- Calculation rule
- total = 1 − (1 − a) × (1 − b) × … (effects are multiplied)
- Occupancy control
- 24 % on average (LBNL meta-analysis, Williams et al., LEUKOS 2012)
- Daylight harvesting
- 28 % on average (LBNL meta-analysis)
- Combined strategies
- 38 % on average (LBNL meta-analysis)
- Halls and warehouses with low occupancy
- up to 75 %, project-dependent (project values)
- Verification
- measurement before and after, same reference period, standby deducted (measurement plan)
What do the savings depend on?
The starting point is the reference consumption: the electricity the lighting uses without controls. It is the installed load multiplied by the annual operating hours. A measured value, for example from a sub-meter on the lighting circuits, is better than any estimate.
The control system then removes shares of this consumption. The main strategies are:
- Occupancy control: light only where someone is present, otherwise background light or off.
- Daylight harvesting: dimming as far as daylight through windows or rooflights allows.
- Institutional tuning: limiting the light level centrally to what is required, for example in over-dimensioned installations.
- Personal tuning: users adjust the light at their own workstation.
How large each share is depends on use, zoning and the building. Good zoning is therefore the precondition for high values.
Why are savings multiplied rather than added?
Each strategy only acts on the consumption the previous one has left. Once occupancy detection has switched the light off, daylight harvesting has nothing left to save there. The formula is therefore:
Total savings = 1 − (1 − a) × (1 − b) × (1 − c)
Two strategies of 30 % each do not give 60 %, but 1 − 0.7 × 0.7 = 51 %. Adding them overstates the savings, and the more strategies are combined, the larger the error.
Which savings values are realistic?
The best-known data source is the meta-analysis by Lawrence Berkeley National Laboratory (LBNL, Williams et al., LEUKOS 2012), covering 240 values from 88 studies.
| Strategy | Average savings (LBNL) |
|---|---|
| Occupancy control | 24 % |
| Daylight harvesting | 28 % |
| Personal tuning | 31 % |
| Institutional tuning | 36 % |
| Combined strategies | 38 % |
These are averages across many studies and buildings. In halls and warehouses with low occupancy, project values are often much higher. A guide value there is above 60 %, with sensor-based savings of up to 75 %, always depending on the project. Conversely, simulations often overestimate daylight. That is exactly why measurement after installation matters.
Worked example: a 5,000 m² warehouse
All values in this example are illustrative assumptions, not a commitment for any specific project.
| Item | Illustrative assumption | Result |
|---|---|---|
| Floor area | 5,000 m² | – |
| Installed LED load | 25 kW (5 W/m²) | – |
| Operating hours | 6,000 h/year | – |
| Reference consumption | 25 kW × 6,000 h | 150,000 kWh/year |
| Occupancy incl. background light (a) | 50 % | – |
| Daylight under rooflights (b) | 20 % | – |
| Institutional tuning (c) | 10 % | – |
| Total savings | 1 − 0.5 × 0.8 × 0.9 | 64 % |
| Gross savings | 150,000 kWh × 0.64 | 96,000 kWh/year |
| Standby of sensors | 40 W × 8,760 h | approx. 350 kWh/year |
| Net savings | 96,000 − 350 kWh | approx. 95,650 kWh/year |
| Electricity price | €0.20/kWh | approx. €19,100/year |
Adding the shares by mistake gives 80 % instead of 64 %. The difference of 24,000 kWh a year shows how much the calculation rule affects the result. The sensors' standby consumption runs around the clock, over 8,760 hours, and is deducted.
Calculations like this are also the basis whenever someone wants to treat the German obligation as economically unreasonable. The yardstick is the economic efficiency requirement in § 5 of the GModG (German Building Modernisation Act, Gebäudemodernisierungsgesetz). In essence, the expenditure must be recoverable through the savings within the usual service life. The exact assessment is made case by case. This is not legal advice; the statutory text is authoritative.
How are savings verified after installation?
A forecast is an expectation. Verification needs measured data. The approach follows the principle of measurement and verification:
- Baseline period before: measure the lighting's consumption before the retrofit over a period that covers typical shifts and weekdays. If there is no meter, a temporary measurement or a documented reference value from load and operating hours can be used.
- Same period after: measure over the same reference period after commissioning, ideally in a comparable season, because daylight varies.
- Same operating hours: if shifts or opening hours change, results are normalised to operating hours.
- Monitoring per zone: shows which areas save how much and reveals anomalies, such as bypassed sensors or hold times that are too long.
- Deduct standby: the own consumption of sensors and wireless nodes is subtracted from the result.
- Report: method, periods, meters, adjustments and the result in kWh and per cent.
With SensMotion, monitoring via the AU-Manager is included in every project, and consumption is recorded per zone. A report of this kind helps beyond the energy balance, for example in ESG reporting, financing, acquisitions and due diligence. If you want the data analysed on an ongoing basis, see maintenance and energy management.
What role do DIN EN 15193-1 and DIN V 18599-4 play?
Both standards calculate a demand, not a measured consumption. DIN EN 15193-1 describes the energy requirement for lighting as the LENI indicator (Lighting Energy Numeric Indicator) in kWh per square metre and year. Occupancy and daylight control enter through factors, the occupancy dependency factor F_O and the daylight dependency factor F_D.
DIN V 18599-4 is the established German energy balance standard for lighting. The GModG already refers to DIN/TS 18599-11:2025-10 in Section 56(3), and from 1 January 2027 DIN/TS 18599:2025-10 also replaces DIN V 18599:2018-09 in the energy balance. Other rules, such as the BEG EM funding guideline, still cite DIN V. These methods are useful for planning and comparison. Only measurement shows the savings actually achieved.
Frequently asked questions
Can I apply the LBNL values to my building?
As a guide, yes; as a project value, only with caution. The LBNL figures are averages across many studies and buildings. In halls and warehouses with low occupancy, project values are often much higher, while in permanently fully occupied areas they can be lower.
Why can't I add the savings shares together?
Because each strategy only acts on the remaining consumption. The correct formula is 1 − (1 − a) × (1 − b) × …. Adding overstates the result, by 16 percentage points in the worked example.
How long should the baseline period be?
Long enough to reflect typical operation, covering all usual shifts and weekdays. The measurement afterwards should be equally long and, if possible, fall in a comparable season. Differing operating hours are normalised.
Does the standby consumption really have to be deducted?
Yes. Sensors and wireless nodes draw power around the clock, even when the light is off. The share is small, but it belongs in an honest balance.
Does a standards-based calculation replace measurement?
No. DIN EN 15193-1 and DIN V 18599-4, or their successor editions, calculate a demand under standard assumptions. The savings actually achieved require measured data before and after the retrofit.
Last reviewed: 18 November 2026. General information, not legal advice. The legal text is authoritative.