageable work.

The gap

A local speed project is being selected because it is easy to measure, without explaining how more prepared units will pass through test.

• The rates are supplied comparable teaching values for the same mix; observe the real constraint mechanism.
• Necessary safety and quality problems require response wherever they occur, including nonconstraint processes.

1. Start with the system outcome

   Farah defines accepted delivered units and compares the supplied rates with demand. Test at eight per hour is the first internal constraint hypothesis, with a modeled gap of two.

   Why: The same starting point lets the team ask different method questions without arguing about which label should win.

   Evidence: Customer/system boundary and the provisional constraint comparison are recorded.

2. Use Lean questions to expose the work

   Follow a kit to test with the operator. Identify missing information, search, waiting, rework, uneven release and burden. Check what the customer needs and what makes the work difficult.

   Why: The method should help people see a mechanism in the actual work, not merely assign a waste label to a department.

   Evidence: Incomplete-kit examples, their causes to investigate and their effect on test readiness are retained.

3. Use TOC questions to focus the response

   Ask how the kit issue limits useful test capability and how supporting release/preparation rules can serve the system. Compare that with increasing preparation speed alone.

   Why: More nonconstraint output may create additional WIP without raising accepted delivery. It can still be useful if it addresses a necessary condition or a demonstrated supporting dependency.

   Evidence: Kit-readiness trial has an explicit path to protecting test; preparation-speed proposal lacks that path so far.

4. Keep the selected controls distinct

   Choose a defined readiness and release trial. If the site uses consumption-based kanban, retain its authorization semantics. If it uses constraint-paced release with protective buffers, retain that method’s definitions and owner.

   Why: Similar-looking cards do not make kanban and drum-buffer-rope the same system. Mixing triggers can create duplicate or uncontrolled release.

   Evidence: One accountable release rule is written for the trial; no informal parallel authorization is added.

5. Test and evaluate the whole effect

   Predict fewer test interruptions, then measure eligible test completions, delivered output, WIP, quality and operator workload. Reconsider the focus if the limiting condition changes.

   Why: A local reduction in searching is useful evidence but not by itself proof of a customer-delivery improvement. A bounded trial can reveal either outcome.

   Evidence: Adopt/adapt/stop decision ties observed mechanism changes to system results and necessary conditions.

A hazard is found at preparation

The observation reveals an uncontrolled safety condition at the nonconstraint preparation station.

Use the required safety response immediately through its owner; do not defer it because test is the current throughput hypothesis.

System improvement includes necessary conditions. Constraint focus does not authorize unsafe work elsewhere.

Safety action and the separate flow-improvement hypothesis are both visible.

Run a practice session

Materials

• Route-rate cards, kit/document examples and two proposal cards
• Trial prediction and system-outcome sheet

1. State the outcome · 4 minutes

   What must reach the customer, and under which conditions?

2. Ask both sets of questions · 8 minutes

   What do operators see, and what limits useful system output?

3. Compare the office proposals · 8 minutes

   Could a readiness check simply move the queue?

4. Define the control and review · 5 minutes

   Which release rule remains authoritative?

Debrief

• Reject tool-first answers without an observed mechanism.
• Use the safety exception to check that necessary conditions remain explicit.

Write the same problem through both question sets, then choose one test with a falsifiable prediction.

Transfer into the work

Owner: System improvement owner with affected operators and process owners

Record: Problem evidence, trial/release rule and whole-system result review

Review: After a bounded representative trial and any changed constraint evidence

Evidence: Correct delivery, controlled WIP, reduced relevant interruptions and sustainable work

Revisit the mechanism or constraint hypothesis; preserve necessary controls and avoid multiplying incompatible release rules.