Lean tools

5S

In French: 5S

5S organises the workstation in five steps — sort, set in order, shine, standardize, sustain — so that an anomaly is visible at once.

What it is

5S takes its name from five Japanese words: seiri, seiton, seiso, seiketsu, shitsuke. In English: sort, set in order, shine, standardize, sustain.

It is often presented as a tidying method. It is one, but that is not the aim. The aim is to make the deviation visible: at a workstation where everything has a defined place, a missing part, a moved tool or a leak is seen in a second. 5S is the base of visual management, and a practical prerequisite for TPM as for standard work.

The five steps

Sort (seiri) — separate the useful from the useless, and clear out the second. It is the step that frees space and the only one that produces an immediately visible effect.

Set in order (seiton) — a place for everything, defined by frequency of use and by the motion, not by looks. Floor marking, tool shadows, labelling.

Shine (seiso) — clean, but above all inspect while cleaning. A leak is found by wiping, not by reading an indicator.

Standardize (seiketsu) — write down what was decided so that the first three S can be reproduced by someone who was not there.

Sustain (shitsuke) — short, regular audits, ownership by the team. It is the S that fails, and it almost always fails.

Holding the fifth S

A useful 5S audit fits in five to ten closed questions, takes ten minutes, and is scored on the spot. As soon as it takes half an hour and thirty criteria, it becomes a monthly chore handed to whoever could not say no — and it stops measuring anything.

Two settings make the difference. Who audits: the team assesses its own workstation, and an outside eye comes by periodically. A purely top-down audit produces compliance on the day of the visit. What is done with the deviation: a point raised with no action and no deadline teaches that the result has no consequence, and the score starts climbing on its own.

The result is displayed at the workstation, not in a spreadsheet. Same logic as visual management: what is not visible where the work is done does not exist for those who do it.

A 240 m² workshop, one project at a time

A machining workshop, 240 m², six workstations, three months of project. The figures below are not targets: they are the readings that were used to decide what to do next.

Sort. Of 210 tooling references listed at the workstation, 62 had not been used in twelve months — nearly one in three. Sorting cleared out 3.1 tonnes of equipment, obsolete parts and scrap combined, and freed 34 m², i.e. 14 % of the floor area. It is the step that produces the most spectacular figure, and also the one that demands the most decisions: half of the project time went into deciding, not moving.

Set in order. The time to find a tool was timed before and after: 4 minutes 20 on average, then 40 seconds. At eight searches per workstation per day over six workstations, that is roughly 2.9 hours recovered every day across the workshop — not on one person, spread over six. That nuance decides what to do with it: this time does not turn into productivity by itself, it simply makes it possible to absorb disruptions without overtime.

Audit. The checklist chosen holds 25 points, five per S, and takes ten minutes for the team itself to fill in. The score went from 9 out of 25 at the start to 19 out of 25 after three months.

And here is the figure that really counts: six months later, with no follow-up, the score had fallen back to 14. Not to 9 — what sorting and floor marking gained holds, because it is physical. But the five points lost are all on the last two S, the ones that rest on nothing but a habit. That relapse is not a failure of the project, it is the demonstration of what the fifth S costs: ten minutes of audit a week and an item in the team routine, indefinitely.

The practical lesson lies in comparing the two curves. What was gained through a physical change — tooling cleared out, locations marked, shadows cut out — stayed. What rested on discipline eroded within six months. A 5S project is therefore designed by trying to shift as many gains as possible from the second register to the first.

The common mistakes

Launching 5S across a whole workshop at once. The follow-up is diluted, nothing gets finished anywhere, and the team is left with the memory of a project that fizzled out. One pilot workstation taken all the way, visible from the others, convinces better than six half-done projects at the same time — and it provides an audit checklist already proven for what comes next.

Confusing the score with the result. From the moment the audit score becomes a target followed in meetings, it rises: criteria get interpreted kindly, and the workshop learns to pass the audit. The score is a thermometer for the team that uses it, not a steering indicator for management. The day it goes up into a dashboard, it stops measuring anything.

Treating 5S as a big spring clean. A Saturday is mobilised, before/after photos are taken, and six months later the workstation is back to where it started — with a team now convinced that "5S does not work". The fifth S is not a final step: it is a rhythm, and it is decided before starting, not after.

On the shop floor — Precision machining

A workshop cut the time spent looking for tooling during changeovers by marking locations and removing from the workstation everything that had not been used in the past month. The gain did not come from tidying but from sorting: half of what was cluttering the workstation had no business being there.

The other tools illustrated by this trade: machining, mechanical engineering and finishing.

Frequently asked questions

What do the 5S mean?

5S takes its name from five Japanese words beginning with the same letter: seiri, seiton, seiso, seiketsu, shitsuke. In English they are rendered as sort, set in order, shine, standardize and sustain. The order is not decorative: setting in order before sorting means carefully organising what should have been thrown away, and it is the most common sequencing mistake.

How long does a 5S project take?

Sorting and setting a workstation in order takes a few concentrated days, often a week. The first three S give an immediately visible result; the last two are not a project but a rhythm, and they have no end date. That is why it must be decided before starting who will audit, how often, and what will be done with the deviations raised — afterwards it is too late, the momentum has gone.

Why does 5S fail so often?

Because the first three S are an event and the last two a habit, and only the first are funded. A project run on a Saturday, photographed before and after, produces a real result that erodes within months without follow-up. The second cause of failure is imposition: a workstation reorganised by a methods department without the people who work there returns to its previous state, often for good ergonomic reasons nobody had asked about.

5S