1What is OEE?
Overall equipment effectiveness (OEE) is a measure of how well a manufacturing operation is utilized (facilities, time and material) compared to its full potential, during the periods when it is scheduled to run. It identifies the percentage of manufacturing time that is truly productive. An OEE of 100% means that only good parts are produced (100% quality), at the maximum speed (100% performance), and without interruption (100% availability).
Measuring OEE is a manufacturing best practice. By measuring OEE and the underlying losses, you will gain important insights on how to systematically improve your manufacturing process. OEE is the single best metric for identifying losses, bench-marking progress, and improving the productivity of manufacturing equipment (i.e., eliminating waste).
2Why should I implement the OEE in my factory?
OEE (Overall Equipment Effectiveness) is a universally accepted method to measure, with a single value, the potential improvement of a process. Without measurements it is impossible to improve productivity. Measuring the OEE allows to know the duration and frequency of the Losses and therefore, helps to carry out effective corrective actions and therefore increase productivity.
3Can OEE be used for manual processes?
Yes, but it would be better to use the OLE (Overall Labor Effectiveness) index. While the OEE was developed to measure the efficiency of equipment (machines, production lines), the OLE was developed to measure the efficiency of Labor. Another option would be to simply measure productivity as products per man hour.
4Is the OEE value sufficient to improve?
OEE measures efficiency, indicating how close to the ideal is the process, when it is produced without stoppages, at the highest possible speed and only good parts. Monitoring OEE by itself can increase productivity by the mere fact of controlling, but the strength of OEE as an indicator is that it allows us to quickly understand where the Losses come from, differentiating between Losses by Availability (Stops, downtime, failures, etc. .), Performance Losses (speed or performance) and Quality Losses (non-conforming products, scrap, rework, etc.)
By focusing on Losses, and especially corrective action, OEE value will naturally increase.
5What speed should I consider to calculate Performance?
In order for OEE to show us exactly how much we could improve our process, it is necessary to indicate the highest speed at which the machine could work for that particular product. Ideally, use the value indicated by the machine manufacturer and not the speed at which the company decides to work. If it did, the true capacity of our process would be hiding. On the other hand, performance measurement and, therefore, the OEE would increase artificially. On some occasions we saw yields higher than 100% (illogical) just because the machine ends up working faster than what is indicated as standard.
6How should I consider product changes?
Unfortunately, there are very few factories that may have a machine dedicated to a single product. The vast majority must make product changes and this implies a waste of time, since that time could have been used to produce.
If we remove the product changes we will increase the Availability and the OEE, but we would also be hiding (hidden factory) that we have the opportunity to increase efficiency by reducing the times of change or setup. To know how to reduce these times we suggest reading about SMED (Single Minute Exchange Die).
7Doesn't taking too much data conspire against productivity?
If data collection is done manually, it will surely be a new task that we will assign to the operators and this may result in a demerit of efficiency. It is an arduous task to take the real times of the stops, count the productions and the defective parts, and turn them into spreadsheets. In this case, we suggest taking the least amount of data possible so as not to impair the operator's performance.
On the other hand, if data is taken electronically with a system, the times, quantities and records are made without operator intervention and therefore it is very useful.
8Is it valid to use OEE for benchmarking?
Not always. Although it is an indicator that allows us to compare the efficiency between machines, sectors or plants, it should not be taken into account that OEE of a machine that makes a couple of products per day is not the same, against another that does 20 Different products daily. In general, this is what happens when comparing industrial plants in developed countries (large markets) with other emerging countries, where production runs are much shorter.
On the other hand, that two plants have the same OEE of XX%, the product mix Availability x Performance x Quality may change and this leads us to draw wrong conclusions. 60% Availability is not the same as 60% Quality.
Compare, but do it carefully.
9Do operators understand the concept of OEE?
No one is born knowing. That is why it is very important to train operators and supervisors, emphasizing how different losses affect the process. Ideally, explain the concept of the 6 large losses (Stops / Breakdowns, Preparation and Adjustments, Small Stops, Speed Reduction, Rejections for Commissioning and Production Defects). Then, progress is made on other probable Groups and Losses in particular. In this way, operators understand how the OEE relates to day-to-day work and adopt it quickly.
As an alternative, Efficiency can be measured as Produced in the Turn / Objective. It is a very motivating index, especially when the measurement and indication are done in “real time”, since everyone wants to reach 100% of the Objective. It is a good indicator but it is not enough in case the Objective is not reached, as the operators do not know why and do not provide suggestions to make improvements.
10In what process or on what machine of my production line should I measure OEE?
The OEE should be measured in the process that constitutes the “bottleneck”. In any production line (bottling machines, assembly lines, packaging, metalworking, etc.) there will always be an operation, step or machine that determines the amount of production possible. It is in this process that we must measure OEE, identifying both the losses of that process and the losses related to other equipment, machines or processes, either upstream or downstream of the “bottleneck”.
For those lines that are fully balanced so that all the processes that compose it run at identical speeds, the ideal is to measure in the first process, for example, in a bottling line to measure at the filler.
11How do I get line OEE instead of the OEE from a single machine?
Performance of the “bottleneck” is the performance of the complete production line. If the “bottleneck” runs at 1000 u / h and the palletizer runs at 2000 u / h, you can only manufacture 1000 per hour. Other processes can stop and start but while the “bottleneck” process is running you will be making money.
Many industries are usually guided by the number of pallets or boxes they send to the customer. Because of this, they often believe that they must measure the OEE at the end of the line. The unit count and the OEE can be treated independently. Only measure the OEE at the end of the line if that last process is your “bottleneck”.
12Why not measure OEE not only on the "bottleneck" but on all machines?
Measuring OEE at multiple points will only give you conflicting information and will probably lead you to focus on less critical aspects of your overall process. The best practice is to measure the OEE in your "bottleneck" and measure the mechanical efficiency of the rest of the machines. This will give a unique OEE value for the Line, as well as good information about the behavior of all processes, without complicating or creating confusion about OEE.
13What happens if my “bottleneck” changes to another process because I made improvements?
If that happens, it is the best thing that can happen. Celebration! It means that you made improvements to the “bottleneck” and it no longer presents restrictions. Change your OEE measurement point to the new “bottleneck” and start over. Do not forget that it is about Continuous Process Improvement.
14What happens if my “bottleneck” changes to another process when manufacturing different products?
If the “bottleneck” changes when manufacturing different products, in theory it would be correct to also move OEE measurement point. However, the economic cost of moving the measurement will have to be analyzed.
15Do product changes affect the OEE?
Yes. Product, format or reference changes should be included when calculating OEE as they affect Availability. These product changes require setup and adjustments; and they mean wasted time, since at that time it could be producing. In fact, Product Changes are one of the 6 Big Losses. To reduce the time of Product Change, the implementation of SMED (Single Minute Exchange of Dies) is recommended.
16Does Preventive Maintenance affect the OEE?
Probably. If Preventive Maintenance consumes time that could be used in value-added production - Producing on demand (MTO) instead of Producing for Stock (BTS) - it would have to be included in the OEE as it impacts Availability.
17Do coffee breaks and lunches affect OEE?
Including breaks and lunches as losses that affect OEE is generally considered correct. Consider that a work shift accumulates approximately 20 hours per month. If this time is excluded from the Availability calculation, and as a result of the OEE, then it would be hidden. However, if it is included, it is good to demonstrate that there is an opportunity to improve, for example, by creating relay groups so that the process does not stop.
18How should I consider Re - Work?
Reworked parts should be considered rejected or nonconforming. OEE quality index should be considered as the parts or units that are good in the first instance (First Pass Yield), that is, production units that come out with deliverable quality without being reworked. If we wanted to incorporate them as Good Parts after being reworked, it would only add complexity to the OEE calculation and we would be limiting the possibility of making good products from the beginning.
19How do I calculate OEE if I cannot detect rejections during the process, but hours or days later?
If you publish the OEE in "real time", the value will be reduced when you add the quality measurement. It is advisable to have quality measurements at intervals of the order of 30 to 60 minutes. This can be achieved with declarations made by the staff during the controls, or using statistical process control techniques, based on PPM (defective parts per million) that allow inferring the quality of the entire population from samples taken regularly.
If none of this were possible, it would be better to rely only on availability and performance, leaving aside quality; or use other types of indexes such as Produced / Objective.
20Should the OEE be based on quantities or times processed?
It depends on the nature of the process. For discrete manufacturing (for example: stamping, packaging, etc.), it is typical to measure quantities of manufactured parts. For process industries (for example: refineries) the typical thing is to measure in units of time.
21Do I have to rely on units, boxes or pallets to indicate the OEE?
Ideally, the minimum unit of product (part) should be used as the measurement helps more to understand the reasons or causes of the loss.
It also depends on which production unit the staff is most used to. If you always counted the tables, show OEE based on these.
22What period of time should I use for OEE calculation?
Use the same one with which they are accustomed to inform. Usually a shift, day, week, month is used; or calculate by production batches.
23To calculate OEE Performance, should I use the budgeted or ideal machine cycle?
The standard or budgeted cycle time is used for production scheduling and for cost calculation. The ideal cycle time is used in the OEE to identify opportunities for improvement.
The ideal cycle time is that indicated by the machine manufacturer for a given product and is the theoretical maximum that could be manufactured.
24Is it possible that any of the OEE variables give me more than 100%?
Definitely not! Having an OEE of 100% means perfect production, that is, producing only units of acceptable quality, as quickly as possible and without machine stops. Of course, the Perfect Process does not exist and will always have some Loss. Often, it can be seen in certain cases that the performance is greater than 100%, but this error is due to the use of a machine cycle or a slower speed than the machine can actually operate.
25Which OEE factor should be addressed first?
The factors that make up the OEE (Availability, Performance and Quality) have exactly the same importance. The really practical thing is to dump the problems in a Tree of Losses. Tree of Losses is a multi-level Pareto that orders losses of greater or lesser importance and, in turn, divides them into groups to better identify the factors (Availability, Performance and Quality). This Pareto analysis can be performed according to the time of each loss or can be weighted by frequency or severity. In any case, we suggest assigning an economic value to each Loss, that is, clearly indicating that said loss means so many pesos. This creates a greater commitment and understanding of the problems. Much more than indicating losses only as times or percentages.
26What happens if attacking the problem indicated as more serious in the Lost Tree is very expensive or cannot be enforced?
Tree of Losses gives us information about the real importance of each Loss and, therefore, if we could start attacking in the order indicated, we would increase OEE very quickly.
However, when solving a particular problem requires a strong investment or a change too deep at the moment, the ideal is to lose less weight but it is easier to solve.
To exemplify this, we usually talk about the apple harvest. Start with the apples on the floor or the low branches that are easier to access!