Tidy Access to Brazilian Federal Highway Police ('PRF') Data

Download and read Brazilian Federal Highway Police ('PRF') open data on traffic accidents by person, by occurrence, and traffic violations < https://www.gov.br/prf/pt-br/acesso-a-informacao/dados-abertos/dados-abertos-da-prf>. Data are retrieved on demand as 'Parquet' files distributed via 'GitHub' releases and cached locally for reuse.

tidyprf

Leia em Português

tidyprf provides access, from R, to public road safety datasets made available by the Polícia Rodoviária Federal (PRF). These include data on accidents by person, accidents by occurrence, and traffic violations.

The package allows users to select the desired dataset and year, returning the data in tabular format. Files are distributed in Parquet format via GitHub Releases and stored in a local cache after the first download.

Quick example

Map fatal accidents across Brazilian states in 2024:

library(tidyprf)
library(geobr)
library(dplyr)
library(ggplot2)

fatal <- get_crashes(2024, severity = "fatal") |>
  count(uf, name = "acidentes")

read_state(year = 2020, showProgress = FALSE) |>
  left_join(fatal, by = c("abbrev_state" = "uf")) |>
  ggplot() +
  geom_sf(aes(fill = acidentes), color = "white", linewidth = 0.3) +
  scale_fill_distiller(palette = "Reds", direction = 1, name = "Accidents") +
  labs(title = "Fatal accidents by state (2024)") +
  theme_void(base_size = 13) +
  theme(plot.title = element_text(hjust = 0.5, face = "bold"))

Installation

Install the released version from CRAN:

install.packages("tidyprf")

Or the development version from GitHub:

# install.packages("remotes")
remotes::install_github("bonijoao/tidyprf")

Datasets

All three functions support filtering by:

• uf — state abbreviation(s), e.g. "SP", c("SP", "RJ")
• br — federal highway number(s), e.g. 101, 116
• severity — "fatal", "injured", or "no_victims" (accidents only)

Use info_accidents(), info_crashes(), or info_violations() to see all variable descriptions in English or Portuguese.

Cache

Parquet files are cached locally after first download:

prf_cache()               # show cached files and sizes
prf_cache_clear()         # delete all cached files
prf_years("accidents")    # available years and row counts per dataset

The data repository is updated weekly with new PRF releases. When a cached year has a newer version, it is downloaded again automatically; offline, the cached copy is used. To turn this check off, use options(tidyprf.check_updates = FALSE).

Data source

Raw data are published by the PRF on the Brazilian federal government open data portal. The consolidation pipeline that converts the raw CSV files into the Parquet files consumed by this package lives at bonijoao/tidyprf-dados.

License

MIT