Datasetpublic

CyberMax-tools/haulroll-new-motor-carriers

Haulroll: new US trucking companies (FMCSA new carrier registrations) Need it fresh, filtered or via API? This free file is a snapshot (new FMCSA motor carriers at the last refresh), last updated 2026-09-25. Get an email when this dataset updates: free, double opt-in, unsubscribe any time. Every motor carrier, broker and freight forwarder that received a new USDOT number in the last 90 days, from the FMCSA Company Census File, one row per company, company-level fields… See the full description on the dataset page: https://huggingface.co/datasets/CyberMax-tools/haulroll-new-motor-carriers.

sourceHugging Faceotherupdated 5h agoView on Hugging Face

0likes

Dataset Card

Haulroll: new US trucking companies (FMCSA new carrier registrations)

Need it fresh, filtered or via API? This free file is a snapshot (new FMCSA motor carriers at the last refresh), last updated 2026-09-25. - [Get an email when this dataset updates](https://data.cybermax-tools.workers.dev/notify?ds=haulroll-new-motor-carriers&s=hf-haulroll-new-motor-carriers): free, double opt-in, unsubscribe any time.

Every motor carrier, broker and freight forwarder that received a new USDOT number in the last 90 days, from the FMCSA Company Census File, one row per company, company-level fields only.

This build (see meta.json): 43,085 new registrations from 2026-06-27 to 2026-09-23 (25,343 interstate, 17,742 intrastate; 13,454 with an MC/MX/FF docket; 3,219 hazmat). Top states: TX 5,261, CA 4,848, FL 3,678, GA 2,042, NY 1,891.

Who uses a list like this: freight brokers onboarding new capacity, truck insurers and factoring firms, fuel-card and ELD vendors, compliance-service firms, and analysts tracking how many new trucking companies start each week by state.

Configs

Config	Rows	What
`new_carriers` (default)	one per USDOT number	registration date, status, operation (interstate / intrastate), entity type, city, state, ZIP-5, county FIPS, power units, trucks, buses, drivers, CDL drivers, hazmat flag, fleet-size code, first docket (MC/MX/FF) and its status, cargo types, a link to the FMCSA SAFER snapshot
`daily_counts`	per day × state × operation	how many new registrations

python

from datasets import load_dataset
ds = load_dataset("CyberMax-tools/haulroll-new-motor-carriers", "new_carriers", split="train")
nj_interstate = ds.filter(lambda r: r["state"] == "NJ" and r["operation"] == "Interstate")

Or query without downloading (Hugging Face dataset viewer API, no key): https://datasets-server.huggingface.co/filter?dataset=CyberMax-tools/haulroll-new-motor-carriers&config=new_carriers&split=train&where="state"='TX'&orderby="registered" DESC

Privacy: what is NOT in this dataset

—No phone, fax, cell phone or email address, no company officer names, no street addresses, no D&B numbers. They are never downloaded (the query selects company-level columns only).
—Legal and DBA names that look like an individual's name (no business word such as LLC, Inc, Corp, Trucking, Transport, Logistics) are withheld (null). The USDOT number and the SAFER link still identify the registration on FMCSA's own site.
—If you represent a listed company and want a correction, correct it at FMCSA (MCS-150 update); the next build picks it up.

Source, licence and limits

—Source: FMCSA Company Census File, data.transportation.gov dataset az4n-8mr2. registered = the census add_date (USDOT registration date).
—Licence: a work of the US Government (FMCSA/USDOT); no US copyright (17 U.S.C. §105). CyberMax's selection and code are offered under the same open terms. Not affiliated with FMCSA or USDOT.
—Limits: fields are as the registrant reported them on the MCS-150; the census lags FMCSA systems by a day or more; a new USDOT number does not mean operating authority was granted. Not a safety rating, credit check, insurance verification or endorsement of any carrier. Verify on SAFER before relying on a record.

Refresh

Rebuilt from the census file by products/haulroll/dataset/build.py (a few seconds, no API key); same file names each build, history kept in the dataset's git.

Also from CyberMax

Quickstart notebook

Load this dataset with pandas and see real outputs in the quickstart notebook.