November 1996 € Volume 6 € Number 11

Overall Equipment Effectiveness
and Evolutionary Operation

By George Johnson, CFPIM

This department is provided to answer technical questions regarding problems in production and inventory control. Readers are invited to contact George Johnson, APICS National Research Committee, Rochester Institute of Technology, College of Business, P.O. Box 9887, Rochester,

Dear APICS: What is overall equipment effectiveness?

Reply: I believe you are referring to a performance measure from the literature of total productive maintenance (TPM). In the book TPM Development Program, edited by Seiichi Nakajima, the author explains that the goal of TPM is to improve equipment effectiveness so that it can be operated to its full potential and maintained there. There are two main thrusts to achieving this goal: Quantitative, emphasizing improvement in total available and in productivity per period; and Qualitative, emphasizing reduction in the number of defective products and stabilization of quality.

Several losses interfere with the overall objective. They are described as:

1. Breakdown losses: These are losses of quantity via defective products and losses of time due to decreased productivity from equipment breakdowns. (Note the term breakdown.)
2. Setup and adjustment losses: These losses stem from defective units and downtime that may be incurred when equipment is adjusted to shift from producing one kind of product to another.
3. Idling and minor stoppage losses: Typically, these kinds of small losses are relatively frequent. They result from brief periods of idleness when between units in a job or when easy-to-clear jams occur.
4. Reduced speed losses: These losses occur when equipment is run at less than its design speed. Design speed may not be known. Materials or tooling may be off-spec and require special treatment. There may be fear of ruining equipment if it is run too fast.
5. Quality defects and rework: These are product-related defects and corrections necessitated by malfunctioning equipment. (Note the term malfunction.)
6. Startup losses: These are yield losses incurred during early production, i.e., from machine startup to steady-state.

Overall equipment effectiveness (OEE) is defined as:

OEE = Availability × Performance Rate x Quality Rate

The previously described losses relate to this measure as follows:

• Availability (operating rate): is enhanced by eliminating equipment breakdowns, setup/adjustment losses and other stoppages.
• Performance rate: is enhanced by eliminating minor stoppages, idling and speed losses.
• Quality rate (rate of quality products): is enhanced by eliminating quality defects in startup and in process.

Actual calculation of OEE is as follows:

OEE = Availability x Performance Rate x Quality Rate
where

Availability =	Loading time - Downtime
	Loading time

Loading time = Total available time - Planned downtime
Downtime = Unexpected losses

Performance Rate =

Output x Actual Cycle time	x	Ideal Cycle time
Loading time - Downtime		Actual Cycle time

Quality Rate =	Number of good products
	Input

OEE indicates the overall capability of a plant.


References
Banker, Shailen, "Revitalizing the Workplace: Total Productive Maintenance," APICS -- The Performance Advantage, August 1995, pp. 54-58.

Maggard, W., and D. Rhyne, "Total Productive Maintenance: A Timely Integration of Production and Maintenance," Production & Inventory Management Journal, Fourth Quarter 1992, pp. 6-10.

Nakajima, Seiichi, Editor, TPM Development Program, Productivity Press, 1989.

Patterson, J. W., W. Kennedy and L. Fredendall, "Total Productive Maintenance Is Not For This Company," Production & Inventory Management Journal, Second Quarter 1995, pp. 61-64.

Dear APICS: What is evolutionary operation?

Reply: Evolutionary Operation, or EVOP, is an application of experimental design to the production setting, originally proposed by statistician G.E.P. Box. The gist of the idea is to unobtrusively obtain information about plant processes that will allow planned improvements with minimal work and risk to operations.

Juran and Gryna describe EVOP as follows: "Essentially, a simple experimental design, run repeatedly, provides a routine of small systematic changes in a production process. The objective is to force the process to produce information about itself while simultaneously producing product to standards.

"Only small changes in the process factors are allowed, and the consequences of these changes must be detected in the presence of the many natural variabilities that surround the process. The repetition of an experimental design, commonly a 2 x 2 factorial with center point, permits the blocking of many of the disturbances that commonly influence production. Through the process of replication, the design provides steadily improving estimates of the main effects and interactions of the studied factors." (Juran & Gryna, 1988, pp. 26.29-26.30)

Results of these experiments can be visualized and analyzed using response surfaces (essentially, topographical maps linking the two factors being manipulated to the output effect being measured).

The procedures for EVOP are described in Juran's Quality Control Handbook, Fourth Edition. See Chapter 26, "Design and Analysis of Experiments." Source references are also cited for those with technical interests.


References
Box, G.E.P., "A Simple System of Evolutionary Operation Subject to Empirical Feedback," Technometrics, Vol. 9, 1966, pp. 10-26.

Box, G.E.P. and N.R Draper, "Isn't My Process Too Variable for Evolutionary Operations?", Technometrics, Vol. 10, 1966, pp. 439-444.

Box, G.E.P., and J.S. Hunter, "Condensed Calculation for Evolutionary Operation Programs," Technometrics, Vol. 1, 1959, pp. 77-95.

Juran, J.M., and F.M. Gryna, Juran's Quality Control Handbook, Fourth Edition, McGraw-Hill, 1988.