Overall Equipment Effectiveness in PCB Manufacturing

Overall equipment effectiveness is a single figure built from three ratios, and its value lies in the way it forces a shop to look at losses it would otherwise treat separately. The three parts are availability, which measures the time the equipment was able to run, performance rate, which compares the actual output with the ideal rate, and quality rate, which measures how much of the output was good. Multiplying them produces a number that is almost always lower than anyone expects, and that surprise is the beginning of the improvement.

The Three Components

Availability is the proportion of the planned production time during which the equipment was actually available to run, so it excludes breakdowns and changeovers. Performance rate compares the actual number of units produced with the number that could have been produced at the ideal cycle time during the time available. Quality rate is the proportion of the units produced that passed without rework or scrap.

Multiplying the three produces overall equipment effectiveness, and the multiplication is why the figure feels harsh. A machine that is available eighty five percent of the time, runs at eighty percent of its ideal rate and produces ninety five percent good parts has an effectiveness of about sixty five percent. Each of the three figures looks acceptable in isolation, and the product of them describes a machine that is producing two thirds of what it could.

Measuring Availability Honestly

Availability depends entirely on what is counted as planned time. If planned time includes every hour of the year, the figure will be low and will say more about the demand pattern than about the equipment. If planned time excludes changeovers and maintenance, the figure will look excellent and will conceal the largest losses. The definition has to be written down and applied consistently.

Equipment downtime should include the small stops as well as the breakdowns. A machine that stops for two minutes every twenty minutes loses more time than one that stops for an hour once a shift, but only the second appears in the maintenance log. Recording short stops requires either an automatic counter or a simple tally kept by the operator, and the tally is worth more than it appears. Short stops are the largest single loss at many machines and the one that no maintenance system records, which is exactly why they are worth counting by hand.

Measuring Performance Rate

Performance rate requires an ideal cycle time, which is the theoretical best rate for the process. The ideal time should come from the equipment specification or from the best observed run, and it must be fixed rather than adjusted to match current performance, otherwise the metric measures nothing. An ideal rate that is revised downwards whenever performance falls turns the metric into a description rather than a target. Where no reliable ideal exists, the best sustained rate over a full shift is a workable substitute.

A falling performance rate usually means small stops, slower running to avoid a defect, or a process that has drifted from its setting. In a board shop, a plating line running below its nominal rate may be compensating for a bath that is out of specification, and the metric will show the loss before anyone notices the cause. That early visibility is the practical benefit of measuring the rate rather than only the breakdowns.

Measuring Quality Rate

Quality rate covers scrap and rework, and it should be counted at the equipment rather than at the end of the line. A board that is scrapped in final inspection because of a defect created at drilling counts against the drilling process, not against the inspection step. Allocating the loss to the step that caused it is what makes the metric actionable.

The quality rate is also where the connection to the rest of the shop is strongest. A process with a high scrap rate consumes the capacity of every downstream step as well as its own, so improving quality at one machine raises the effective capacity of the whole line. This is why quality rate is often the component with the most improvement potential even when it appears to be the highest of the three figures.

Using It in a Board Shop

A board shop rarely runs a single machine in isolation, so the measurement has to be applied to a defined asset such as a drilling centre, an exposure unit, a plating line or a press. Applying it to a whole area mixes different processes and produces a figure that nobody can act on. Starting with the constraint is the most productive choice, because that is where a percentage point of effectiveness produces the most output.

The measurement should also be kept simple enough to sustain. Automatic collection from the machine controller is ideal, and where it is not available, a shift sheet with the run time, the stop time and the good count is enough. The purpose of the figure is to direct attention, not to be exact, and a rough number collected consistently beats an accurate one collected for three weeks.

Acting on the Result

The result is only useful if it is decomposed. A low availability points at breakdowns and changeovers, a low performance rate points at small stops and speed losses, and a low quality rate points at defects and rework. Each of those has a different owner and a different remedy, and the single figure is only a way of deciding which to work on first. Without the decomposition the metric becomes a score rather than a tool.

Improvement should start with the largest loss rather than the easiest one. In most shops the first large gain comes from changeover time and short stops, because both are large and both respond to discipline rather than capital. The second usually comes from quality at the bottleneck, and the third, the replacement of an unreliable asset, is the most expensive and should be last.

Practical Rules

Define the planned time, the ideal rate and the loss allocation once, and apply the same definitions every month. Measure the constraint first, keep the collection simple, and record the figures with the production records so that the trend can be reviewed.

Act on the largest loss and check the effect on the figure after the change. Overall equipment effectiveness is not a report for management but a way of ranking the losses at a machine, and it earns its place when it leads to a change that shows up in the quality and delivery data within the next month.

FAQ

What is a good effectiveness figure? It depends on the process, but many plants run between sixty and seventy five percent. The useful comparison is against the same asset a year earlier.

Why is the figure always lower than expected? Because three figures that each look reasonable multiply together. A small loss in each of the three produces a large loss in the product.

Should changeovers count against availability? They should be shown separately and included in the calculation, otherwise the largest loss at many machines remains invisible.