NHTSA Pe Recall
Will a newly opened NHTSA preliminary evaluation (PE) into a vehicle defect result in a linked safety recall campaign within 24 months?
Versioned research data. Read the suitability requirements before interpreting scores.
Suitability and evaluation
Evaluation setting: retrospective research. Read the prediction setting and limitations below. Public validation is development data; downloadable holdout labels must stay outside the evaluated agent workspace.
At a glance
| Question | Will a newly opened NHTSA preliminary evaluation (PE) into a vehicle defect result in a linked safety recall campaign within 24 months? |
|---|---|
| Field | National Highway Traffic Safety Administration (NHTSA) / binary-classification |
| Rows | 1,153 |
| Documented splits | 3 |
| This version | v-76d064ad7e164c41 |
| Released | 2026-11-09 |
| Validation ROC AUC | 0.6195033955857386 |
| Validation Brier | 0.25882025409720283 |
| Artifact | 7119c52b6fe5171677a06aebffca01b29a9960e1d0534755b8126137bbcf82ef |
Get this dataset
How to load it
from datasets import load_dataset
ds = load_dataset('earino/factory-nhtsa-pe-recall', revision='v-76d064ad7e164c41')
The version links and checksums identify this artifact. Exclude the labelled holdout from the agent's workspace when running an evaluation.
What is inside 
| Split | Rows | Positive labels |
|---|---|---|
| eval | 307 | — |
| holdout | 88 | — |
| train | 758 | — |
Files and checksums
| Download | Bytes | SHA-256 |
|---|---|---|
| LICENSE-EVIDENCE.json | 675 | ce4e625ff789cb84be0402e7a8aa885defd9099c7417dc6fecd33cbc2ea7459d |
| LICENSE.md | 866 | 4c4b6f7c8022255c051abf7db10cd6d13066aff62fd190e2cb7c2c282a3707d9 |
| README.md | 4,233 | 93a9b83cba91b0a3b4fe56f0a3c82ac3ecb56064d5f4fa4cd6e91a2910274c02 |
| SOURCES.json | 1,389 | d2c9fc9b1d11613d4d0620e5b8d6952fa19f1c2d5a17dc314f2dd1ce2cddb0f4 |
| evaluation.json | 3,972 | ed8fcb37b383323a347701eb232df52f46bf2bfba5fdc1159b3c64c89015d3b2 |
| extract__meta.json | 1,260 | c068d9fdb21af3d22b0bce6ae815d57a3e5788afed90c942202e905180ca0f5b |
| extract__private__holdout.csv | 8,107 | 93ca7bfbf095099c642d34471eda0be742b7f928eed74d080be6e5429ffc804d |
| extract__public__eval.csv | 30,634 | fdf29ddc574057c2d14efe13c4f62dc67a151cb339452f1888467d40e96860ca |
| extract__public__train.csv | 71,583 | 02390d2b80d997e73b1e1e258e8c2f72d52df06fc7c31ca0d60ce916e2f00970 |
| extract__quality.json | 17,759 | e0f9b8fbed4591387fff37f3a397df4783fc3c347bbde700f263b282cdd38479 |
| get_dataset.py | 3,680 | fad89222d50f9d9709225c34909be74c591df409c08b8b5951eba9d3cadfdd48 |
| qualification.json | 18,901 | 77a52242a6ce43fdc44b125b536e8afd53182223836b20da848be34ea7f1dc38 |
| reproduction__LICENSE-CC0-1.0.txt | 7,048 | a2010f343487d3f7618affe54f789f5487602331c0a8d03f49e9a7c547cf0499 |
| reproduction__LICENSE-MIT.txt | 1,124 | 56b932ff064e5da113ae10175c8546325f64f26e1ece71683ffcbf8b2304fe1c |
| reproduction__artifact_scrub.py | 11,369 | 8c663253e691a7b8751f9c853b40a3752a833360caafb1ea8e8f534e64af4b68 |
| reproduction__build.py | 15,122 | d662fc58fa051edd80cf641a9a76933bca875ac4b5e0dadef76a386394b26948 |
| reproduction__construction.json | 17,474 | b51ddfbe87d13a30aabf7a515b7bf1c27e1130cc65a3d55108a04613b19cd003 |
| reproduction__dataset_package.py | 7,927 | c2364defe58dcc99322be59a1bb04894ce00e823ce37afeb37bc31012f744327 |
| reproduction__dataset_run.py | 12,381 | 8085fae14d6e93b852e692e18c1ef72113a3601f48f78901b6d1ae6db278de57 |
| reproduction__get_dataset.py | 3,680 | fad89222d50f9d9709225c34909be74c591df409c08b8b5951eba9d3cadfdd48 |
| reproduction__leakage_battery.py | 17,379 | adbf7ee0aca0012b78cc1f1e4aeb968a23717cae30f66a02e080215e75f830e3 |
| reproduction__provenance.json | 573 | 768bebe8c4b2a024625465fb8453c587e61691707251168963ed7f1851a9b476 |
| reproduction__qualify_dataset.py | 66,067 | bff6d43d0904182e2612c32bab702e41addbf389904a236c209e78be39e37a6b |
| reproduction__recovery_recipes.py | 17,609 | f64c542fcda232169c95ace17421b45275194b934938c0ea4c9b68e6f844ad0e |
| reproduction__requirements.txt | 160 | 5c72f3268174d20364624f76680e189c408a05dba2e7b8cfe12d077b3865e7f0 |
| reproduction__source.json | 10,751 | c0896d7594b232d2b61be922e6b223491fcaf6ee497892faa80c4ab299b5be46 |
| reproduction__source_io.py | 5,607 | 0c60eed4968cd342148efbaac7e4fa98f00708d8e6c14c0929928687b9af1ea3 |
| reproduction__train.py | 2,500 | a3c6bcf13735bc85c52129ded68f839090dffdc266ebc3810dc61b2e6ea5e7e8 |
| reproduction__validate.py | 1,277 | b597e7f84fed614e64b4a86fbecbd6ec0145916a24d0d221a796586976418e7b |
| reproduction__validate.sh | 482 | 3f06ca48f11c2e05d331b4d8394284925de660c1d4b086d2f8243959e83b7e6d |
| source-checks.json | 6,154 | 9bbc899c1a888b3be261299c2140eca0f7bc8073c1cb6951fc90314cc0ea61f1 |
| source-snapshot.tar.gz | 23,689,483 | c00bb26d6f46694d6e396bb44a4cb1bcb80bffcdc6d53f02575ce6e1751035f4 |
| verify_dataset.py | 1,276 | f4a4bb7f565cdbd075c520621f881876b491eaed91a5e6e70afb31e4753f9834 |
Release notes · 2026-11-09
Will a newly opened NHTSA preliminary evaluation (PE) into a vehicle defect result in a linked safety recall campaign within 24 months?
Population: PE investigations opened 1995-01-01 through 2019-12-31 with a parseable ODATE, non-empty MAKE and MODEL on at least one make/model line, and YEAR from 1900 through 2019; multiple lines sharing an action number are collapsed to one row per action number using the lexicographically first valid line and a model count across lines; excludes YEAR 9999 equipment, tire, and child-restraint records and rows without a parseable ODATE.
Target: 1 if the PE record's CAMPNO is non-empty and the linked recall's RCDATE is after ODATE and on or before ODATE + 730 days; 0 otherwise. CAMPNO, CDATE, status, and recall records are label-only and excluded from features.
Prediction time: The PE open date (ODATE), before investigation findings, closure, or a recall campaign number are known.
Splits: Chronological by ODATE: train 1995-01-01 to 2007-12-31; eval 2008-01-01 to 2014-12-31; holdout 2015-01-01 to 2019-12-31. No embargo; each row is a distinct PE and no feature uses another row's outcome. All 24-month label windows close by 2021-12-31, before the 2026 snapshot.
Outcome horizon: 24 months (730 days) from ODATE.
Inputs: make: Vehicle make from the deterministic representative line of the PE (INV.txt field 2 MAKE).; model: Vehicle model or model list from the deterministic representative line of the PE (INV.txt field 3 MODEL).; model_year: Model year of the representative vehicle line under investigation (INV.txt field 4 YEAR).; component: Component or system description from the deterministic representative line (INV.txt field 5 COMPNAME).; manufacturer: Manufacturer from the deterministic representative line of the PE (INV.txt field 6 MFR_NAME).; model_count: Number of distinct model names across the PE make/model lines, a proxy for the breadth of the PE.; vehicle_age: Open year minus the representative model year, a proxy for vehicle age at the probe.
Recorded structural checks and baselines
PASSED Gate 1.6.1.
Existing unmodified runner baseline on public validation only; no holdout scoring
These historical checks do not establish availability at each prediction time or protection against every shortcut. Scores below describe the recorded protocol.
Limits
Current-vintage retrospective snapshot: the 2026 ODI file may reflect later corrections and does not prove the fields were operationally available in real time during 1995-2019.
CAMPNO attribution is imperfect: one recall may be linked to multiple PEs or models, and a PE may be followed by an EA; using PE only reduces but does not eliminate shared-event dependence.
Recall timing uses RCDATE from the recall file; if CAMPNO is present but unmatched or RCDATE is missing, the row is labeled 0, which can understate escalation.
Territorial limit: Public Domain U.S. Government status rests on 17 U.S.C. § 105 and does not establish worldwide public domain; manufacturer names and trademarks remain third-party.
No production or sales exposure; recall propensity is not normalized by fleet size.
YEAR=9999 is used to exclude equipment, tire, and child-restraint records, but some vehicle PEs with unknown model years are dropped.
Multiple make/model lines sharing one NHTSA action number are collapsed to one row per PE; the deterministic representative line discards within-PE variation in make, model, model year, component, and manufacturer.
Recall timing uses RCDATE from the recall file; if CAMPNO is present but unmatched or RCDATE is missing, the row is labeled 0 under the frozen rule, which can understate escalation.
Territorial limit: Public Domain U.S. Government status rests on 17 U.S.C. 105 and does not establish worldwide public domain; manufacturer names and trademarks remain third-party.
YEAR 9999 is used to exclude equipment, tire, and child-restraint records, but some vehicle PEs with unknown model years are dropped.
MODEL delimiter interpretation is assumed; if the source uses a different list separator, model_count may undercount or overcount.
No full-row availability ledger exists, so training_label_timing is unverified and feature availability is documented-field evidence plus explicit assumptions.
No uncertainty interval; observed differences do not establish universal headroom
Source, licences, and citation
National Highway Traffic Safety Administration (NHTSA)
National Highway Traffic Safety Administration
Source: Public Domain U.S. Government. Compilation: CC0-1.0. Code and documentation: MIT.
Our additional compilation rights only; upstream terms remain in force.