This article is for maintenance managers, plant managers, production managers and continuous improvement leads who have put a preventive maintenance plan in place, in house or with a contractor, and who have to answer a simple question a few months later: is it working? It does not cover how to build the plan. It covers how to follow it over time: how to check, machine by machine, that preventive work is really reducing unplanned stops, and what to do when it is not.
- Plan completion measures the effort, not the result.
- The result indicator is unplanned stops on each machine, related to run time.
- Planned downtime used by preventive work belongs in the balance.
- A baseline and a stable list of causes make the comparison reliable.
What Tecnia Industries writes about the maintenance plan
Tecnia Industries, an industrial maintenance contractor based in Malay-le-Grand near Sens (Yonne, France), published the article Maintenance préventive industrielle : construire et suivre un plan d’entretien (“Industrial preventive maintenance: building and following a maintenance plan”, in French) on 20 July 2026. It sets out the starting definition: “La maintenance préventive consiste à intervenir avant la défaillance, pour en réduire la probabilité” (preventive maintenance means acting before the failure, to reduce its probability).
Drawing on the NF EN 13306 standard, the article distinguishes time-based preventive maintenance, triggered at fixed intervals (operating hours, cycles or calendar), from condition-based preventive maintenance, triggered by the actual state of the equipment. Tecnia Industries then describes how a plan is built: an on-site assessment (machine criticality, breakdown history, manufacturer recommendations), followed by maintenance routines with intervals, checkpoints and wear parts to track. One point stands out: the interval set at the start is then adjusted based on the readings. This is where measurement comes in. To adjust a routine, you need to know what it has changed on the machine. This article adds that piece to the method: measuring, over time, the effect of the plan on downtime.
Plan completion and plan effectiveness: two different indicators
The first indicator most plants track is the completion rate: how many scheduled routines were done on the scheduled date. It is useful, but it measures the effort, not the result. A plan can be fully completed without breakdowns going down, if the routines do not target the right failure modes. Conversely, a partly completed plan can already have a clear effect on the two or three most critical machines.
The result indicator is the trend in unplanned stops on each machine covered by the plan: their number, their total duration and the mean time between failures (MTBF). Add one indicator that is often forgotten: the planned downtime consumed by the preventive work itself. In OEE, these interventions show up as planned stops. An effective plan saves more time on breakdowns than it uses for interventions. Tracking both on the same machine over the same period is the only honest way to strike that balance.
Step by step: tracking the effect of the plan machine by machine
- Freeze a baseline before the plan. Record the stops on each machine concerned over several weeks, with their duration and cause. Without a baseline, every later comparison relies on memory.
- Separate three families of stops. Planned preventive interventions, breakdowns, and other stops (adjustments, changeovers, waiting). If a preventive intervention is logged as a breakdown, or a breakdown as an adjustment, the assessment is wrong from the start.
- Date every intervention. For each routine carried out, note the machine, the date and the component. This is what lets you link a change in breakdowns to a specific intervention.
- Compare at equivalent load. Relate breakdowns to actual run time, not to the calendar. A machine that ran half as much this month will break down less for that reason alone, whether or not the plan is working.
- Look at the days after each intervention. Breakdowns or minor stops that appear right after a preventive visit often point to a reassembly, setting or restart problem.
- Adjust the routine from the readings. At a fixed interval, for example every quarter, review the interval and checkpoints of each routine using the machine’s figures, with the maintenance team or the contractor who carries out the plan.
Five situations to recognize in the data
Once the measurements are in place, most machines fall into one of the following situations. These are general patterns, not results from a specific plant.
| What the data shows | Likely reading | Action |
|---|---|---|
| Preventive work done, breakdowns down, planned time stable | The plan works on this machine | Keep the routine, then carefully test a longer interval |
| Preventive work done, breakdowns flat | The routine does not target the real causes of breakdowns | Compare the measured breakdown causes with the routine’s checkpoints |
| Preventive time rising, breakdowns flat | Over-maintenance: machine time used with no gain | Lighten or space out the routine, or move to condition-based maintenance |
| Breakdowns or minor stops right after the intervention | Reassembly, setting or restart issue | Review the end of the intervention and the restart with production |
| Interventions postponed, breakdowns up | The plan is not being followed on this machine | Protect intervention slots in the production schedule |
For example, a press that broke down about once a week before the plan and only once a month after a quarter of preventive work is in the first situation. If a neighboring press, covered by an equivalent routine, keeps the same breakdown rate, the problem is not preventive maintenance as such but the content of its routine. Comparing machine by machine avoids jumping to conclusions, in either direction, from a shop-wide average.
Two reading precautions. On a machine that rarely breaks down, a few weeks are not enough to conclude: you need a longer period, or to also track minor stops, which are more frequent. And the list of stop causes must stay the same before and after. If it changes along the way, the measured trend reflects the new classification, not the effect of the plan.
What TEEPTRAK adds to following the plan
TEEPTRAK measures stops and minor stops on every machine automatically and in real time. A module wired into the electrical cabinet and a sensor at the line are all it takes, with no PLC connection. Installation takes about one hour per machine and OEE is available within 48 hours. Every stop is time-stamped with its duration, and the operator assigns the cause at the workstation: preventive intervention, breakdown, adjustment or waiting. Machine by machine, you get the baseline before the plan, the planned time consumed by preventive work and the trend in breakdowns after each intervention. TEEPTRAK does not replace the maintenance plan or the expertise of the people who carry it out: it provides the measurements needed to adjust it. TEEPTRAK is used in more than 700 plants in over 30 countries, with an average of 30% fewer unplanned stops.
Further reading
- Tecnia Industries: Maintenance préventive industrielle : construire et suivre un plan d’entretien (in French)
- Planning maintenance windows around real machine stop patterns
- Opportunistic maintenance: bringing preventive work forward during unplanned stops
FAQ
How do you measure the effectiveness of a preventive maintenance plan? By tracking, for each machine covered, the number and duration of unplanned stops and the MTBF, related to run time, and comparing them with a baseline recorded before the plan. The planned downtime consumed by preventive work is tracked alongside.
Is the plan completion rate enough? No. It shows whether routines are done, not whether they reduce breakdowns. Read it together with a result indicator for each machine.
How long before you can judge the effect of preventive maintenance? It depends on how often the machine breaks down. On a machine that breaks down often, a few weeks give a first trend. On a machine that rarely does, you need a longer period, or to also track minor stops.
0 Comments