In a previous blog post, I described a potential environment for the “Road to Reliability”. Based on the positive feedback received from that posting, I want to take you further on “the road” with a series of posts that dive deeper into the subjects of reliability, availability and risk.
Previously we described the “on-ramp” to that road as measuring the “availability” of the assets in your organization. So what exactly is availability and how should it be measured? In a broad sense availability is the capability of an asset or an organization to produce products that are correct, on time, on budget, and meet the needs of the market.
But to simplify the measurement of availability, we will need to eliminate items such as; redundancy, failure rates and supply chain; and break availability down to its primary elements of production.
Next, let’s put each of the production-related items associated with availability under the lens to get a better feel for what it means to measure availability beginning with the ’on-time‘ component.
To measure the ’on-time‘ component, you will need to define your ’useful production time‘. This is the number of hours per day, per week, per month etc. that you plan to produce products.
Typically this is done in the context of a production plan which contains the quantity of products that you have planned to produce in a given time frame. Multiplying your useful production time by your maximum sustainable production rates will enable you to create a production plan. With the plan firmly established and agreed upon you then will need to enter actual production data. This is the quantity of products that were actually produced in the planned time frame. Any time the actual production quantity is less than the planned production quantity this can be considered as a ’loss‘.
In order to measure the ’on-budget‘ component of the equation, you will need to monetize or normalize your production and loss information to common units of measure that are understood and accepted across the organization.
The final element to be measured is ’meeting the needs of the market‘. This is most easily translated into the actual value of the product, or the adjusted value of the product, and is based on anticipated value (e.g. planned margin) as compared to actual value (e.g. adjusted margin).
Getting back to losses, we will need to evaluate each loss individually to determine the reason that the loss took place and to categorize the loss bases on accepted categories such as; Performance, Availability, and Quality.
It is only through the evaluation of loss that we can begin to determine the reasons that our losses are taking place, and to begin to put systems, and workflows in place to reduce either the probability or consequence of those losses. Having loss information readily available has the added benefit of determining ROI of a particular corrective action.
In many instances, there is resistance at the organization level to disclose detailed accounting information, as this could be considered as “anti-trust or “insider trading”. For this reason, organizations such as SMRP (Society of Maintenance and Reliability Professionals) came up with a standardized format for measuring what is known as OEE (operational equipment effectiveness)
OEE is defined as; “breakdown of the performance of a manufacturing unit into three separate but measurable elements: Availability, Performance, and Quality”. Each element points to an aspect of the process that can be targeted for improvement. Since OEE is measured in percentages, it is possible to compare different production areas and to easily roll-up OEE values, or conversely drill into specific areas and units within the production organization.
An alternative means of calculating OEE is based on dividing the minimum time needed to produce the parts under optimal conditions by the actual time needed to produce the parts. For example:
Total Time: 8 hour shift or 28,800 seconds, producing 14,400 parts, or one part every 2 seconds.
Fastest possible cycle time is 1.5 seconds, hence only 21,600 seconds would have been needed to produce the 14,400 parts. The remaining 7,200 seconds or 2 hours were lost.
The OEE is now the 21,600 seconds divided by 28,800 seconds (same as maximal 1.5 seconds per part divided by 2 actual seconds per part), or 75%.
So why do organizations remain so fixated with measuring downtime, and is downtime really a relevant measure of the availability of your organization to produce valuable products in the safest and most efficient manner?