Shop floor KPIs are the numbers a crew can see and change during the shift. Six cover most lines: good units against the hourly target, stop minutes with the top cause, minutes until help arrives, rejects this hour, changeover time against standard, and safety events. OEE stays above them as the result, because a crew cannot act on a percentage at 10:40.
What makes a KPI a shop floor KPI
A number belongs on the floor only if the people there can move it today. Cost per unit is a real KPI, but an operator can do nothing with it at 10:40.
Run every candidate through three tests. If it fails one, it belongs in the office, on a list like the 12 production KPIs.
- Can the crew change it? A stop they can call help for passes. A material price does not.
- Does it update within the hour? A number that arrives next week cannot change this shift.
- Can they check it themselves? They should be able to count it at the line, without asking the office.
Six KPIs that pass the tests
Each one below answers a question the crew has anyway. Start with the first two, then add the others once those two are filled in every shift.
1. Good units against the hourly target
This is the main floor number, because it says whether the line is ahead or behind right now. The hour-by-hour board shows how to lay it out and set the target.
2. Stop minutes, with the top cause
Show today's stop minutes and the single largest cause. One name is enough to start a conversation, and ten causes on a screen start none.
3. Minutes until help arrives
Measure from the moment the crew calls until the person who can fix it reaches the line. This wait is often the part of a stop the floor can shrink without any new parts, and andon exists to shorten it.
4. Rejects this hour
Use a count, not a rate. "14 rejects since 10:00" tells a crew more than "99.2%", and it is easy to compare with the bin.
Longer term, track first pass yield in the office.
5. Changeover time against standard
Time each changeover from the last good unit of one product to the first good unit of the next. The guide on measuring changeover time gives the start and stop rules.
6. Safety events and near misses
Count them every shift and put them first on the board. A line that is on target after someone got hurt did not have a good shift.
Turn OEE into a number the crew can use
OEE is the score for the shift, so keep it visible. But translate it, or it stays a percentage that belongs to someone else.
The conversion takes one line of math. Example: a line has 480 scheduled minutes and an ideal rate of 50 units a minute.
Now "we need 75%" becomes "we need 1,200 more good units this shift". That is 150 units an hour over an eight-hour shift, a figure anyone on the line can follow.
The same trick works for stops. At 50 units a minute, a 5-minute jam costs up to 250 units, so the crew can see why a small repeat stop matters.
The office number and its floor version
You do not need two sets of truth. Each floor KPI is the live form of something the office already tracks.
| Office KPI | Floor version | What the crew does with it |
|---|---|---|
| OEE | Good units against hourly target | Sees the gap within the hour |
| Availability | Stop minutes and top cause | Calls help, fixes the repeat |
| Performance | Short stops counted, line rate | Restores the set speed, clears the jam source |
| Quality | Rejects this hour | Checks the setting, holds the batch |
| MTTR | Minutes until help arrives | Escalates when the wait runs long |
| Plan attainment | Units left in this order | Prepares the next changeover |
Example: one shift's stop minutes
Example: a filling line logs 50 stop minutes by 14:00. The crew's view shows the ranking below, and nothing else about stops.
Tomorrow's question is plain: what will we do about the misfeed? The crew does not need a chart with 14 categories to answer it.
These causes must come from the people at the line. Keep the pick list short and read the guide on operator downtime logging before you add codes.
How to set the targets
Base targets on your own recent shifts, not on a figure from somewhere else. A target nobody has come close to turns every row red, and people stop looking.
- Hourly units: ideal rate x scheduled minutes in the hour x the OEE you plan for. Example: 50 x 60 x 0.75 = 2,250 units.
- Stop minutes: take your median shift from the last month, then aim a little below it.
- Help arrival: agree a time with maintenance and team leads, and measure it for two weeks before you commit.
- Rejects: use the last month's normal count for the product, because some products reject more than others.
- Changeover: use the current standard for that product pair, or your median until you have one.
Move a target only when the line has beaten it for several weeks. The page on setting OEE targets covers the longer view.
Common mistakes
- Too many numbers. A crew can hold a few in mind during a shift, so keep the rest in the office.
- Yesterday's data only. A morning report is useful for a meeting, but it changes nothing on the line during the shift.
- Numbers without a next step. If red changes nothing, people learn to ignore red.
- Using the number to blame. Once a stop minute gets someone in trouble, stops stop being logged.
- Comparing crews on a ranking. Different products and different shifts make the ranking unfair, and crews know it.
- Showing a rate where a count would do. Counts are easier to check, and people trust what they can check.
Start with two numbers
Pick one line and put two numbers beside it: good units against the hourly target, and stop minutes with the top cause. Fill them in by hand if you have to.
After two weeks, ask the crew which number changed what they did. Keep that one, and add the next from the list.
When you are ready to show them, the guide to a manufacturing KPI board covers the layout. To check how floor numbers connect to the plant's goals, use a KPI tree.
Stop minutes only help when they are right and on time. Fabrico takes stops and counts live from the machine, and operators tag the cause at the line, so a crew can see its numbers during the shift. See how Fabrico tracks stops.
What people ask next
What is the difference between shop floor KPIs and production KPIs?
Shop floor KPIs are the subset a crew can see and move during the shift. Production KPIs also include numbers that only make sense over weeks, such as cost per unit. See the wider list of production KPIs with formulas.
Should operators see OEE?
Yes, as the shift result, with its three factors one level below. Pair it with a number in units, because 240 good units means more at the line than one OEE point. See how to explain OEE.
How many KPIs should a shop floor board have?
We suggest starting with two and growing to about six. If nobody can say what the crew did differently because of a number, take it off. The KPI board guide shows a layout.
How often should shop floor KPIs be updated?
Output and stops should update at least every hour, because that is the window in which a crew can still react. Safety and quality events go up when they happen. Longer trends can update once a day.
Which KPIs matter for a supervisor rather than an operator?
A supervisor adds the shift view: OEE with its factors, the top three stops in minutes, and open actions. A plant manager needs fewer numbers again, as shown in the plant manager KPI list.
Fabrico is a light MES for OEE. It takes stops, speed and counts from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal, and operators tag the cause at the line.