Ask the Expert: Lean Leadership — Can We Talk About OEE?

Manostaxx – Industrial Management Consulting

Answer: Overall Equipment Effectiveness (OEE) is a powerful metric used to improve the effective use of resources, by machine.  Here’s the formula:OEE = Machine Utilization x Efficiency x First-Pass Yield

Utilization (hours of actual machine uptime ÷ scheduled hours): Typically, the biggest reductions to utilization are due to set-ups and maintenance downtime.  To reduce the impact of set-up times on machine throughput, you must measure and report the time spent on set-ups as a discrete measure for each machine.  You can do the same for time spent on PMs.  Data should be collected so that accounting can calculate the variance, +/- the standard.  Importantly, from a capacity standpoint it doesn’t matter on a constraint—it’s still lost production time.  Focus on why machines are down regardless of whether it’s planned or unplanned.  That will identify where the biggest opportunities lie so you can do a kaizen blitz on the right machines.  Always work on the constraint work centers first!  Let’s assume the utilization for the machine is 75% in a 120-hour week, i.e., 40 hours times 3 shifts x .75 = 90.  That means the machine is available for production 90 hours a week.  Lots of opportunity to improve throughput.

Efficiency (some call it Performance) (actual machine speed ÷ rated/standard machine speed) If the machine runs at 10% less than rated speed then the efficiency is just 90%; 90 hours x .90 = 81 hours.

First Pass Yield (% of WIP that flows through the process uninterrupted, i.e. no rework or scrap).  First pass yield is the quality metric and is 100%, minus material interrupted in process such as rejections/rework and scrapped WIP.  Let’s assume FPY is 88%. Note: I strongly recommend the use of the RTY (rolled throughput yield) as the much more accurate measure. [See my November 2016 article regarding rolled throughput yield vs. FPY.]    Typical FPY numbers I regularly see are in the high 80s/low 90s and always overstate reality.  Typical RTY numbers are in the 60s and 70s and are much more accurate and useful. Pareto-ize this data and improve the whole process, not just one operation of the process.  You’ll get to the root causes of flow interruptions much more quickly, and the improvement in performance will be much larger.

Using the above examples, OEE would be .75 x .90 x .88 = 59% OEE.

Next Steps:  Manufacturing/value stream managers should meet with the first line supervisors, quality manager, accounting and other staff as necessary to develop the plan of attack.  The first thing to do is to start collecting OEE data, by machine, on the constraint equipment in each value stream or department.  If you can get accurate information out of your current systems, great.  Many operations cannot from legacy systems.  If not, collect it manually and get started while challenging the quality manager to work with the necessary people to create a formal authorized system for use by all.  For the short-term, start with a 24-hour worksheet attached to a clip board at each machine and ask the operators to mark the page each hour, e.g., green marker for “machine running at rated speed,” yellow marker for “machine in set-up” or red marker, “machine is down.”

Collect these data first thing each morning and record into a spreadsheet.  Begin to track until the data points clearly to where the largest opportunities are.  This is where you will start.  It won’t take very long for the worst performing machines to be identified for action.  Then it’s time to simply create the improvement project with bold deliverables, assign scarce resources and get it done.  When the first project is complete go to the next biggest waste of capacity on the No. 2 constraint in the plant and so on.  The leverage on these opportunities is huge for the bottom line so staff multiple improvement projects if resources can be made available.

Please be sure when you ask operators for help that you educate them on what you’re going to do with the information they collect and why it’s important.  Be sure and let them know you also want to collect their insights into the process that go along with the data they’re recording.

Be sure to hold the first line supervisors accountable for regular follow up on gemba walks.  They can and must respond in real time to machines that are not performing well.  My favorite way to do that is with andon lights at each constraint.  Green light means quality, run speed, all good.  Yellow means there are issues with one or more components of OEE and corrective actions should be in progress.  Red means the machine is down. (Red andon lights should also include an alarm to command immediate attention.)

Finally, any operator or maintenance person I’ve ever talked to has a much better day when the machine is running properly than when it is not, so it’s usually easy to get their support.  They are a critical part of the team so make sure they feel that in all interactions — no finger pointing.

And salaried folks:  ACT WITH URGENCY and common purpose.  A lost machine hour on a constraint work center is an immediate hit to operating margin and maybe to your best customer’s order delivery promise.

“What you do has far greater impact than what you say.” — Stephen Covey

“Watch the little things; a small leak will sink a great ship.”  — Benjamin Franklin

Continue at:

The text above is owned by the site above referred.

Here is only a small part of the article, for more please follow the link

Also see:

Web Data Extraction

Leave a Reply

Your email address will not be published. Required fields are marked *