Level 3: Structured
A Level 3 plant scores between 400 and 599 on the Factory Excellence Index. Its routines and definitions are written down and followed; it reacts within the shift and attacks its main losses every week.
A 30-minute floor walk in a Level 3 plant
At shift change the outgoing team leader fills in the standard handover form: status, open issues, safety. It is read in the team leaders' office rather than at the line, so questions about an open issue wait until the next break.
The board at each line is updated every shift with output against plan and the main issues. It tells you how the last shift ended, not whether the current one is behind.
The plant meeting starts on time with a fixed agenda: safety, quality, output, issues. Stop reasons were entered during the shift from a reason list, and last week's Pareto is on the wall with the top items assigned to named people. The meeting still discovers some issues for the first time, because nothing happens at line level before it.
When the bottleneck stops, the operator logs it with a reason code and follows the written escalation rule for breakdowns. If the stop is a major one, a 5 Whys is written and the countermeasure goes on the action list with an owner and a date. It will be tracked until done, but nobody yet checks with data whether it worked.
Short stops are mostly missed, speed losses are invisible, and the line only knows it fell behind once the shift is over.
How the six pillars typically look at Level 3
The table follows the third anchored answer on each question of the model. Set it next to the pillar indices in your assessment report; the pillar pages give the full anchors.
| Pillar | Typical practice at this level |
|---|---|
| Performance visibility | Operators log stops during the shift from a reason list; short stops are mostly missed. The line knows it is behind at the end of the shift. One written performance definition across the plant, including planned time. Results posted at the line per shift or per day. |
| Daily management and shift routines | A daily plant meeting with a fixed agenda. A standard handover form covering status, open issues and safety. Boards updated every shift. Written escalation rules for major events. Supervisors follow a written daily routine: floor walk, meeting, checks. |
| Loss and downtime problem-solving | A weekly Pareto of losses by reason, with the top items assigned. 5 Whys or fishbone used for major events. Actions tracked until done. Planned maintenance respected and breakdown data shared. Changeover times recorded; short stops estimated. |
| Improvement engine | An annual improvement plan linked to plant targets. A central action list with owners and dates. Standard work instructions for key operations, updated when problems occur. Gains calculated with a common method; major projects followed up. |
| People and frontline ownership | A skills matrix per line, updated at least yearly. An idea system with a guaranteed answer time. Structured onboarding for new team leaders. Results recognised in team meetings. Regular surveys followed by actions. |
| Data and connectivity foundations | Bottleneck machines connected, data collected centrally. Ideal cycle times and reason codes defined for main lines. Some automated exports. A central database with access on request. Office and production networks separated. |
What keeps plants at Level 3
- Knowing at the end of the shift. By then the hours are gone. A Level 3 plant sees its losses accurately but too late to recover them within the shift.
- Treating 'done' as 'worked'. An action list that closes items when the task is finished rewards activity. Without a before-and-after check on the loss, the same reasons return to the top of the Pareto.
- One meeting doing everyone's job. When the plant meeting is the only tier, it discovers problems instead of receiving them, and it overruns. Line issues wait a day for a decision a team leader could take in five minutes.
- Standards that change only after a problem. If instructions are updated when something goes wrong but not when something improves, every improvement lives in the heads of the people who made it.
- Routines that hold only on good days. The supervisors' written routine is the first thing dropped when a line is down. Without adherence checks, nobody notices the drift.
Five moves that take a plant to Level 4
These moves take the third 'next move' on the relevant questions of the model, starting on the bottleneck.
- Month 1: an hour-by-hour board on the bottleneckReplace the end-of-shift entry with an hour-by-hour plan-versus-actual board, filled in by the team, with a reason next to every missed hour. A missed hour becomes something to act on during the shift, not something to explain the next day.
- Month 1: add a line-level tierHold a short line meeting before the plant meeting, so issues arrive from the floor instead of being discovered. Move the handover to the line as a five-minute face-to-face meeting in front of the board, where every open issue leaves with a name.
- Month 2: test escalation, capture stops at sourceLog every escalation for a month with the time it was raised and the time it was answered, then adjust the triggers. On the bottleneck, connect one machine signal (run/stop or a cycle pulse) so stops are detected automatically, and keep operators on one job only: assigning the reason.
- Month 2: price the losses, check the fixesConvert the top losses into hours and money with a cost per hour agreed with finance. Write triggers for a full analysis, for example any stop over two hours or the same reason three times in a week. Close an action only when the loss data shows a change over two to four weeks.
- Month 3: lock in the gainsPut standards under version control and make 'update the standard' a required step before closing any improvement. Keep the key measure of each closed project on the line board with a threshold that reopens it. Extend standard work to managers, with short layered process audits at the line.
At the end of the quarter, take the assessment again.
What Level 3 means for recognition
Level 3 is the first level eligible for Factory Excellence Recognition. The self-assessment alone is not enough: the plant requests verification, which is optional and free. It uploads evidence and holds a 45-minute video interview with two jurors.
Good evidence at Level 3 is what the plant already produces: a photo of the line board, the weekly loss Pareto, the action list with owners and dates, a skills matrix. Verified status is dated and valid for 12 months, and verified plants can appear in the public register if they consent.
Twice a year, each regional league jury reviews verified plants and grants recognition at Level 3, 4 or 5. The first cycle runs in the France and Western Europe league and the Americas league, with a cut-off of 31 March 2027. The other leagues run the assessment and report now, and open recognition once their juries are in place.
The overall level cannot be more than one level above the weakest pillar. To be shown at Level 3, every pillar must be at Level 2 or above. A plant verified at Level 3 now can also aim for the Most Improved distinction, which needs two verified assessments at least six months apart.
Questions
Is it worth requesting verification at Level 3?
If you want an outside view of your practice, yes. Verification tests your answers against evidence and a conversation with two practitioners.
Do we need connected machines to be at Level 3?
Not necessarily. The Level 3 answer on the Data pillar describes bottleneck machines connected with data collected centrally, but the overall level allows one pillar to sit a level lower. A plant with its other pillars at Level 3 and its Data pillar at Level 2 is still Level 3 overall. To reach Level 4, the Data pillar would need to be at Level 3 or above.
What evidence should we prepare?
Evidence of routines that run, not documents written for the occasion. Jurors will ask how each item is used, so choose ones your team leaders can explain. The detailed evidence requirements will be published on the rules page.