NL Traffic Congestion Index Monitor (NDW) 🇳🇱🚗

Live traffic congestion index per Dutch province, computed from the official NDW (Nationale Databank Wegverkeersgegevens) open-data travel-time feed. No configuration required, no API key, no scraping — this actor fetches NDW's real-time DATEX II travel-time publication, compares actual measured travel time against each measurement point's free-flow reference value, and aggregates the result into one congestion score per province (plus separate national-highway and municipal-road summaries).

Who is this for?

• Logistics planners & fleet operators — quick daily congestion signal per province, no need to parse raw NDW feeds yourself
• Real-estate & site-selection analysts — accessibility context for a region, alongside other location data
• Media/data journalists — a ready-made national congestion index for reporting
• AI agents — structured JSON output, ideal as a tool for agentic workflows (MCP-ready)

When should an AI agent use this?

• "What's the current traffic congestion level in Zuid-Holland?"
• "Which Dutch province has the worst traffic right now?"
• "Give me a national traffic congestion snapshot."
• "Alert me when congestion in Noord-Brabant crosses a certain threshold."

What this Actor does

• Fetches NDW's live traveltime.xml.gz feed (updated roughly every minute, DATEX II/XML format)
• For every measurement point, compares the actual/reconstituted travel time against its static free-flow reference value
• Classifies each measurement point to a province using the road authority's own site-ID naming convention (see Methodology below) — measurement points on national highways or municipal roads that can't be attributed to a specific province are reported as separate summary groups instead of being silently dropped
• Aggregates to one record per group: number of measurement points, average travel-time ratio, and a derived 0-100+ congestion score
• Optional threshold alerting: flag groups whose congestion score crosses a value you set

Methodology (please read before use)

NDW's own measurement-site IDs for provincial roads follow the pattern P<province-code><number>_... (e.g. PNB05_BRE_Keizerstraat_N01). This is the road authority's own convention, not something this actor invents. It was empirically verified on 2026-07-15 by resolving 11 measurement points (spanning 10 of the 12 province codes) to their real coordinates via NDW's own location table and reverse-geocoding each one with the official PDOK Locatieserver — all 11 matched the expected province.

• Known edge case: a handful of specific road segments run right along a province border (e.g. a bridge/dam connecting two provinces). Such a segment is grouped under the province that administers it, which can occasionally differ from strict geographic containment.
• Measurement points on the national highway network (RWS...) generally don't carry a province code at all — these are reported as "Rijkswegennet (niet provinciaal ingedeeld)", a separate group.
• Measurement points on municipal roads (G...) are reported as "Gemeentelijk wegennet".
• Anything that doesn't match a known prefix is reported as "Overig/onbekend wegbeheerder" — nothing is ever silently excluded from the totals.

This actor does not perform per-point geocoding at runtime — that would require tens of thousands of lookups per run and isn't practical. The province classification is a static, empirically spot-checked mapping from the road authority's own ID prefix.

Input

Output

One record per group (province, or non-provincial category):

{
  "groep": "Zuid-Holland",
  "classificatieMethode": "provincie-wegbeheerder-prefix",
  "aantalMeetpunten": 2807,
  "gemiddeldeReistijdRatio": 1.253,
  "filedrukScore": 25,
  "hogeFiledrukAlert": true,
  "dagdeel": "ochtendspits",
  "publicatietijd": "2026-07-15T09:37:42.012Z",
  "trend": null,
  "bron": "NDW Open Data — traveltime.xml.gz (opendata.ndw.nu, DATEX II)"
}

• gemiddeldeReistijdRatio: average of (actual travel time / free-flow reference time) across the group's measurement points that have a reference value
• filedrukScore: round((ratio - 1) * 100), floored at 0 — 0 means at-or-below free-flow speed on average, 25 means travel times are running ~25% above free-flow, etc.
• classificatieMethode: "provincie-wegbeheerder-prefix" for province groups, "niet-provinciaal" for the national-highway/municipal-road/unknown groups
• hogeFiledrukAlert: null when alertDrempel isn't set, otherwise true/false
• dagdeel: "ochtendspits"/"avondspits"/"dal", derived from publicatietijd converted to Europe/Amsterdam local time (DST-aware). null if publicatietijd is unknown
• trend: null unless trackTrend is enabled; otherwise "oplopend"/"dalend"/"gelijk" compared to the previous run, or null on the very first run (no prior snapshot yet)

Trend between runs (trackTrend, optional, separately charged)

With trackTrend: true, the actor stores each group's filedrukScore in its key-value store after every run and compares against it on the next one. Each record then gets a trend field: "oplopend" (higher than last run), "dalend" (lower), or "gelijk" (same). The very first run (no stored snapshot yet, or a group that's new) always gets trend: null. Charged as trend-signaal, only for groups actually flagged "oplopend" or "dalend" — an unchanged group is never charged.

Top-N busiest measurement points (topNPerGroep, optional, separately charged)

With topNPerGroep set (e.g. 5), each group record gets a topMeetpunten array with the N individual measurement points that have the highest travel-time ratio within that group, sorted descending:

{
  "groep": "Zuid-Holland",
  "topMeetpunten": [
    { "siteId": "PZH04_A20_Gouda_L01", "reistijdRatio": 2.31 },
    { "siteId": "PZH04_A12_Zoetermeer_L02", "reistijdRatio": 2.08 }
  ]
}

Charged as meetpunt-detail, once per measurement point actually included (e.g. topNPerGroep: 5 on a run with 10 groups charges up to 50 meetpunt-detail events, on top of the 10 filedruk-groep-record events).

National summary (free extra dataset record)

Whenever there's at least one group record in the run, one extra record with

"type": "landelijke-samenvatting"

{
  "type": "landelijke-samenvatting",
  "aantalGroepen": 12,
  "gemiddeldeFiledrukScore": 18.4,
  "drukstProvincie": "Zuid-Holland",
  "drukstProvincieScore": 25,
  "aantalAlerts": 2,
  "dagdeel": "ochtendspits",
  "publicatietijd": "2026-07-15T09:37:42.012Z",
  "bron": "NDW Open Data — traveltime.xml.gz (opendata.ndw.nu, DATEX II)"
}

Use cases

Just the provinces, no national-highway/municipal noise:

{ "includeOngeclassificeerd": false }

Watch specific provinces with an alert:

{ "provincies": ["Noord-Holland", "Zuid-Holland", "Utrecht"], "alertDrempel": 30 }

Pricing

This Actor uses Apify's Pay-Per-Event (PPE) pricing model.

• Actor Start: $0.00005 (Apify default)
• filedruk-groep-record: $0.008 per group record delivered
• hoge-filedruk-alert: $0.02 extra, only when alertDrempel is set and a group's congestion score reaches it
• trend-signaal: $0.0125 extra, only with trackTrend enabled and a group actually flagged "oplopend" or "dalend"
• meetpunt-detail: $0.005 per measurement point, only with topNPerGroep set, per point actually included

Legal

• Data source: NDW (Nationale Databank Wegverkeersgegevens), opendata.ndw.nu — a joint open-data service of Rijkswaterstaat, provinces, and municipalities, explicitly published for reuse. No authentication required.
• Purely aggregated traffic-flow measurements on the road network — no license plates, no individual vehicle identification, no personal data.
• Province classification is a best-effort, empirically spot-checked heuristic based on the road authority's own ID convention (see Methodology above), not an officially published NDW mapping table.

FAQ

Q: How current is the data? A: NDW updates the travel-time feed roughly every minute. Each run reflects a live snapshot at the moment it runs — schedule the actor periodically (e.g. hourly) to build a time series.

Q: Why do some groups have far more measurement points than others? A: Coverage reflects NDW's actual sensor/camera/floating-car-data network, which is denser in the Randstad and along major national highways. This isn't adjustable.

Q: What if a province name I pass in provincies doesn't match anything? A: You'll get zero records for that run — province names must match exactly (e.g. "Noord-Holland", not "Noord Holland" or "NH").

Related Actors

• PDOK Locatieserver — used during this actor's own development to verify the province-classification methodology above.

Zoekwoorden: filedruk, verkeersdrukte, reistijdindex, NDW verkeersdata, files per provincie, verkeerscongestie Nederland, filedruk monitor, actuele filedruk.

Keywords

traffic congestion, filedruk, ndw, travel time index, netherlands traffic, road congestion, datex ii, verkeersdata

Changelog

0.3.0 - Agent-readiness: task-oriented description

• Rewrote the Store description/short description (task-oriented first sentence, explicit NL/EN synonyms "filedruk"/"verkeersdrukte"/"congestion", a decision-benefit sentence, and a plain-text price indication) as part of the portfolio-wide agent-readiness pass. No code, schema, or pricing change — .actor/actor.json and STOREINFO.md are kept identical.

0.2.0 - Trend detection & top-N busiest measurement points

• Added trackTrend — compares each group's filedrukScore against the previous run via a named key-value-store snapshot, flags trend as "oplopend"/"dalend"/"gelijk". New trend-signaal event ($0.0125), only charged for groups actually flagged rising or falling.
• Added topNPerGroep — includes the N individual measurement points with the highest travel-time ratio per group (topMeetpunten). New meetpunt-detail event ($0.005 per point), only charged per point actually included.

0.1.1

• Added dagdeel (ochtendspits/avondspits/dal) per record, derived from publicatietijd in NL-local time. No pricing change.
• Added a free landelijke-samenvatting dataset record per run (national average, busiest province, alert count). No pricing change.

0.1.0

• Initial release: live NDW travel-time feed aggregated into a congestion index per province, with national-highway/municipal-road summary groups and optional threshold alerting.