For Data Scientists & ML Engineers · Cargo Ground Transportation
Cargo Ground Transportation Data for Data Scientists
Cargo ground transportation data for data scientists starts with the CFS 2022 PUMS file and FAF5 origin-destination flows.
financial time series api for backtesting · alternative data for quantitative research · where to get training data for cargo ground transportation models
API, files, or your warehouse. Daily, weekly, or hourly.
What cargo ground transportation data can data scientists actually train on?
This slice takes the 9 trucking and freight records that clear Datadory's relevance bar for data scientists and ML engineers — 7 of them score maximum or high relevance for modeling work, and none fall below quality 6. Quality runs hot for the domain: an 8.4-of-10 slice average beats the 7.81 catalog-wide mean, with 7 of 9 records scoring 8 or higher versus 62.8% across all 1,744. Browse the cargo-ground-transportation data hub for the full industry view, or see all data-scientists resources for sibling industries.
The 9 best trucking datasets for machine learning
Ordered by relevance to modeling work, then Datadory quality score. The one-line verdict under each name is written for pipeline builders.
1. Commodity Flow Survey (CFS) 2022 — Train freight-demand and lane-choice models on the January 2026 PUMS file of 37.6M+ shipment records.
2. Freight Analysis Framework Version 5 (FAF5) — Model freight flows to 2050 from full FAF5 downloads of origin-destination tonnage, value and ton-miles. Distribution spans CSV, Microsoft Access, XLSX, SHP and PDF; annual release; quality 10.
3. USDA Agricultural Truck Rate Report (AgTransport) — Forecast reefer rates from 206,201 weekly rows pulled through the Socrata SODA API. JSON, CSV and XLSX responses, weekly refresh, quality 9 — the only rate series here that pipelines programmatically.
4. Eurostat Road Freight Transport Statistics Explained — Download road_go_ta_tott annual totals 1999-2025 as the backbone of European freight models. Exports arrive as SDMX-CSV, JSON, TSV, SDMX-XML and XLSX on a quarterly cadence; quality 9.
5. UK Road Freight: Domestic and International Statistics — Analyse CSRGT goods-lifted series and RFS01-RFS03 tables back to 2012 for UK freight models.
6. Data.gov Freight Catalog — Discover DCAT-documented freight datasets and pull them into pipelines via the JSON search API.
7. EU Open Data Portal - Road Freight Transport — Pull Eurostat road_go_* bulk SDMX distributions documented in DCAT-AP into ELT jobs. Daily refresh across JSON, DCAT-AP RDF, SDMX-CSV, TSV, SDMX-XML and HTML; quality 8.
8. FHWA Freight Performance Measurement Program — Explore NPMRDS-derived truck travel-time reliability measures for congestion modelling. Access runs through a request form, delivering HTML, PDF, XLSX and interactive Tableau dashboards monthly; quality 7.
9. ATA Economics and Industry Data Hub — Headline tonnage releases are press-format; the full monthly index requires purchase, limiting pipeline use.
How do you combine these sources into one modeling panel?
The slice composes around geography and time. Start from Commodity Flow Survey (CFS) 2022 shipment microdata for demand and lane-choice targets, then add Freight Analysis Framework Version 5 (FAF5) zone-pair flows — tonnage, value and ton-miles projected to 2050 — as the structural prior your model beats. USDA Agricultural Truck Rate Report (AgTransport) supplies the weekly price dimension neither flow source carries, making it the natural target for reefer-rate forecasting. For European coverage, pair Eurostat's road_go_ta_tott country totals with the UK's CSRGT goods-lifted series back to 2012, and let FHWA Freight Performance Measurement Program NPMRDS reliability measures stand in for congestion covariates.
Straight answers
Does truck freight volume work as alternative data for quantitative research?
It can. CFS 2022 PUMS microdata decomposes 37.6M+ shipments by origin, destination and commodity, FAF5 adds zone-pair tonnage, value and ton-miles projected to 2050, and AgTransport tracks weekly reefer rates. Two caveats: both flow sources update annually, and FHWA reliability measures require a request form rather than open download.
Where can I get training data for freight demand forecasting models?
Combine grains: CFS 2022 PUMS for shipment-level demand, FAF5 zone-pair flows with 2050 projections as the baseline to beat, and AgTransport weekly rates as the price feature. For Europe, Eurostat's road_go_ta_tott totals run 1999-2025 and the UK CSRGT series reaches back to 2012.
Rows before rollout
Sample rows from any shelf entry — the field dictionary and coverage notes ride along. If the shelf misses what you need, say so; sourcing requests are half our job.
Talk to us