This article is for plant managers, production managers, shift leaders and the people in charge of lean, continuous improvement and maintenance who already track OEE but notice that little changes after the monthly review. It shows how to tell whether OEE is only reported in a plant or actually used, and how to translate it into shift meetings and corrective actions.
- A monthly OEE figure is a result metric.
- Measured per machine and shift, it becomes a steering metric.
- Thresholds, owners and due dates turn losses into actions.
- The effect is checked on the same measured OEE.
What TeamThink writes about result metrics and steering metrics
Weert Jacobsen-Kramer of TeamThink, a German consultancy for lean management in production and indirect areas, distinguishes two kinds of metrics in his article Kennzahlen Lean Management (Lean management metrics, in German). Result metrics (Ergebniskennzahlen) describe outcomes that lie in the past and can no longer be influenced, such as monthly sales. He defines steering metrics this way: “Bei Steuerungskennzahlen handelt es sich um Kennzahlen, die Merkmale anzeigen, die sich in der Zukunft auswirken.” (Steering metrics are metrics that show characteristics which will have an effect in the future.)
His example is the share of commitments kept: when that value drops, it is a warning for the manager, who can step in before the result suffers. Among the key lean manufacturing metrics, the article also lists OEE, and it recommends making deviations visible on digital or physical team boards so that the team and its managers can correct course early. This article takes that distinction and applies it to a single metric: when is OEE a steering metric, and when is it only reported?
The same metric, two roles
OEE is not by nature one or the other. It depends on how it is collected, when it is shown and what it is used for. A monthly figure for the whole plant, in the report to senior management, is a result metric: it describes what happened, and nobody can change it anymore. The same OEE, measured per machine and per shift and broken down by loss type, shows the next morning where the line is losing time right now. At that point a shift leader can still do something about it.
| Characteristic | OEE as a reporting metric | OEE as a steering metric |
|---|---|---|
| Time frame | Month or week, looking back | Last shift, current day |
| Level | Plant or department | Machine, line, shift |
| Content | A single percentage | Losses by availability, performance, quality and by cause |
| Data source | Shift logs and production reports, compiled after the fact | Captured automatically from the machine signal |
| Who looks at it | Senior management, finance | Shift leaders, team, maintenance |
| What follows | A comment in the report | A decision with an owner and a due date |
A simple test: ask which decision was made last week because of OEE. If the answer is that it was in the monthly report, OEE is a reporting metric. If someone can name a specific action on a specific machine, it is being used to steer.
Why OEE so often ends up only in reports
- It arrives too late. When OEE is calculated at the end of the month from shift sheets and production reports, the events are weeks old. Nobody remembers why the line ran slower on a particular Tuesday.
- It is too coarse. A plant average evens out good and bad machines. A number that sums up everything shows no team where to start.
- It shows no cause. A percentage does not say whether time was lost in long breakdowns, in many micro-stops or in a slower cycle. Without the loss type there is no matching action.
- Nobody owns it. If no threshold is set at which someone must act, the deviation stays a comment.
- It is estimated. Manually logged stops are rounded, and short stops are usually missing altogether. A number the team does not trust does not lead to decisions.
This is where TeamThink’s criterion helps: a steering metric shows something that has yet to take effect, early enough to intervene. Monthly OEE rarely meets that test. Its building blocks do: the number of micro-stops per shift, a growing average stop duration at one station, a cycle time that slowly drifts. These signals come before the weak monthly result.
Five steps from OEE to corrective action
- Set the level at which decisions are made. Steering happens at the machine or line, per shift. Show OEE where the team works, not only as a plant average.
- Break OEE down into its losses. Availability, performance and quality separately, plus the main stop reasons by duration and frequency. The shift meeting discusses a loss, not a percentage.
- Define thresholds that trigger a response. For example: one stop reason tops the list for several shifts in a row, or the number of micro-stops at a station rises well above its usual level. The threshold decides when to act, so it does not depend on how the day is going.
- Assign every deviation to a person and a date. In the shift meeting, a loss becomes an action: who checks what, by when? On the team board, the action sits next to the metric that triggered it.
- Check the effect on the same metric. After implementation, the measured OEE of the same machine shows whether the loss has gone down. Only then is the action closed. This is how the team learns which actions actually work.
Which targets make sense per machine is covered in Realistic OEE target: set it machine by machine, in steps. How to present the monthly result to management in hours rather than percentages is explained in Presenting production losses to management.
What changes in the shift meeting
When OEE is used as a steering metric, the question in the meeting shifts. Instead of “What was OEE yesterday?” it becomes: “Which loss cost us the most time yesterday, and what are we doing about it today?” An illustrative example: on a packaging line, the previous day shows many short stops at the labeler, each lasting only a few seconds. In the monthly report, this would have shown up as a slightly weaker performance rate. In the shift meeting, it becomes a task for maintenance and the line leader, with a due date, and the next day shows whether the number of stops is falling.
The reporting metric does not lose its value. Senior management still needs the monthly figure to assess trends and capacity. The difference is that this figure is now the result of many small decisions made on the shop floor, not the starting point of a discussion that comes too late.
How TEEPTRAK provides the data for steering
TEEPTRAK measures OEE, downtime and micro-stops automatically and in real time. A module is wired into the electrical cabinet and a sensor is mounted at the edge of the line, with no connection to the PLC. Installation takes about one hour per machine, and OEE is available within 48 hours. Every stop is recorded with its start, end and duration, and operators assign the reason directly at the line. The figures per machine and per shift are ready before the meeting starts. TEEPTRAK is used in more than 700 plants in over 30 countries, where OEE rises by 10 to 15 points on average.
Further reading
- TeamThink: Kennzahlen Lean Management (in German)
- Realistic OEE target: set it machine by machine, in steps
- Presenting production losses to management: from percentages to hours and capacity
FAQ
Is OEE a result metric or a steering metric? It can be either. As a monthly figure for the plant, it describes the past and is a result metric. Measured per machine and per shift and broken down by loss, it shows early enough where to intervene and becomes a steering metric.
How often should OEE be discussed in production? Where it is used to steer, in every shift or daily meeting, with the data from the last shift. The monthly figure belongs in the report to senior management.
Which signals show early that OEE is about to drop? Typical signals are a rising number of micro-stops, a growing average stop duration at one station and a cycle time that slowly drifts. These signals only become visible when every stop is captured automatically.
0 Comments