Smart Street Lighting in Turkey: Inventory, EN 13201 and M&V
Blog
Smart Municipality

Smart Street Lighting in Turkey: Inventory, EN 13201 and M&V

smart street lightingsmart lighting systemsmunicipal LED conversionstreet lighting inventoryEN 13201lighting asset managementtelemanagement systemGIS luminaire inventorysmart municipality

Street lighting is the most visible infrastructure a municipality has to keep running, and one of the quietest lines in its budget. In the summer of 2026 that quiet ended in Turkey: a single decision tripled the municipal share of the lighting bill. This guide explains what smart street lighting systems actually are, which standard governs them, and how a municipality can measure rather than merely believe a savings claim.

Why now: the municipal share tripled

In Turkey the cost of general lighting is covered from the Ministry of Energy and Natural Resources budget together with municipalities' and provincial special administrations' share of general budget tax revenues. The municipality does not pay the invoice directly — it is deducted from its share.

Presidential Decision no. 11573, published in the Official Gazette on 9 August 2026 and based on provisional article 6 of Electricity Market Law no. 6446, reset those deduction rates:

  • Metropolitan municipalities and municipalities in the adjacent area: 20% → 60%
  • Other municipalities: 10% → 30%
  • Provincial special administrations outside municipal boundaries: 10% → 60%

The decision applies from 1 January 2026. The practical consequence: every kilowatt-hour saved on lighting now returns directly to the municipality's own budget. Payback periods calculated before this decision are now materially shorter.

For scale: according to Ministry figures, Turkey consumed 5,024 GWh of electricity for general lighting in 2023; covering that bill took 23.6 billion lira from the Ministry budget and 5.2 billion lira from municipal and provincial budgets. For the number of luminaires the primary source is TEDAŞ's 2024 annual report: a national total of 9,177,558 luminaires and lamps as of 2023 (8,868,190 held by distribution companies plus 309,368 by third parties and industrial zones). The “roughly 10 million” figure repeated in the press has no traceable source.

What is a smart lighting system?

A smart street lighting system lets each luminaire be monitored and controlled individually from central software. It has three layers:

  • Light source: usually an LED luminaire and its driver
  • Control node and network: a node on each luminaire and the network linking them to the centre
  • Central management system (CMS): where dimming profiles are defined, faults arrive and consumption is reported

One confusion is worth clearing up: LED conversion and smart lighting are not the same thing. LED conversion replaces the luminaire and delivers most of the saving on its own — the Turkish Ministry cites roughly 50% when replacing high pressure sodium lamps. Smart lighting adds a control and measurement layer on top, and its real contribution is threefold:

  • Dimming: lowering light levels when traffic thins, without violating the lighting class
  • Fault management: learning about a dark luminaire from the system rather than from a citizen complaint
  • Measurability: per-luminaire consumption reporting, so the saving is data rather than a claim

Progress under the Ministry-coordinated conversion stood at 276,000 LED luminaires and 415 million lira of annual saving as stated by the Minister of Energy and Natural Resources to the Parliamentary Plan and Budget Committee on 7 November 2025.

More important than the running total is that the conversion is not a target but a binding programme. Annex 1 of the Ministry's procedures and principles on LED conversion sets a year-by-year luminaire quota for every distribution company: 500,000 in 2025 and 1,000,000 a year from 2026 to 2029, 4,500,000 luminaires in total. The consequence for a municipality is this: converting the road network to LED is already legally programmed and funded from the distribution company's investment plan. The municipality's leverage is not paying for the swap itself but deciding, in the lighting commission, which street enters the programme in which year — which requires holding a prioritised inventory.

Note: the ministerial-level target of “1 million luminaires by 2028 and 1.2 million by 2030” differs from Annex 1's 4.5 million quota by a factor of about 3.75, and no source reconciling the two could be found. This page treats the annex to the procedures in force as the primary source.

Lighting class: EN 13201

The technical backbone of road lighting is the EN 13201 series (adopted in Turkey as TS EN 13201). Whether a luminaire is "adequate" is not decided by its wattage but by whether it satisfies the requirements of its assigned lighting class. The series has five parts:

  • CEN/TR 13201-1 — selection of lighting classes. Note: this is a technical report, not a standard; it guides, it does not require.
  • EN 13201-2 — performance requirements (the numeric criteria per class)
  • EN 13201-3 — calculation of performance
  • EN 13201-4 — methods of measuring performance
  • EN 13201-5 — energy performance indicators

Parts -2 to -5 are dated 2015, and -2:2015 superseded the 2003 edition. That distinction has a practical consequence you will see in the warning below.

What M, C and P classes actually measure

  • M series — motorised traffic roads. The measured quantity is road surface luminance (cd/m²), what the driver's eye actually sees. Surface type and observation angle enter the calculation.
  • C seriesconflict areas such as junctions, roundabouts and queuing areas. The luminance convention does not hold in those geometries, so the criterion reverts to illuminance (lux).
  • P series — pedestrian and cycle areas. The measured quantity is horizontal illuminance (lux).

The standard also defines three series most Turkish sources omit: HS (hemispherical illuminance), SC (semi-cylindrical, for areas where facial recognition matters) and EV (vertical plane illuminance).

The criteria tables

The values below are transcribed from EN 13201-2:2015. These are the numbers to check a bid against:

M classes — EN 13201-2:2015, Table 1. Dry road surface unless noted. Uow applies to wet conditions and is a national option.
ClassL̄ (cd/m²) minUo minUl minUow minfTI (%) maxREI min
M12.000.400.700.15100.35
M21.500.400.700.15100.35
M31.000.400.600.15150.30
M40.750.400.600.15150.30
M50.500.350.400.15150.30
M60.300.350.400.15200.30

A common error: many pages still print SR (surround ratio) and TI. Those are parameters of the 2003 edition. In the 2015 edition SR was replaced by REI (edge illuminance ratio) and TI by fTI.

If a specification or compliance report shows “SR”, it is citing a withdrawn edition. REI values are marked tentative in the standard and apply only where the adjacent area has no lighting requirement of its own.

C and P classes — EN 13201-2:2015, Tables 2 and 3. Every C class requires Uo ≥ 0.40. P7 exists but its performance is not determined.
ClassĒ (lx) minUo / Emin
C050Uo 0.40
C130Uo 0.40
C220.0Uo 0.40
C315.0Uo 0.40
C410.0Uo 0.40
C57.50Uo 0.40
P115.0Emin 3.00 lx
P210.0Emin 2.00 lx
P37.50Emin 1.50 lx
P45.00Emin 1.00 lx
P53.00Emin 0.60 lx
P62.00Emin 0.40 lx

Over-lighting is non-compliance, not a safety margin

  • In P classes the achieved maintained average illuminance may not exceed 1.5 times the class minimum (EN 13201-2:2015, Table 3 footnote).
  • TEDAŞ's LED road lighting design procedures cap average road luminance at 1.2 times the class minimum.
  • EN 13201-5 requires the calculated level not to exceed the requirement of the next class up.

So lighting a road that requires M4 to an effective M2 is not “safer” — it is non-compliant, and it bills the municipality for years.

How a class is assigned: 6 − ΣVW

CEN/TR 13201-1 assigns a weighting value (VW) to a set of parameters and derives the class number as 6 − Σ(weighting values). Some parameters are time-dependent: traffic volume, traffic composition, ambient luminance and weather all change through the night.

This is exactly where adaptive dimming gets its legitimacy. When traffic falls after midnight, ΣVW changes, the required class drops, and less light becomes compliant. A reduction of 2 units in ΣVW corresponds roughly to halving the required level.

Hence the correct formulation: a dimming profile is specified as “which class is met in each period”, not as “dimmed to 50%”. These are not the same thing, and a municipality that writes the second cannot later measure non-compliance.

The Turkish overlay: TEDAŞ design procedures

Alongside adopting the EN 13201-2 tables verbatim, TEDAŞ sets design rules specific to Turkey. They bind a project submitted for approval:

  • Maintenance factor is fixed at 0.80 (LLMF 0.90 × LSF 1.00 × LMF 0.89, CIE 154).
  • Absent measurement, the road surface class is taken as R3.
  • A junction takes one class above the highest approach road (M1→C0, M2→C1 … M5→C4).
  • Photometric data is required from an accredited laboratory in EULUMDAT (.ldt) format.
  • Out of scope: motorways and access-controlled roads, and the surroundings of astronomical observatories and protected natural areas.

That last exclusion is not a footnote: lighting in protected areas is a distinct design problem because of its ecological effect — a topic we touch on in protected areas and environmental mapping.

Components and protocols

Writing open standards rather than brand names into a specification is what keeps a municipality from being locked to one supplier:

  • TALQ: the interoperability interface between central management software and outdoor device networks — how luminaires from different manufacturers are driven from one CMS.
  • DALI-2 / D4i: the digital interface between driver and control node. D4i exposes energy and fault data from the driver, which is what per-luminaire consumption reporting is built on.
  • Zhaga Book 18: the physical standard for the sensor/node connector on the luminaire — critical if you install LED now and add control later.
  • NEMA / ANSI C136.41: the seven-pin control socket, the common alternative for carrying the dimming signal.
  • Communication: NB-IoT (carrier network, wide coverage), LoRaWAN (municipality-owned network), RF mesh (dense luminaire runs).
  • CLO (constant light output): LEDs lose flux over time; CLO holds output steady over life instead of over-lighting on day one.
  • Astronomical calendar: switching driven by computed sunrise/sunset for the location rather than a photocell.

The skipped first step: the asset inventory

This is the most important section of this guide, and the one absent from supplier catalogues.

Ask a municipality how many luminaires it has, at what wattage, on which pole, fed from which circuit — and a clear answer rarely follows. Turkish Court of Accounts audits have reported findings such as general-lighting assets not being transferred to the distribution company and parks being tariffed under the wrong subscriber group. These share one root cause: no asset inventory.

Without it, none of the following can be done properly: tender quantities, lighting-class verification, baseline consumption, or checking the distribution company's billed consumption against the wattage actually installed.

How the inventory is produced

  • Mobile mapping: vehicle-mounted LiDAR and panoramic cameras scan the road network once; pole positions, mounting heights and luminaire types are extracted from the point cloud and imagery.
  • Aerial LiDAR and photogrammetry: bulk production of pole positions and road geometry over wide areas.
  • GIS integration: relating the inventory to zoning, parcel, road and feeder-circuit layers in one database.

The output is not a spreadsheet but a spatial asset catalogue: coordinate, type, wattage, pole height and associated road section for every luminaire. Produced once, it serves the tender, the EN 13201 calculation and maintenance management simultaneously.

The “80% savings” claim and the band you can actually defend

Collect the figures circulating on Turkish supplier pages and you get: 30–50%, 50–70%, “up to 70%”, 60%, 70–80%, 25–60% — with three mutually inconsistent numbers on a single page in at least one case.

The defect is not the size of the percentages. None of them states what was compared with what, the burning hours or the tariff, and — decisively — none normalises to equal lighting-class compliance.

An example makes it concrete: you can save 70% on a road that requires class M3 by effectively lighting it to M5. That is not a saving, it is non-compliance. The number is true and meaningless.

The one sourced Turkish band: 42–60% at equal luminance

There is a peer-reviewed study done under Turkish conditions with an explicit method: Onaygil, Güler and Yurtseven, “LED Conversion of Existing Roads”, 12th National Lighting Congress (2019). For classes M1–M4, using TEDAŞ luminaire classes, a maintenance factor of 0.80, the ≤1.2× luminance cap and 11.5 hours per night:

42–60% power saving at equal luminance compliance, replacing high-pressure sodium (20,000 h) with LED (60,000 h).

Use this as the realistic reference band. Note that the tariff inputs date from 2019, so the financial payback needs recalculating.

Separate savings into three layers

High headline percentages usually come from stacking three distinct sources on top of an over-lit baseline. They should be separated:

  1. LED conversion at equal class — the bulk of the saving, and the most reliable part.
  2. Calendar / astronomical dimming — fixed-hour or sunrise-linked profiles.
  3. Sensor / traffic-adaptive dimming.

Layers (2) and (3) do not deliver as promised. The U.S. Department of Energy tested 21 0–10 V LED street light drivers in December 2023:

  • Relative power draw at the same control voltage varied by about 53 percentage points between products.
  • Nine of 19 products did not conform to ANSI C137.1-2022.
  • A midnight dimming strategy targeting 18% delivered 12% on average.
  • Two CLO strategies produced a 2% increase and zero saving against expectations of 5% and 10%.

The specification clause that follows is simple: dimming savings are accepted when measured in the field, not when promised.

The regulation's own payback rule: 5 years and 4,200 hours

The Ministry's procedures on LED conversion put a ceiling on this discussion. Under article 5/7 the payback period of a conversion investment may not exceed 5 years. Annex 2 gives the calculation:

Payback = (investment × 1000) / [(HPS system W − LED W) × annual hours × electricity price]

Annual operating hours are fixed at 4,200 h/year. Cross-check: 11.5 h/night × 365 = 4,197.5, which is what the regulation's own lump-sum consumption formula implies. Article 5/6 of the same document also requires luminaires ten years old or more to be replaced with LED within five years.

The EN 13201-5 indicators are not a savings measurement

The standard defines two indicators that are widely misread:

  • Power density indicator Dp = P / Σ(Ei·Ai), in W·lx⁻¹·m⁻².
  • Annual energy consumption indicator DE, in Wh·m⁻², including dimming profiles.

System power P includes not only the lamps but the control gear, control units, switches and photocells. Where CLO is implemented in the driver, average system power is used for Dp.

The critical distinction: Dp and DE are indicators calculated at design stage — they are not measurement results. The standard also states plainly that installations with different geometry or different lighting requirements cannot be compared directly.

They exist to compare competing designs for the same road. Proving a realised saving needs the next section.

Proving the saving: measurement and verification

The governing principle of the International Performance Measurement and Verification Protocol (IPMVP) surprises most municipalities: savings cannot be measured, only inferred. A saving is energy that was not consumed, and no meter records absence. It is the difference between adjusted baseline energy and reporting-period energy.

The practical consequence: the baseline period, the measurement boundary and the adjustment rules must be written before the conversion. Written afterwards, they become a negotiation.

Building the baseline for street lighting

  1. Verified luminaire count — from the inventory.
  2. Measured luminaire power — not the nameplate; driver losses included.
  3. Operating hours — derived from the astronomical calendar.

Where no meter exists, the regulation assigns consumption by formula: M = Pt × 11.5 × GS (connected load × nightly hours × days). The existence of that formula reveals something important: a large share of Turkey's lighting bill rests on declaration, not measurement. That is precisely where the financial value of an inventory sits.

Which IPMVP option

IPMVP options applied to street lighting.
OptionWhat it doesFor street lighting
ARetrofit isolation, key parameter measurementThe honest default. Measured luminaire power at each dimming step, hours stipulated from the astronomical calendar
BRetrofit isolation, all parameter measurementWhere a panel meter or D4i per-luminaire energy reporting exists
CWhole facility, meter/bill analysisUsually unavailable — lighting is largely unmetered and billed lump-sum
DCalibrated simulationNew roads, where no baseline data exists at all

That option C is usually unavailable also explains why supplier claims are so rarely falsified: there is no meter series to compare against.

Acceptance measurement: a simulation is not a statement of compliance

The output of DIALux or similar is a simulation. Compliance is demonstrated by field measurement to EN 13201-4 with a calibrated ILMD, following TEDAŞ's procedures for LED lighting measurement. A compliance statement should also record which r-table and which maintenance factor were used.

In public contracts, M&V is a contractual condition

Where the work runs under an energy performance contract, measurement and verification is already mandatory: carried out every year of the contract under TS ISO 50006 and IPMVP or TS ISO 50015, certified by a qualified Measurement and Verification Professional. Bids come from energy efficiency consultancy (EVD) companies, the winner is selected on Net Present Value, and at least two valid bids are required.

What is actually in the field in Turkey

Every figure here comes from a primary source; what could not be sourced is marked as such.

  • Luminaires and lamps on the network: 9,177,558 (TEDAŞ 2024 annual report, 2023 data).
  • General lighting consumption: 5,024 GWh (2023).
  • Conversion to date: 276,000 LED luminaires, 415 million lira annual saving (November 2025).
  • Binding quota: 500,000 in 2025 and 1,000,000 a year for 2026–2029 — 4,500,000 luminaires in total, allocated per distribution company.

Comparing distribution companies' press announcements against those quotas produces a striking overlap: KCETAŞ announced 11,200 for 2025 against a quota of 11,296; OEDAŞ announced 20,397 against 19,887; YEDAŞ announced 35,000 against 35,663. These announcements are effectively compliance reporting, not voluntary efficiency initiatives.

What we could not find: no source publishing measured(as opposed to targeted) savings for an LED-plus-central-management deployment in Turkey. The published national percentages derive from Annex 2's fixed 4,200 hours and nameplate wattage arithmetic — they are definitional, not measured.

This is an observation rather than a criticism, and it shows why a municipality establishing its own M&V framework gains so much.

A six-step roadmap for municipalities

  1. Build the spatial inventory — position, type, wattage, mounting height and feeder circuit in one GIS layer.
  2. Classify roads under EN 13201-1 — an M, C or P class for each section.
  3. Compute baseline energy indicators — document current DP and DE.
  4. Write the specification around the class — bind the luminaire to a class and indicator, not a wattage; state socket and control interface as open standards.
  5. Design the control layer — astronomical calendar, CLO and dimming profiles that never violate the class.
  6. Measure and verify — recompute the indicators after commissioning and report the saving as a measurement.

Do not go to tender before step four. A specification written without an inventory and without classes commits the municipality to work whose outcome cannot be measured.

Where Verigo fits

To be clear: Verigo does not sell luminaires, drivers or management software. Verigo produces the spatial data layer underneath — which is exactly the first three of the six steps above.

  • Inventory production via mobile mapping, aerial LiDAR and photogrammetry
  • GIS integration with zoning, parcel, road and utility layers, connected to the existing city information system through open standards (OGC, WMS/WFS, GeoJSON)
  • Road geometry and classification base data — the road width, pole spacing and mounting height an EN 13201 calculation needs
  • 3D city model and digital twin in CityGML / 3D Tiles, managing the lighting layer alongside other infrastructure

In other words: the supplier installs the luminaire; Verigo makes it measurable — where it is, what it lights, and whether it is really saving anything.

Without a lighting inventory, a smart lighting project is not a measurable project.

For scope, the data model, the field method and the clauses to put in a tender document, see our lighting asset inventory and GIS integration service; the inventory template is downloadable there as a CSV.

Sources

Every numeric claim on this page rests on the sources below. Market figures whose provenance could not be established were deliberately excluded and are referred to only as criticised examples.

  • EN 13201-2:2015 — Road lighting, Part 2: Performance requirements (Tables 1, 2, 3)
  • EN 13201-3:2015 / -4:2015 / -5:2016 — calculation, field measurement and energy performance indicators
  • CEN/TR 13201-1 — selection of lighting classes (technical report)
  • TEDAŞ — procedures for LED road lighting design, and procedures for LED lighting measurement
  • Turkish General Lighting Regulation — art. 6, art. 8/3 and Annex 1 (lump-sum consumption formula)
  • Ministry of Energy — procedures for LED conversion under general lighting — art. 5/6, art. 5/7, Annex 1 (per-company quotas), Annex 2 (payback formula)
  • Presidential Decision no. 11573, Official Gazette 9 August 2026 — lighting cost deduction rates
  • Electricity Market Law no. 6446, provisional article 6
  • TEDAŞ 2024 annual report — Table 3, pole and luminaire counts (2023)
  • Onaygil, Güler, Yurtseven — “LED Conversion of Existing Roads”, 12th National Lighting Congress, 2019
  • U.S. DOE — Impacts of 0–10 V LED Street Lights, December 2023
  • EVO — IPMVP Core Concepts (2022); TS ISO 50006, TS ISO 50015
  • Communiqué on Energy Performance Contracting in the Public Sector, Official Gazette 15 April 2021, no. 31455
  • Energy Efficiency Law no. 5627, additional article 1

Regulation checked: 12 August 2026. Turkish lighting regulation changed more than once during 2026; re-verify the primary source before relying on it for a decision.

Frequently asked questions

What is a smart street lighting system?

A smart street lighting system lets each luminaire be monitored and controlled individually from central management software. On top of the LED luminaire there is a control node, a communication network and a central management system (CMS). It can apply dimming profiles, report faults on its own and meter energy consumption per luminaire.

Is LED conversion the same as smart lighting?

No. LED conversion replaces the luminaire and delivers most of the saving on its own. Smart lighting adds a control and measurement layer on top; its real value is midnight dimming, shorter fault-detection time and consumption that can be documented. A municipality can go LED first and add control later, as long as the luminaire socket is specified for it.

Who pays for street lighting in Turkey?

General lighting cost is covered from the Ministry of Energy and Natural Resources budget together with municipalities' and provincial special administrations' share of general budget tax revenues. Presidential Decision no. 11573 of 9 August 2026 raised the deduction from the municipal share from 20% to 60% for metropolitan municipalities and from 10% to 30% for others.

How is a lighting class assigned?

Under EN 13201-1 a class is selected from traffic volume, speed limit, user type and conflict conditions. Motorised traffic roads are assessed in the M series by road surface luminance (cd/m²), pedestrian areas in the P series by horizontal illuminance (lux), and conflict areas such as junctions in the C series. Luminaire selection cannot precede this step.

Does smart lighting really save 80%?

That figure is usually quoted without a source. The Turkish Ministry of Energy cites roughly 50% when replacing high pressure sodium lamps with LED. Dimming profiles add to that, but the real figure depends on existing luminaire wattage, burning hours and the selected lighting class. Savings should be computed with the power density indicator (DP) and annual energy consumption indicator (DE) of EN 13201-5 and confirmed by field measurement.

Why is a lighting inventory necessary?

Most municipalities do not know the exact position, wattage, type and feeder circuit of their poles and luminaires. Without an inventory the tender quantity cannot be costed correctly, the lighting class cannot be verified and savings cannot be measured. The inventory can be produced at centimetre accuracy with mobile mapping, LiDAR and photogrammetry, then loaded into GIS.

Which standards and protocols apply?

EN 13201 for lighting performance, TALQ for interoperability between central management software and field devices, DALI-2/D4i for driver control, Zhaga Book 18 for the sensor-node connector and NEMA/ANSI C136.41 for the seven-pin socket. NB-IoT, LoRaWAN and RF mesh are the common communication choices.

Where should a municipality start?

With the inventory. Map the poles and luminaires spatially, classify the roads under EN 13201-1, then compute the energy indicators of the current situation. Only then can a realistic technical specification be written and the outcome measured.