alinutzal/CA-HVF2017

---
license: cc
language:
- en
tags:
- synthetic-data
- transportation
- vehicle-fleet
- california
- agent-based-models
- emissions
pretty_name: 'CA-HVF2017: California Household Vehicle Fleet Dataset'
size_categories:
- 10M<n<100M
---

[](https://creativecommons.org/licenses/by/4.0/)

# CA-HVF2017: California Household Vehicle Fleet Dataset (2017)

A statewide synthetic vehicle fleet dataset containing 13 million households and over ~25+ million vehicles across California. This dataset provides detailed household-level vehicle ownership information, including vehicle type, powertrain, vintage, and body type, generated using a Multiple Discrete Continuous Extreme Value (MDCEV) model and a geographically explicit synthetic population.

This dataset is designed to support transportation modeling, energy/emissions analysis, policy scenario evaluation, EV adoption studies, and agent-based simulations.

---

## Overview

Modern transportation models require realistic household vehicle fleets, but privacy constraints limit access to micro-level vehicle ownership data.
CA-HVF2017 fills this gap by providing a statewide, publicly available, synthetic, validated representation of vehicle ownership across all California census tracts.

The dataset includes:

- Household characteristics (demographics, size, income)
- Detailed vehicle fleets for each household
- Geographic attributes at the census-tract level
- Location-based accessibility and built environment indicators

All values are synthetic but statistically consistent with real-world data.

---

# Data Dictionary: Input

Below is the full set of household-level variables included in the CA-HVF2017 dataset.

---

## Household Demographics

| Variable | Description | Values |
|---------|-------------|--------|
| child | Household has at least one child | 0, 1 |
| HOUSEID | Household ID | numeric |
| HHSIZE | Household size | numeric |
| HHSIZE1 | Household size = 1 | 0, 1 |
| HHSIZE2 | Household size = 2 | 0, 1 |
| HHSIZE3 | Household size = 3 | 0, 1 |
| HHSIZE4 | Household size = 4 or more | 0, 1 |
| NUMADLT | Number of adults | numeric |
| NUMCHILD | Number of children | numeric |
| NUM_WORKERS | Number of workers in household | numeric |
| retired | Household has at least one retiree | 0, 1 |

---

## Race / Ethnicity Indicators

| Variable | Description | Values |
|---------|-------------|--------|
| hhwhite | Householder identifies as White | 0, 1 |
| hhasian | Householder identifies as Asian | 0, 1 |
| hhblack | Householder identifies as Black or African American | 0, 1 |
| hhothers | Householder identifies as another race (not White/Black/Asian) | 0, 1 |

---

## Income Categories

| Variable | Description | Values |
|---------|-------------|--------|
| income1 | Income < $25,000 | 0, 1 |
| income2 | $25,000 ≤ income < $50,000 | 0, 1 |
| income3 | $50,000 ≤ income < $75,000 | 0, 1 |
| income4 | $75,000 ≤ income < $100,000 | 0, 1 |
| income5 | Income ≥ $100,000 | 0, 1 |

---

## Life Cycle Categories

| Variable | Description | Values |
|---------|-------------|--------|
| LIF_CYC1 | 1 adult, no children | 0, 1 |
| LIF_CYC2 | 2+ adults, no children | 0, 1 |
| LIF_CYC3 | 1 adult + child age 0–5 | 0, 1 |
| LIF_CYC4 | 2+ adults + child age 0–5 | 0, 1 |
| LIF_CYC5 | 1 adult + child age 6–15 | 0, 1 |
| LIF_CYC6 | 2+ adults + child age 6–15 | 0, 1 |
| LIF_CYC7 | 1 adult + child age 16–21 | 0, 1 |
| LIF_CYC8 | 2+ adults + child age 16–21 | 0, 1 |
| LIF_CYC9 | Household has at least one senior (65+) | 0, 1 |
| LIF_CYC10 | Household has 2+ seniors (65+) | 0, 1 |

---

## Housing & Tenure

| Variable | Description | Values |
|---------|-------------|--------|
| hhown | Household owns home | 0, 1 |
| perrent | % rental housing in tract | numeric |
| perrent1 | Rental housing < 25% | 0, 1 |
| perrent2 | Rental housing 25–45% | 0, 1 |
| perrent3 | Rental housing > 45% | 0, 1 |

---

## Work Status

| Variable | Description | Values |
|---------|-------------|--------|
| work0 | No members employed | 0, 1 |
| work1 | 1 worker in household | 0, 1 |
| work2 | 2 workers in household | 0, 1 |
| work3 | 3+ workers in household | 0, 1 |

---

## Geographic Identifiers

| Variable | Description | Values |
|---------|-------------|--------|
| county_fips | County FIPS code | numeric |
| county_name | County name | character |
| state_fips | State FIPS code | numeric |
| state_name | State name | character |
| tractid | Census tract ID | numeric |

---

## Transit & Accessibility Variables

| Variable | Description | Values |
|---------|-------------|--------|
| emp_zscore | Standardized jobs reachable by 30-min transit | numeric |
| tractmean | Average number of jobs reachable from tract | numeric |
| tas_acres | Total acres accessible via 30-minute transit | numeric |
| tci | Transit Connectivity Index (0–100) | 0–100 |
| hi_tps | AllTransit Performance Score ≥ 8 | 0, 1 |
| transit_performance_score | Transit Performance Score (0–10) | 0–10 |

---

## Built Environment Variables

| Variable | Description | Values |
|---------|-------------|--------|
| job_density | Jobs per km² | numeric |
| pop_density | People per km² | numeric |
| res_density | Housing units per acre (unprotected) | numeric |
| pct_ag_land | % agricultural land | numeric |
| pct_water | % water area | numeric |
| urban_cbsa | Census tract is urban | 0, 1 |
| walkndx | Walkability index (0–20) | 0–20 |

---

## Log-Transformed Built Environment / Transit Indicators

| Variable | Description | Values |
|---------|-------------|--------|
| log_job_density | Log(job_density) | numeric |
| log_job_above8 | Log(job_density) > 8 | 0, 1 |
| log_job_below4 | Log(job_density) < 4 | 0, 1 |
| log_pop_density | Log(pop_density) | numeric |
| log_pop_above9 | Log(pop_density) > 9 | 0, 1 |
| log_pop_below3 | Log(pop_density) < 3 | 0, 1 |
| log_res_density | Log(res_density) | numeric |
| log_pct_agland | Log(pct_ag_land) | numeric |
| log_pct_water | Log(pct_water) | numeric |
| log_lastyear_zevpct | Log(previous-year ZEV share) | numeric |

---

## Zero-Emission Vehicle (ZEV) Exposure

| Variable | Description | Values |
|---------|-------------|--------|
| lastyear_zev_pct | Percentage of ZEVs in prior year | numeric |

# Data Dictionary: Output

## Household-Level Variables

| Variable name | Description | Value |
|---------------|-------------|--------|
| county_fips | County FIPS code | numeric |
| tractid | Census tract ID | numeric |
| HOUSEID | Household ID | numeric |
| nvehicles | Number of vehicle(s) owned by the household | numeric |

## Vehicle-Level Variables

| Variable name | Description | Value |
|-------------------|-----------------------------------------------|--------|
| county_fips | County FIPS code | numeric |
| tractid | Census tract ID | numeric |
| HOUSEID | Household ID | numeric |
| VEHID | Vehicle ID associated with the household | numeric |
| bodytype | The vehicle's body type | car, van, suv (Sport Utility Vehicle), pickup (Light-duty pick-up truck) |
| vintage_category | Vehicle age range | 0–5 years, 6–11 years, 12+ years |
| annual_mileage | The vehicle's annual mileage | numeric |
| pred_power | The vehicle's powertrain | ICE (Internal Combustion Engine), AEV (All-Electric Vehicle), PHEV (Plug-in Hybrid Electric Vehicle), Hybrid (hybrid vehicle) |
| modelyear | Vehicle model year (year manufactured) | numeric |

---

## Methodology Summary

### 1. Synthetic Population

Generated using PopulationSim, producing approximately 13 million California households with demographics matched to ACS distributions.

### 2. Multiple Discrete-Choice Vehicle Ownership Model

The dataset extends the MDCEV-based fleet composition model by Garikapati et al. (2014).
The model jointly predicts:

- Number of vehicles per household
- Vehicle category combinations
- Powertrain shares
- Vintage distributions

Predictors include:

- Income, household size, and workers
- Built environment metrics
- Accessibility indices
- Regional land-use patterns

### 3. Validation

The synthetic fleet is externally validated against:

- California DMV vehicle registration data
- County-level vintage and powertrain distributions
- Household vehicle count statistics

The dataset reproduces observed distributions with high fidelity.

---

## Geographic and Temporal Scope

- Region: California
- Spatial resolution: Census tract (GEOID)
- Households: ~13 million
- Vehicles: ~25+ million
- Base demographic year: 2017
- Fleet calibration year: 2017

---

## License

This dataset is licensed under the **Creative Commons Attribution 4.0 International (CC BY 4.0)** license.

You are free to:

- **Share** — copy and redistribute the material in any medium or format
- **Adapt** — remix, transform, and build upon the material for any purpose, even commercially

Under the following terms:

- **Attribution** — You must give appropriate credit, provide a link to the license, and indicate if changes were made.
You may do so in any reasonable manner, but not in a way that suggests the licensor endorses you or your use.

Full license text: https://creativecommons.org/licenses/by/4.0/

---

## How to Load the Dataset

### Python (pandas)

```python
import polars as pl

hh = pl.read_parquet("households.parquet")
veh = pl.read_parquet("vehicles.parquet")