OEE Waterfall Chart in Excel: Build and Interpret Results

To make an OEE waterfall chart in Excel, put the starting 100% and each staged loss in a two-column table, insert a Waterfall chart, then set the first and final bars as totals. The chart shows how availability, performance, and quality losses bridge ideal production to the final OEE, making the calculation easier to explain and investigate.

What an OEE waterfall chart shows

OEE, or Overall Equipment Effectiveness, combines three measures: availability, performance, and quality. It is commonly expressed as OEE = Availability × Performance × Quality. Availability reflects the share of planned production time when equipment was running; performance compares actual output with the ideal output during run time; and quality represents the share of total output that meets quality requirements. The exact definitions and time categories should be agreed within the organization, because inconsistent inputs make comparisons misleading.

A waterfall chart displays a starting value, increases or decreases, and an ending value. In an OEE bridge, the starting point is often 100% of ideal production, while negative bars show the effect of successive losses. The final bar is the calculated OEE. This is useful for communicating where effectiveness was lost, but the bars are not automatically the same thing as downtime minutes, rejected units, or percentages of their own individual categories. They need to be converted to a consistent basis.

For example, if availability is 90%, performance is 83.33%, and quality is 95%, OEE is 71.25%. A staged bridge can show 100%, less 10 percentage points for availability, less 15 points for performance, and less 3.75 points for quality, ending at 71.25%. The order matters: performance loss is calculated after availability, and quality loss after both earlier factors. This makes the waterfall add up to the OEE result.

Choose a meaningful loss bridge

Before charting, decide what the audience needs to understand. A compact chart with three bars explains the OEE factors; a more detailed chart can split a factor into operational causes such as breakdowns, changeovers, minor stops, reduced speed, or rejects. More detail is useful only when the categories are defined consistently and the underlying data supports them.

Factor bridge: Shows availability, performance, and quality effects. It is a clear starting point for an overall OEE review.
Time-loss bridge: Shows how downtime categories reduce available production time. This explains availability, but by itself does not account for speed or quality losses.
Unit or capacity bridge: Starts from ideal units or capacity and subtracts the equivalent lost output. It can be intuitive for production teams if all losses use the same unit and period.

Avoid mixing units in one waterfall. For instance, a 30-minute breakdown, 40 rejected units, and a 5% speed shortfall cannot be added as raw values. Convert each cause to a common measure, such as percentage points of ideal production or equivalent good units, and make the conversion logic visible. If an organization uses a different loss-accounting convention, label it rather than implying it is the only valid OEE method.

Calculate OEE and staged losses

Here is a worked example with a planned production time of 480 minutes, 48 minutes of downtime, an ideal cycle time of one minute per unit, 360 total units produced, and 342 good units. The figures are illustrative. Enter the inputs in Excel as follows: B2 planned production time (480), B3 downtime (48), B4 total units (360), B5 ideal cycle time in minutes per unit (1), and B6 good units (342).

With planned production time in B2 and downtime in B3, calculate availability in B8 with =(B2-B3)/B2. It returns 90%, since 432 of 480 planned minutes were running. This assumes planned production time already excludes any planned stops that your chosen OEE definition excludes.

With total units in B4, ideal cycle time in B5, and run time calculated as B2 minus B3, calculate performance in B9 with =B4*B5/(B2-B3). It returns 83.333%, because 360 units at one ideal minute per unit represent 360 ideal minutes of output over 432 running minutes.

With good units in B6 and total units in B4, calculate quality in B10 with =B6/B4. It returns 95%. With the factor results in B8:B10, calculate OEE in B11 with =B8*B9*B10. The result is 71.25%: 0.9 × 0.83333… × 0.95. Format B8:B11 as percentages for readable calculation cells.

For a percentage-point waterfall, use numeric values rather than percentage-formatted fractions: 100, -10, -15, -3.75, and 71.25. With availability in B8, the availability loss in percentage points is =-(1-B8)*100, which returns -10. With availability in B8 and performance in B9, the performance loss is =-(B8*(1-B9))*100, which returns -15. With availability in B8, performance in B9, and quality in B10, the quality loss is =-(B8*B9*(1-B10))*100, which returns -3.75. Together the changes take the baseline from 100 to 71.25. The formulas assume the factor cells contain decimal percentages, as Excel stores them when formatted as percentages.

Build the chart in Excel for Microsoft 365

Create a small chart-source table with categories in A2:A6 and values in B2:B6:

Ideal production — 100
Availability loss — -10
Performance loss — -15
Quality loss — -3.75
OEE — 71.25

Select A2:B6, then choose Insert > Insert Waterfall, Funnel, Stock, Surface or Radar Chart > Waterfall. Excel draws the categories as a waterfall, but the first and final values may initially appear as changes rather than anchored totals. Right-click the first bar and choose Set as Total. Repeat for the final OEE bar. The first and last bars should now start at the chart baseline, while the three loss bars step down between them.

Check the arithmetic before styling. Starting at 100, subtracting 10, 15, and 3.75 gives 71.25. If the ending bar does not match the OEE cell, investigate the formulas, signs, units, or rounding instead of adjusting a bar manually. On Excel for Mac, the chart type is also available in current Microsoft 365 editions; the contextual menu or ribbon wording can vary slightly, but the chart still needs its start and endpoint identified as totals.

Format the chart for clear reading

Use a descriptive title such as “OEE loss bridge — Line 2, 2 November” and include the period and equipment scope somewhere near the chart. A percentage-point axis or clear labels help readers distinguish the chart’s values from raw percentages. In this example, the vertical values are points on a 0–100 scale, so a bar labeled -10 means a 10-percentage-point reduction in the bridge, not a 10% reduction of the preceding value.

Keep the visual hierarchy simple. Use one color for losses, a distinct color for the final OEE, and a neutral color for the starting total. Add data labels if they remain legible; avoid forcing long cause names into cramped category labels. If the chart has many causes, shorten labels and explain abbreviations in a nearby note. Include the OEE result as a number in the chart title or a callout only if it is kept linked to the source data, so it cannot silently become outdated.

For a chart with many loss categories, sort causes by impact only if that ordering makes the bridge easier to interpret and does not obscure a meaningful process sequence. Category order affects the visual path, even though the final sum is unchanged. Keep the source table adjacent to the chart or on a clearly named worksheet, and retain enough detail to trace each bar back to its source.

Interpret the result without overclaiming

The largest bar identifies the largest loss in the chosen bridge, not necessarily the best improvement opportunity. A large loss may be difficult or costly to address, while a smaller repeated issue could be easier to fix. Use the chart to direct questions: which event or condition caused the loss, how often did it occur, and is the estimate based on reliable records? Then examine the source data, rather than treating the chart as a diagnosis.

OEE is a summary metric, not a complete measure of factory performance. It does not directly report customer demand, production cost, safety, maintenance risk, or whether the selected ideal cycle time is appropriate. Comparing two machines or shifts is meaningful only when the definitions, time windows, product mix, and data collection rules are comparable. A high OEE number does not by itself establish that a process is safe or that its output meets every operational goal.

Common mistakes and troubleshooting

Mixing percentages and percentage points: A waterfall of staged losses uses percentage-point changes on a common 100-point baseline. Label the values and axis accordingly.
Adding the three factor percentages as losses: Availability, performance, and quality multiply. They are not generally additive. Calculate the staged reductions as shown above.
Wrong signs: Losses should be negative when the bridge descends. A positive loss value raises the running total.
Forgetting to set totals: If the starting or ending bar floats instead of anchoring to zero, right-click that bar and select Set as Total.
Using inconsistent counts or times: Ensure good units are a subset of total units, and make sure planned time, downtime, and run time refer to the same equipment and reporting period.
Rounding too early: Keep full precision in the calculations and round only the displayed labels. Rounded intermediate factors can produce a small mismatch at the endpoint.

If an inserted chart seems to have an unexpected direction, inspect the source values and category order first. Excel uses the sign and sequence of the data to build the steps. Recreating the chart from a clean, two-column table is often easier than trying to correct a table that contains subtotals, blank rows, or unrelated values.

When to use another Excel view

A waterfall is strongest when the main question is how a starting capacity becomes a final result through sequential gains or losses. A stacked bar chart is often better for comparing the composition of losses across several machines or weeks, but it does not naturally show a running total. A line chart suits OEE trends over time. A Pareto chart can help rank loss causes by magnitude, especially when prioritizing investigation, but it does not explain the multiplicative relationship between OEE factors. These views can complement a waterfall rather than replace it.

For recurring reporting, keep the source table in a consistent format and update its values from the same approved data source each period. A simple chart is easier to audit than an elaborate dashboard with hidden transformations. If several products have different ideal cycle times, calculate performance using the appropriate product-level values before aggregating; using one unsuitable cycle time can distort the result. Document the definition and assumptions beside the chart so another reader can reproduce the calculation.

Frequently Asked Questions

Can Excel make an OEE waterfall chart automatically?
Excel for Microsoft 365 includes a Waterfall chart type, but it does not infer OEE losses from raw production data. Prepare the start, calculated loss steps, and ending OEE in a table, insert the chart, and set the first and final bars as totals.

Should the OEE chart start at 100% or at actual production?
For a percentage-point loss bridge, 100% is a useful starting point because the steps can show how staged losses reduce ideal production to OEE. A chart beginning with actual units or capacity can also work, provided every bar uses the same unit and the endpoint reconciles.

Why do my OEE losses not add up to the final OEE?
The three OEE factors multiply, so their raw percentages cannot usually be subtracted as if they were independent additive losses. Calculate the sequential effects on a shared baseline, keep full precision, and check that the chart uses the same period and definitions as the OEE calculation.

Can I show downtime causes in the same chart?
Yes, if each cause is converted to the same basis and the categories are mutually consistent. Downtime causes can explain availability loss; speed and quality losses need their own calculations if the chart is intended to reconcile fully to OEE.

What does a negative waterfall bar mean?
A negative value reduces the running total. In the example, -10 means availability reduces the ideal-production baseline by 10 percentage points. It does not mean downtime was negative or that the machine operated below zero.

Try it: the Percentage Formula Generator: percent of, percent change, increase and decrease, with the formula.

More in Charts & pivot tables

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.