Lately, Overall Equipment Effectiveness (OEE) has taken center stage as the measurement of choice among manufacturers. Most have adopted it as the measure of their performance. A significant segment uses it in “real time” to control shop floor. It is common to see electronic boards hanging over manufacturing lines prominently displaying OEE along with other manufacturing measurements such as parts produced and quality.
A measure of manufacturing efficiency
OEE incorporates three critical aspects of manufacturing efficiency – productivity, equipment availability, and product quality. Following equation is its definition:
OEE = Productivity x Availability x QualityOEE condenses the complexity of a manufacturing operation in a single value. A decline in OEE is often viewed as losing profitability, and, thus, is highly undesirable. Foremen work diligently to meet or exceed the target on shop floors.
The popularity of OEE stems from a strong belief that high productivity coupled with high quality and availability is a prescription for high manufacturing efficiency. And, a highly efficient process operating at or above the target OEE meets the customer requirements of on-time-delivery of good quality products but also does so at low cost. In other words, a high OEE represents little or no waste in the process.
OEE has served manufacturing well in growth market. As demand increased, it became imperative to have high productivity and availability. Of course, meeting quality standards goes without saying; it is essential to stay in business these days. Manufacturers continued to improve OEE performance. Maintaining high OEE became an important strategy to ward off competition from low cost producers. It, therefore, is not surprising that improving OEE has become a cottage industry within the manufacturing improvement consultancies. It is hard to find a manufacturing enterprise software vendor who does not have an OEE module in their offerings.
Changing operating environments
Many manufacturers are facing the wrath of increased oil prices and declining US auto industry in the form of reduced demand for their products. Some have lost more than half of their volumes. They have limited choices: either run at full rates for shorter times, or run at slower speeds. Devising new operating policies are not simple as many other policy factors confound the issue. In any scenario, OEE takes a tumble. High productivity is difficult to maintain in a declining market because of lower demands unless assets are taken out of service.
The correlations between manufacturing cost and OEE, which apparently worked in the narrow range at near capacity operations, are now failing. Absence of such simple relationships is leaving managers to wonder about their decision making process. The apparent “linearity” of correlation in the narrow range is no longer valid as the operations have moved far beyond the vicinity of earlier operating conditions. The relationship between cost and OEE, in many cases, is exhibiting sudden changes like quantum shifts. Some managers are starting to ask how OEE is related to cost. Some have started to wonder whether OEE is at all related to cost. They want to move beyond the simple correlative approach to understanding the functional relationship. Can they continue to depend on OEE as a good measure of their performance? They don’t know any more.
OEE as measurement in setting operating policy
Making correct decisions require apposite measurements. This premise is elementary to every feedback decision-making process notwithstanding the type of decision – whether to control a machine, a process, or the whole business.
Most manufacturing improvement consultants and practitioners strongly believe that maintaining high process efficiency is the way to control cost. In other words, can OEE, which is a measurement of process efficiency, act as a surrogate for business performance? Let’s answer this by comparing OEE as a measure of performance in two instances. The scenarios represent two completely distinct operating policies for the same manufacturing line. Both are equally viable options. The scenarios are designed to have identical OEE but differ vastly in operating conditions.
The table and the figure below summarize the two scenarios. The blue scenario has high productivity and lower quality, while the red scenario has high quality and lower productivity and availability. The question is which one is better to select? The answer, of course, depends upon your business objective and not process performance. If the quality is of greater importance then you pick the red one, whereas if throughput is most critical then you pick the blue one. This selection is without regard to the cost to add value in either case. If OEE and cost are related by a direct and linear relationship then we would expect the cost structures for the two scenarios to be at least similar, if not identical.
Figure 1: Comparing two scenarios with same OEE
Relationship of OEE to cost and business performance
Since both scenarios have identical OEE, do they both have the same cost structure also? The answer is a most definite “NO”. In order to maintain high quality in the red scenario, we have to inspect more parts. This requires additional inspection stations and the associated resources. There is more rework to be done as well as increased scrap in order to meet the higher quality standard. Increased rework takes time away from what otherwise would be first time quality production resulting in lower productivity as manufacturing line produces less good parts per hour. With the lower productivity and increased activities for rework, we would expect the cost to add value per part would be higher.
In addition, a more stringent preventive maintenance program may be necessary to reduce variations in machine performances. This would reduce machine availability for production. In the red scenario in comparison to the blue one, increased quality does not increase the composite value of OEE as the productivity and the availability, both, decrease.
Both scenarios are two distinct modes of operation on the same manufacturing line. Despite both reporting the same composite value of OEE, their process dynamics, resource requirements, and financial performance are vastly different. The single value measurement of OEE does not reflect these differences. This example makes it clear that there is no one-to-one relationship between OEE, which is a measure of process performance, and the financial or the business performance.
In fact, one could question the claim of OEE, a composite value, as a suitable measurement at all for any purposes. Fortunately the situation is not that hopeless. OEE is a good indicator of process performance only when it is closer to the 100% marks on all three dimensions and the deviations are not large. In such cases the triangle on the radar chart approaches its boundary. If any one of the dimensions moves farther away from the 100% mark, OEE tends to lose its sensitivity and starts to break down.
OEE, the simplicity of which lured many to adopt it as a primary measurement, is not a good measure for process performance if the line is operating far away from full capacity. It is rarely a good indicator of the business performance. OEE is a good indicator for the foreman on the shop floor only when deviations in all three dimensions are small. Using OEE, however, for continuous improvement projects or to devise operating policies to face the challenges of the constantly changing marketplace is risky and warrants caution.