Ron Moore of the RMGroup Inc. related to me his best OEE experience was a client with a verified on-stream OEE of 98% over a two-year period.
Using OEE metrics and establishing a disciplined equipment performance reporting system will help any manufacturing area to focus on the parameters critical to its success. Analyzing OEE categories can reveal the greatest limits to success. Forming cross-functional teams to solve these root problems will drive the greatest improvement and generate real bottom-line earnings.
The vast majority of these improvements usually come from non-capital projects. Changes to basic procedures often reduce bottlenecks. Changing supply or distribution policies can help manage bottlenecks. Significant equipment reliability improvement may result by changing maintenance methods or substituting different materials. Focused projects, such as Reliability-centered Maintenance5, can provide major increases in uptime. Improving performance through OEE involves several steps:
1.Calculate the OEE value for current performance (see method 3 in section 2.5).
2.Use discipline and be honest with the results. Compute the financial opportunities of improved throughput (use the model provided in chapter 3). Generate a realistic business plan of closing the OEE gap to world-class levels for your type of industry. At this point, accept the assumption that improvement programs will consist primarily of education efforts and focus teams collecting/analyzing data for root causes. Minimal capital is required and existing resources are usually adequate. Training time and participative education on improvement methods contained in this book are often 90% of the investment.
3.Assuming that the size of the opportunity is significant, commit to a pro-active agenda. Define the hierarchy of critical processes and bottlenecks. Set expectations for plant goals and rewards. (This step may require changing existing measures and reward systems.) Once the key bottlenecks are identified, they must be tackled. OEE and Constraint Management methods should work in synergy.
4.Once the goals are defined, and a plan for addressing the bottle necks is established, share this vision with the workers. Communicate the significance of the improvement and give the community a compelling reason to make the changes. At this time, identify the reward structure.
5.Educate all members of the community about OEE measures and how to collect and reconcile the information. For example, counters, time clocks and chart recorders may be needed on key equipment systems. Reports may need to be modified to categorize downtimes. Everyone should have a major portion of their performance appraisal and compensation linked to achieving the OEE goals. By understanding the categories for data collection and how losses impact OEE, synergistic teams will form. These teams can quickly eliminate root problems. Associated departments can support additional improvements.
6.Generate the resources (e.g., money, people, time, and training) to make the changes happen. Introduce new techniques and programs, as appropriate, including condition-based, predictive maintenance and reliability programs, Total Productive Manufacturing and Best Practices techniques, Statistical Process Control, mistake proof and fail safe techniques, supplier quality requirements and follow up, and quick changeover techniques for operations and maintenance repetitive tasks.
7.Use the OEE metrics at all levels of the plant. Share the results with all parts of the plant community. With good data collection, each improvement project should demonstrate the projected increase in OEE. By frequently posting OEE metrics, any distur bances to high productivity will surface and can be quickly investigated.
A Case Study Using OEE Metrics
In 1996,I was assigned to a highly automated film finishing work center staffed with about 140 people. It was organized into high performance work teams manning four similar equipment flowlines, 7 days a week, 24 hours a day. The area was lead by a cross-functional business team and was data driven. This area finished many different sizes and formats of product and was challenged with many ‘new’ formats and products, as well as methods of operation to improve inventories. Daily meetings were held to review ongoing performance of the “factory”, which included the output quantity, flow line availability, and equipment reliability expressed as an index of the four lines. Meeting the projected production schedules was critical for just-in-time delivery and avoiding overtime in related work centers.
Figure 1-1 Equipment Index versus Output Index
One advantage for the area was an automated Equipment Performance System (EPS) to gather information. The EPS system provided all of the information suggested for categories to compute a detailed OEE, including frequencies of events. However, in early 1996, OEE was computed only monthly and submitted to plant management. It was not being actively used as an online guiding metric.
New levels of output had been achieved at the end of 1995 and carried into week one of 1996. See week one of figure 1-1. Output projections for 1996 were even higher. This was because prototype equipment improvement projects appeared to be successful. Early in 1996, the equipment improvement upgrades were migrated over all four flowlines.
Although the impact of shutdowns on operating schedules was minimal, the equipment changes required procedure changes and retraining for operators and mechanics.
Figure 1-2 Operational MTTR
Almost from week one, 1996, the output began to decline from expected levels. The second quarter results were very serious with an under production of 10 percent. By week sixteen of 1996 the investigating team had reached a conclusion.
In general, the feeling was that equipment reliability was not good. It was reported that the modifications were causing more problems, and couldn’t reliably handle new products.
With this information, the technical community worked hard to make sure processes and systems were working properly. With intense focus, the equipment reliability index improved over the second quarter by approximately 10 percent. See the increase in the equipment index following week sixteen in figure 1-1 Equipment Index vs. Output per Day Index.
Yet, factory output was still going down.
A more thorough investigation followed.
In review of all parameters, the root cause was found to be operational downtime. However, a study of this category did not reveal any unique or significant single items of downtime, by machine section, crew or product.
Only after plotting Operational Mean Time To Restore (MTTR) did the understanding that the many little events, which used to take 0.8 minutes to restore, were now taking 1.1 minutes.
Over time, poor habits and interruption of concentration had diverted the attention from “making product”. See figure 1-2 Operational MTTR. This underscores the importance of being able to collect time and frequencies of category events.
Week twenty eight of 1996 was the specific “intervention” date when the results of the detailed investigation were shared with each crew. That date is noted in figures 1-1 and 1-2.
Once the community was presented with the information and convinced that they really could influence the outcome by re-focusing their attention to detail, the output per day began to recover. In fact, output did reach the higher levels as predicted with the equipment modification project.
This understanding