OEE for the Enterprise

  • September 22, 2014
  • SAP Control Systems & Engineers PVT. LTD.
  • News

By Salvatore (Sam) Castro, Director, SAP Line of Business Manufacturing, SAP Measuring Performance As Robin S. Sharma once said, “What gets measured gets improved.”  In the industrial space, the process of continuous improvement is founded on this concept.  There are always new and challenging ways to drive improvements into processes and workflows, but understanding their priority, cost, and impact on the rest of the system isn’t always easy to measure.  It comes down to tracking the resources – people and machines – which are performing the work, and then managing variations that arise from the standard process.  People are needed at the operations level to deal with variation, and there’s no shortage of machine tools to provide technical feedback on how well the process is running.  In fact, these tools have been refined so clearly that one can easily apply the concepts of manufacturing process improvement to improving things in your own daily life.   Take the example of commuting to work in the morning rush and trying to avoid traffic.  I might start the process by planning out a route via a GPS or an online maps website, and noting the expected duration of the trip.  Then after taking the route, I can compare the actual time to the planned time, and also the quality of my experience during the commute.  From there, I may try alternate routes depending on the given time of day and other changing factors, such as the weather, construction, annoying intersections, etc.  This is essentially what the Overall Equipment Effectiveness (OEE) metric is measuring, but applied to a manufacturing process.  While OEE isn’t perfect and can’t be looked at in a vacuum, it does cover many scenarios and it can be a tool that provides guidance to an overarching continuous improvement strategy like Short Interval Control, which we’ll get to in a bit. What is OEE Overall Equipment Effectiveness, or simply OEE, is a very common metric used to indicate performance.  The OEE calculation is a simple one and it is the product of a couple of other metrics which are measured as percentages (%): OEE = Production Rate x Availability x Quality In the commuting example, I use the GPS plan to represent a production capacity plan, the actual drive time to be the production rate, and stress level to represent the quality of the drive.  In short, this is how I determine the strength of my commute based on how good I feel, offset by how quickly I was able to reach my goal, and compared to how quickly I should have reached it.  Then applying the concept of Short-Interval Control as a continuous improvement strategy I can look for ways to improve the process. What is Short Interval Control Short Interval Control (SIC) is  has its roots in industrial process control and continuous improvement.  It was originally applied to equipment control feedback loops as a way for machines to auto-correct their operation based on feedback from various sensors and operator input. Consider this graphic: Graphic: http://www.leanproduction.com/short-interval-control.html

When applying Short Interval Control to improving business processes, it’s hard to overestimate the value of having consistent and easily accessible KPIs, like OEE, available at various levels of the organization.  The first step and foundation is the ability to consistently and accurately look back at what has happened up to ”now,” which is dependent on KPIs, like OEE.  The next step is to look forward, which is where analytical models and engines that are capable of predictive analytics come in.  Finally, the linkage of corporate goals and objectives based on cost to the business from actual and expected deviations enable people to setup a clear plan of action on how to address current and potential problems as they occur.  This type of end-to-end planning and guidance ensures that the personnel involved are all on the same page and have a clear business driver behind them. OEE Calculations and Complexities The OEE concept is not a new one and the vast majority of equipment vendors offer tools that will calculate an OEE, or at the very least, a technical efficiency metric (OEE minus quality) for that specific piece of equipment.  This is done in a bit of a vacuum from the rest of the process, but it does provide local feedback to the operator how well the equipment is running in accordance to its programmed plan.  However, this isolated OEE calculation brings up several questions: what about the rest of the facility? And how can the operation of equipment upstream or downstream from this equipment influence the OEE for the entire line?  And even more advanced - what about comparing the other lines in the same plant or other plants to each other?  The seemingly simple approach of determining how efficient “this” machine is running can become a very complex calculation when you scale it out to an Area, Line, Plant, or the Enterprise.  Variation in the configuration of lines also means different classifications of the same physical event. Even more complex, the movement of a bottleneck machine can occur based on a wide variety of circumstances.  Tracking where and when efficiency losses occur is a useful guide on how to improve the overall flow, obviously. But it can be difficult to achieve this in a meaningful way and on the varying timescales required for this analysis.  It’s one thing to present the real-time view of OEE from the equipment, but what about the ten year historical view by shift and material across one or many plants, for example?  These multiple views of OEE are important for process improvements, but to different people and for different reasons depending on job function and the issues they are tracking down.  The common view of OEE is the localized one of an operator trying to control the machine better during the shift. The more modern spin on OEE is trying to improve all operations to run more efficiently with the lowest overall cost back to the business. Enterprise Reporting and Visibility Increasingly, and with the introduction of new and better looking technologies, managers and executives are noticing the limitations of the traditional ways of capturing this data.  They are also keenly aware that many reports they receive are already outdated when they see them, making it impossible to take action on any issues within the right interval of time. The end result is that problems cannot be prevented; only mitigated.  Additionally, the technologies that are available today are often limited to only a single time interval and job function, and can lead to further complications around how to link various intervals together.  As a result, the data can become fragmented and isolated in such a way that upper management and operations have different views of exactly the same thing.  This underpins the need for an enterprise view that demonstrates how a common set of standards and tools provides a holistic view across the corporate landscape, and specifically around manufacturing performance management.  Without this type of harmonized view, it’s going to be tough for manufacturers and industrial engineers to build business cases around process improvements for their executives to sign off on. And after all, that’s the whole reason for OEE , whether it’s a plant operations, or just getting yourself to work on time.

Did you enjoy this great article?

Check out our free e-newsletters to read more great articles..