Value Stream Map Uptime: Uptime, Availability or OEE in the Box?

Key takeaways

• Put measured Availability in the uptime box: run time divided by planned production time, over real shifts.
• Add OEE to the data box, and use it for the capacity check against takt time.
• Uptime alone hides slow running, short stops and rejects, so the map shows capacity the line does not have.
• In the worked example, remembered uptime showed the filler 20% over demand. OEE showed it 2,160 bottles short each shift.
• The choice can move the bottleneck, and with it each kaizen burst on the future state map.

Write measured Availability in the uptime box of a value stream map, and add OEE beside it. Then check each process against takt time with OEE, not uptime. Uptime alone leaves out slow running and rejects, so the map can put the bottleneck in the wrong place.

This guide is from Fabrico, the OEE and light MES platform. It compares the three numbers on one bottling line and shows how each one changes the map. For the method itself, see our value stream mapping guide.

What does the uptime box on a value stream map mean?

Each process on a value stream map has a data box under it, with cycle time, changeover time, uptime, operators, batch size and shifts. The format comes from Learning to See, the 1998 workbook by Mike Rother and John Shook, published by the Lean Enterprise Institute.

Uptime should say how much of the planned time the process can really run. In practice it is often a number from memory, a good week or the machine supplier's sheet. That is where a map starts to drift from the floor.

Uptime, Availability or OEE: what is the difference?

Uptime and Availability are close, but uptime has no fixed definition. Our guide to availability vs uptime covers the gap. OEE adds the two losses a data box never shows: speed and quality. Our OEE vs availability guide explains why the two can sit far apart.

Which number should go in the uptime box?

Use measured Availability, so the box keeps its usual meaning. Then add one line for OEE, and use OEE for the capacity check, because the customer only takes good units.

• Uptime box: Availability, with the window next to it, such as "80.0%, 15 shifts".
• New OEE line: OEE over the same window, at least for the bottleneck.
• Cycle time box: the ideal cycle time, not a good day's speed.
• Capacity check: effective cycle time = ideal cycle time / OEE. Compare it to takt time.

Christoph Roser of AllAboutLean lists taking a good day's speed as the ideal among the three common ways OEE gets fudged. The same fudge makes a data box look better than the floor.

How does the choice change the map?

The bottleneck is the process with the longest effective cycle time. Change the number in the box and the bottleneck can move. When it moves, so does the future state map:

• Kaizen bursts go to the process that really limits output.
• Buffers sit in front of the real bottleneck, not the process with the lowest uptime.
• The capacity question changes from "do we need a new line?" to "which loss do we remove first?".

Our bottleneck analysis guide shows how to confirm the constraint on the floor.

Value stream map uptime example: one bottling line, three numbers

Sample line (illustrative numbers, not a customer): a sauce bottling value stream with a filler, a labeller and a case packer. A shift has 450 minutes of planned production time. The customer needs 33,750 bottles a shift, so takt time is 27,000 seconds / 33,750 = 0.80 seconds a bottle. Each bottle earns a contribution of $0.10, or €0.10 in a euro plant.

The old map used uptime from memory. The team then measured each process over 15 shifts:

• Uptime view. The filler, at 100 bottles a minute, can make 45,000 x 90% = 40,500 bottles a shift: 20% over demand. Each process beats takt. The labeller has the lowest uptime, so the kaizen burst goes there.
• Availability view. The filler makes 45,000 x 80.0% = 36,000, just 2,250 over demand. The labeller still looks worst, at 78.0%.
• OEE view. The filler also runs at 90.0% Performance and 97.5% Quality, so OEE is 80.0% x 90.0% x 97.5% = 70.2%. That is 45,000 x 70.2% = 31,590 good bottles, 2,160 short of demand. Its effective cycle time, 0.60 / 0.702 = 0.85 seconds, is over takt.
• The labeller is fine. Its OEE is 78.0% x 96.0% x 99.5% = 74.5%, about 40,200 good bottles a shift from 54,000 possible.

What it costs. Priced at $0.10 a bottle, the filler's gap is $216 a shift (€216), or $162,000 (€162,000) over 750 shifts a year. Today the plant makes it up with overtime. The old map sent the team to the labeller, which was never the limit.

The new filler data box reads: cycle time 0.60 s, changeover 25 minutes, uptime (Availability) 80.0% over 15 shifts, OEE 70.2% over 15 shifts.

How do you measure uptime for a value stream map?

1. Use the same time model as your OEE. Planned production time is the base, and changeovers count as lost time, as in the six big losses.
2. Measure per machine, not per line. A line number hides which box is the limit.
3. Take enough shifts: 15 or more, covering your normal SKU mix.
4. State your short stop cutoff. No standard sets it, so write yours down and keep it fixed.
5. Use machine data, not memory. Paper logs lose short stops, and IoT Analytics finds 54.4% of small and medium manufacturers still run on pen and paper or spreadsheets.

Then compare each effective cycle time with takt time and redraw the timeline.

What mistakes make a value stream map lie about uptime?

• Writing uptime from memory, or from the best week of the year.
• Using the machine's rated speed as cycle time while it really runs slower.
• Mixing time bases: one box on shift time, the next on all hours.
• Putting the line's uptime in each box.
• Leaving rejects out, although the customer only takes good units.
• Drawing only material flow. Rother and Shook's map also shows the information flow, the part many maps skip.

Where Fabrico fits

Fabrico is a computer vision enabled Manufacturing Performance Platform. OEE and light MES come first: stops are logged automatically, and OEE is live by line and shift. For a value stream map, that means measured data boxes.

• Availability, Performance, Quality and OEE by machine, over the shifts you choose.
• Losses priced in money and ranked, so kaizen bursts go to the costliest loss at the bottleneck.
• The operator confirms the reason in one tap, so changeovers and breakdowns are counted apart.
• A camera clip as evidence for a stop, where a camera watches the machine.

Fabrico measures the boxes. Your team draws the map and decides the future state. There is no free plan and there are no tiers: each plant gets a quote. See how a light MES fills in the information flow of the map.

Frequently asked questions

What is uptime on a value stream map?

Uptime is the share of planned time a process can really run, written in the data box under each process. It should be measured, not remembered. The clearest choice is Availability: run time divided by planned production time, over a stated number of shifts.

Is uptime the same as OEE?

No. Uptime, like Availability, counts only the time a machine is stopped. OEE also counts slow running, short stops and rejects. In the worked example the filler has 80.0% Availability but 70.2% OEE, and only OEE tells you how many good units it makes.

How do you calculate uptime for a value stream map?

Take each process's planned production time over at least 15 shifts. Subtract the stop time, changeovers included, to get run time. Divide run time by planned production time, and write the result in the uptime box with the number of shifts.

Should a value stream map use OEE instead of uptime?

Use both. Keep Availability in the uptime box so the map reads the usual way, and add an OEE line. Use OEE for the capacity check against takt time, because it shows the good units each process really makes in a shift.

What is a good uptime on a value stream map?

There is no single good number. What matters is whether each process, at its measured OEE, makes what the customer needs in a shift. A fast process with low uptime can be fine, and a process with high uptime can still be the bottleneck.

See your data boxes in a demo

Bring one line and a few recent shifts. We will show Availability, Performance, Quality and OEE by machine, ready for your value stream map.

Request a demo · Contact us