Production efficiency: what it is, the formula, a worked shift and how to raise it

The production efficiency formula

Production efficiency (%) = actual output / standard output x 100, where standard output = standard rate x planned production time. Both counts must cover the same time and the same product. If the line ran for 25 days of a 30-day month, compare 25 days of output with 25 days of standard, rather than stretching one figure to fit the other.

When a line makes several products: standard hours earned

Units of different products cannot be added. A cell that makes a 2-minute part and a 5-minute part has done more work making 40 of the slow part than 40 of the quick one. The fix is to turn output into time: give each unit the standard time it is worth, add those up, and compare the total with the time worked. Production efficiency = standard hours earned / hours worked x 100. Labor reports usually call this labor efficiency, and it is the same ratio as actual over standard output whenever the line makes one product.

The same table shows why efficiency needs a delivery measure beside it. The cell earned 97.8% of its time, but it made 30 fewer brackets than planned and 10 housings nobody asked for. Schedule attainment, capped at plan for each item, is (120 + 30) / 180 = 83.3%. The manufacturing KPIs guide explains the cap.

Production efficiency vs productivity vs OEE vs utilization

These words are used as if they meant the same thing. They do not. Each one divides a different pair of numbers and answers a different question, so the same shift scores very differently on each. Here is the packing shift from the example, measured six ways.

Why 92% and 75% on the same shift? They use different yardsticks. Efficiency measures against the standard rate of 100 cases an hour; OEE measures against the ideal 120 and counts only good cases. Set the standard rate at the ideal rate and count only good cases over planned production time, and the two become one number: 675 / 900 = 75.0%. So a plant that reports both should say which standard it uses. The OEE calculation guide works OEE through by hand, the OEE calculator does it from shift counts, and the TEEP calculator adds the unscheduled hours.

Productivity is output per unit of input and carries a unit, such as cases per labor hour or tonnes per kilowatt hour. Efficiency is a percentage against a standard. The two can move apart: a plant that buys a faster packer raises productivity at once, but if the standard rate is raised to match, efficiency can fall while the new line beds in. For people rather than machines, the OLE calculator splits labor effectiveness into availability, performance and quality.

Utilization is the one most often read the wrong way. A machine that runs every hour of the day is not a goal in itself: if the orders are not there, running it builds stock, ties up cash and hides problems. Raise utilization only on the step that limits output, and only when the demand is there to use it.

How to improve production efficiency in six steps

Start by measuring, not by buying. A new machine on the wrong step adds cost and no output. These six steps work on one line at a time.

1. 1

   Check the standard

   Time the current method with enough cycles to trust the average (the time study sample size calculator says how many) and add the allowances. A standard the line beats every week cannot show a loss.

2. 2

   Measure the gap hour by hour

   Write the plan and the actual on the hour-by-hour board each hour, with a reason for every hour that misses. A shift total tells you that output was lost; the hours tell you when and why.

3. 3

   Rank the losses

   After two to four weeks, add up the lost minutes by reason with downtime reason codes and draw them with the Pareto chart maker. Put a cost on the top ones with the downtime cost calculator.

4. 4

   Work on the constraint first

   Find the step that limits the line with the bottleneck calculator. An hour gained at the bottleneck is an hour gained for the whole line; an hour gained anywhere else mostly adds stock in front of it, which is the central lesson of The Goal by Eliyahu Goldratt and Jeff Cox.

5. 5

   Remove the biggest loss with the right tool

   SMED for changeovers, preventive maintenance and TPM for breakdowns, error proofing for scrap, line balancing for a station that cannot keep up, standard work for methods that vary by shift. Run each one as a kaizen with a before and after measure.

6. 6

   Hold the gain

   Update the standard work, the standard rate and the target, then review the line's efficiency, OEE and schedule attainment at every shift huddle so a slip is seen the next day.

The kaizen savings calculator turns the minutes and scrap saved into yearly savings and payback, which helps when the fix needs money. For more ideas on what to try once the top loss is known, read the seven ways to improve production efficiency and the plant-wide cost reduction page.

Common mistakes when measuring production efficiency

• A standard set from history. If the standard rate is last year's average output, the line runs over 100% whenever it has an ordinary good week, and the losses disappear from view.
• Never updating the standard after an improvement. The line in the example would read 98% for years while a better method was available.
• Dividing by run time instead of planned production time. On the packing shift, 690 cases against 100 an hour for 6.5 run hours reads 690 / 650 = 106%, and the 60 stopped minutes vanish.
• Adding units of different products. Use standard hours earned whenever the products have different standard times.
• Counting bad units as output. Either count good units only, or report first pass yield right beside efficiency.
• Chasing efficiency on every step. Above the bottleneck it builds stock; below it there is nothing to run. Push it at the constraint.
• Averaging percentages across lines. Add up the outputs and the standards first, then divide; the manufacturing KPIs guide works an example.
• Using the number to rank operators. Most of the loss sits in machines, material and methods. Blame teaches people to stop writing reasons on the board.