PDCA Cycle in Manufacturing: Examples and Templates

By Mirko Peters · 2026-11-11

The PDCA cycle in manufacturing explained: the four steps, a one-page template, a worked example with numbers and the mistakes that stop plan-do-check-act.

The PDCA cycle is a four-step model for carrying out change: plan, do, check, act. In manufacturing, you plan a change to a process, try it on a small scale, check the result against a baseline, and then either make it the new standard or start the cycle again. This guide is for continuous improvement managers who want to use PDCA in a plant. It gives you the steps, a one-page template, a worked example with numbers and a list of common mistakes.

Key points

• The PDCA cycle is a four-step model for change: plan a change, do it on a small scale, check the result and act on what you learned.
• In manufacturing, every cycle needs a baseline, one clear measure and a short trial, so the check step shows whether the change really helped.
• A one-page template with problem, baseline, target, cause, countermeasure, result and decision keeps each cycle readable and comparable.
• A failed trial is useful: the check step shows what did not work, and the next cycle starts with a better plan.
• The act step is where gains are won or lost, so write the new standard down or restart the cycle.

What is the PDCA cycle?

The American Society for Quality (ASQ) describes the PDCA cycle as a four-step model for carrying out change, meant to be repeated for continuous improvement. Its steps are to recognize an opportunity and plan a change, test the change on a small scale, review the results, and then act: fold the change into broader work if it succeeded, or restart the cycle with a different plan. See the ASQ page on the PDCA cycle.

PDCA has other names. ASQ links it to the Deming cycle, the Shewhart cycle and the plan-do-study-act (PDSA) variant. Many people call it the Deming wheel, after W. Edwards Deming, whose name is tied to the method. As a management method, it is plan-do-check-act in plain words: think, try, look, decide. The idea comes from the scientific method: form a hypothesis, test it, learn.

Why does it matter in a plant? Because most improvement fails not for lack of ideas but for lack of checking. PDCA forces the team to state what it expects, test it small and look at the data. It is an iterative process: each loop starts with what the last one taught.

What happens in each step?

Plan. Define the problem with a number, such as "scrap on line 2 is 3.1%". Collect baseline data, find the likely root cause (the 5 whys or a fishbone diagram help), and choose one countermeasure. Decide the target and the single measure you will use. The item24 article on the continuous improvement process gives a useful rule here: define just one measure for each PDCA cycle. If you change three things at once, you cannot tell which one worked.

Do. Test the change on a small scale. One shift, one machine or one product family is enough. Keep the trial short. Record what actually happened, including deviations from the plan.

Check. Compare the trial data with the baseline and the target. Ask what you learned, not only whether you won. Was the effect real, or did something else change, such as a different material batch? Be data-driven, and treat the check as an honest review.

Act. Decide. If the change worked, standardize it: update the work instruction, train the people and fold it into the wider process. If it did not, restart the cycle with a different plan. Either way, write down what you learned and use it to begin the next cycle.

Template: a one-page PDCA sheet

A template keeps cycles readable and comparable. Copy these fields onto one page, a whiteboard or a spreadsheet row:

• Title and owner: what the cycle is about and who leads it.
• Problem statement: one sentence with a number.
• Baseline: the current value, the period and the source.
• Target: the value you want and the date.
• Root cause: the cause you will test, and how you found it.
• Countermeasure: the one change you will try.
• Trial: where, when and for how long.
• Result: the measured value after the trial.
• Learning: what worked, what did not, what surprised you.
• Decision: standardize, adjust or stop.
• Standard updated: which instruction or document changed, and who was trained.
• Next cycle: the next problem or the next test.

Many plants pin one such sheet to the team board, so that the cycle is visible to everyone. A larger problem can use an A3 report, which follows the same logic on one sheet of paper.

Example: reducing label misprints

This example uses assumed numbers.

Plan. A packing line shows 12 label misprints per 1,000 units, a rate of 1.2%. The team finds that operators sometimes load the wrong label roll. The countermeasure is a barcode scan that checks the roll against the order. The target is 0.8% or less. The measure is misprints per 1,000 units.

Do. The team runs the scan check for two weeks on one shift. In that time the line packs 5,000 units.

Check. The trial records 35 misprints. That is 35 ÷ 5,000 = 0.7%. The reduction is (1.2 − 0.7) ÷ 1.2 = about 42%. The target of 0.8% is met. The team also notes that 20 of the 35 remaining misprints occurred on the first hour after a roll change, which points to the next cause.

Act. The scan check becomes the standard on all shifts, the work instruction is updated and operators are trained. A new cycle starts on the first hour after a roll change.

The same sheet would also be right if the trial had failed. A result of 1.1% would tell the team that the roll was not the main cause, and the next plan would look elsewhere.

How do you run PDCA as a weekly routine?

One cycle is a tool. A routine is a system. Implementing the PDCA cycle across a plant is mostly a matter of habit: the PDCA process works when the same few steps repeat in the same place at the same time. Start with one team, and let the first results spread the method. A culture of continuous improvement grows from many small, checked changes in business processes, not from a large launch. The check phase deserves protected time, because teams under pressure skip it first.

1. Keep a short list of problems on a board, ranked by effect on safety, quality, delivery or cost.
2. Hold a brief daily check at the board. Review trials that are running.
3. Protect a weekly slot of about an hour to plan the next test and review the last one.
4. Limit open cycles per team, so that work gets finished. Two or three open cycles are usually enough.
5. Review monthly with a supervisor or manager: which cycles closed, which standards changed, which problems return.

This routine is the practical engine of continuous improvement in many plants. In kaizen, the Lean Enterprise Institute notes that PDCA cycles are used to identify, implement and check changes. See the LEI lexicon entry on kaizen.

How does PDCA relate to quality management and Six Sigma?

PDCA is widely used in quality management. An Advisera guide to PDCA in the ISO 9001 standard maps the standard's clauses onto the four steps: planning, operation, monitoring and evaluation, and improvement. The same guide notes that a main goal of ISO 9001 is the continual improvement of the quality management system.

Six Sigma uses a longer structure called DMAIC (define, measure, analyze, improve, control). PDCA is lighter and suits fast tests on the shop floor, while DMAIC fits projects that need more statistics. Some plants use both: PDCA for daily process improvement, DMAIC for complex problems, and both help you optimize flow and quality. Organizational support, such as time and a place on the shop floor to look at the data, decides whether either method sticks. If you aim for operational excellence, you need a method that people actually use every week, and PDCA is often the simplest one.

Common mistakes

• Skipping the baseline. Without a before number, the check step proves nothing.
• Changing several things at once. You cannot tell which change helped.
• Making the trial too big. A large test is slow and risky. Start small and refine.
• Treating check as a formality. If the result is always "fine", the cycle teaches nothing.
• Forgetting the act step. A good result that is not standardized disappears at the next shift change.
• Never closing a cycle. Open cycles pile up. Close each one with a decision.

How software can help

A whiteboard and a template are enough for a start. When many cycles run in parallel, software can keep owners, dates and results in one place, remind people of due checks and show which standards changed. Look for a simple template view, a clear status for each cycle and easy access for people on the shop floor. Pick a tool after your team has run a few cycles on paper, so you know what you really need.

Next step

Choose one problem on your line this week, fill in the one-page PDCA template and run a short trial. After the check, decide: standardize, adjust or stop. That is the quickest way to learn how the PDCA cycle works in your plant.