Construction Machinery & Heavy Transportation Equipment Data: Production, Trade, Fleets and Resale · Head-to-head
Eurostat Road Freight Transport Statistics vs NHTSA vPIC Vehicle Product Information Catalog
Which construction machinery & heavy transportation equipment data: production, trade, fleets and resale data fits your job: Eurostat road freight transport statistics, or NHTSA vPIC vehicle product information catalog. API, files, or your warehouse. Daily, weekly, or hourly.
Eurostat road freight transport statistics (road_go_ta_tg)
NHTSA vPIC vehicle product information catalog (heavy truck specs)
Where the fields line up
No shared field names. These two answer different questions.
| Field | Eurostat road freight transport statistics | NHTSA vPIC vehicle product information catalog |
|---|---|---|
freq | Time frequency of the observation; A (annual) throughout this table. | not in this set |
tra_type | Type of transport: TOTAL, OWN (own account - the shipper hauling with its own fleet), HIRE (hire or reward - paid third-party haulage) and NSP (not specified). | not in this set |
nst07 | Standard goods classification for transport statistics, NST 2007: twenty groups GT01-GT20 plus TOTAL and UNK. | not in this set |
unit | Unit of measure: THS_T (thousand tonnes) or MIO_TKM (million tonne-kilometres). | not in this set |
geo | Reporting country, with Eurostat-computed aggregates such as EU27_2020 riding the same dimension as sovereign states. | not in this set |
time | Reference year of the observation, 2008 through 2025. | not in this set |
OBS_VALUE | Freight volume or tonne-kilometre value for the observation, in the row's unit. | not in this set |
OBS_FLAG | Observation status flag marking estimate, provisional or break-in-series states. | not in this set |
Make | not in this set | documented |
Model | not in this set | documented |
ModelYear | not in this set | documented |
VehicleType | not in this set | documented |
What each contains
Pick by fit, not by loyalty.
| Eurostat road freight transport statistics | NHTSA vPIC vehicle product information catalog | |
|---|---|---|
| Documented fields | 8 | 14 |
| Field types | Categorical dimensions plus one numeric measure: freq, tra_type, nst07, unit, geo, time and an observation flag around OBS_VALUE | Identifiers and engineering attributes: Make, Model, ModelYear, VehicleType, BodyClass, GVWR, DriveType, EngineModel, EngineCylinders, DisplacementL, FuelTypePrimary, BrakeSystemType, Manufacturer, plant location |
| Signature fields | tra_type splitting TOTAL / OWN / HIRE / NSP; nst07 goods groups GT01-GT20; unit switching THS_T and MIO_TKM; OBS_VALUE carrying the measurement | GVWR band "Class 8: 33,001 lb and above"; DriveType 6x4; EngineModel Detroit Diesel DD15 at 14.8 L; BodyClass Truck-Tractor |
| Category-specific extension | OBS_FLAG preserves data quality - estimate, provisional, break - on every published cell | A complete decode widens to 154 returned variables per single VIN |
| Shared concepts | Heavy vehicles only as subject matter: one dataset counts their work, the other lists their build | No economics anywhere: neither carries price, transaction values or shipment-level revenue |
| Join keys | None shared: Eurostat keys on geo × year × goods group × transport type; vPIC keys on VIN and make-model-year | Bridging runs through a documented fleet-composition assumption, not a column match |
| Overlap verdict | Vocabulary-level overlap, record-level complement: one resolves to a flow, the other to a machine. | — |
What each does better
Eurostat road freight data
It measures the market's actual output. Tonnes give volume; tonne-kilometres give work done weighted by distance - a short urban hop and a 900 km line-haul look identical until the second metric lands. Any analysis of haulage demand, corridor loading or logistics intensity needs that second number, and it exists here for every country-year.
The own-account versus hire-or-reward split is the structural cut. Own account is a manufacturer moving its own goods with its own fleet; hire or reward is the for-hire carrier industry. Twenty NST 2007 goods groups cross that channel split, so Austrian hire-or-reward carriage of group GT01 reads 1,913 million tonne-km as far back as 2008 - private fleets and professional carriers never blur into one number.
Series depth makes trend claims possible. Observations run 2008 through 2025, gathered through vehicle- and time-related sample surveys conducted by national authorities under Regulation (EU) No 70/2012, with precision requirements set by Commission Regulation (EC) No 642/2004. Germany's published haulage moves from 270,325 million tonne-km in 2022 to 293,345 in 2024 - comparable readings, same definitions, year after year.
Observation flags keep the numbers honest. Every value carries a status flag marking estimates, provisional figures and series breaks, so a model can discount exactly the cells that deserve discounting.
NHTSA vPIC
It resolves all the way down to one vehicle. A full decode returns 154 variables for a single VIN - fourteen documented in our dictionary, from Make and Model through GVWR, DriveType, EngineModel, DisplacementL, FuelTypePrimary and BrakeSystemType to the legal Manufacturer entity and assembly-plant location. No freight statistic can tell you which specific tractor pulled the load; this record can.
Weight class and configuration are first-class fields. The GVWR banding - "Class 8: 33,001 lb and above" on the sampled Cascadia - plus a 6x4 driveline and body classes such as Truck-Tractor hand equipment analysts, insurers and valuation teams the exact spec sheet a marketplace listing usually paraphrases.
Identity infrastructure surrounds the vehicles. Beyond decodes, the catalog enumerates makes (12,340 at verification), manufacturers filterable by type, models per make and year, World Manufacturer Identifier codes on a global basis, and Canadian vehicle specifications alongside the US market core. VINs decode back to pre-1981 model years, which turns an aging fleet register into analyzable history.
It is the reference layer other records lean on. Auction listings, registration files and telematics exports all describe vehicles; vPIC is where the canonical spelling of a make, the true model year and the weight class live.
The verdict
Verdict: sample both, pick by fit - they measure different layers of the same industry.
Take Eurostat if your question moves at market level. Sizing European road freight demand, tracking modal shift, separating for-hire carriers from private fleets, forecasting tonnage by goods group, benchmarking one country's haulage intensity against another's. Accept the trade: countries, goods groups and channels, never vehicles.
Take vPIC if your question moves at equipment level. Identifying a specific tractor from its VIN, validating a listing's claimed specification, building fleet-composition panels by weight class and driveline, anchoring valuation or insurance work to factory-recorded truth. Accept the trade: attributes only - nothing here says how much anything moved.
If the noun in your question is a tonne, choose Eurostat. If it is a truck, choose vPIC.
Sample both, pick by fit. See Eurostat road freight transport statistics · See NHTSA vPIC vehicle product information catalog
Or take both in one feed
Yes - Eurostat supplies the "how much", vPIC supplies the "with what".
A first-pass pipeline: take a Eurostat country-year cell - German hire-or-reward tonne-kilometres by goods group - then use vPIC decodes across a fleet sample to estimate how much of that work runs Class 8 6x4 tractors versus rigid rigs, and allocate the flow over equipment cohorts for capacity planning or emissions modeling.
Three cautions from the pair. First, the bridge is yours: nothing joins a country-goods cell to a VIN, so document the fleet-mix assumptions doing the connecting. Second, respect the units asymmetry - thousand tonnes and million tonne-kilometres on one side, categorical specs with litres and cylinders on the other - meaning any per-vehicle figure you derive is an allocation, not an observation. Third, date-stamp separately: annual reference years on one side, a continuously maintained catalog whose decodes reach back before 1981 on the other.
Or take both in one feed: Datadory normalizes each to its documented field dictionary, attaches sample rows for validation, and ships them alongside the rest of the construction machinery & heavy transportation equipment catalog, delivered daily, weekly, or hourly - your call.
API, files, or your warehouse. Daily, weekly, or hourly.
Fair questions
Is Eurostat road freight data better than NHTSA vPIC?
They answer different questions. Eurostat road_go_ta_tg wins on flows: about 71,000 annual observations of EU haulage in tonnes and tonne-kilometres across twenty NST 2007 goods groups and own-account versus hire-or-reward channels, scoring 9/10. NHTSA vPIC wins on vehicles: 154 variables per VIN decode covering make, model, GVWR class, driveline, engine and plant of manufacture, also scoring 9/10.
Do Eurostat and NHTSA vPIC cover the same vehicles?
Only conceptually. Eurostat counts freight moved by heavy goods vehicles registered in EU Member States, EFTA and candidate countries; vPIC catalogs US market vehicles with Canadian specifications adjacent and a global WMI registry behind them. The records share no geography, and neither shares the other's units - tonnes on one side, vehicle attributes on the other.
Which dataset has history, Eurostat road_go_ta_tg or vPIC?
Different kinds of history. Eurostat publishes dated annual series running 2008 through 2025 under stable definitions, made for year-over-year trend work. vPIC holds the current catalog state, but its decodes reach back to pre-1981 model years, so an aging vehicle register still resolves. Through Datadory either can be delivered daily, weekly, or hourly on your schedule.
Can I combine EU freight tonnage with US vehicle specification data?
Yes, with an explicit bridge. A common pattern allocates Eurostat tonne-kilometre flows across equipment cohorts characterized from vPIC decodes - weight class, driveline, fuel. No shared join key exists between a country-goods cell and a VIN, so the fleet-mix assumptions doing the connecting need documenting, and the result is an allocation rather than an observation.
Can I get both datasets from Datadory?
Yes - sample both and pick by fit, or take both in one feed. Datadory normalizes each to its documented field dictionary, attaches sample rows for validation, and ships them alongside the rest of the construction machinery & heavy transportation equipment catalog, delivered daily, weekly, or hourly - your call.