Language selection

Dataset from January 1983

With a view to advancing transportation safety, the TSB is publishing data from its Rail Occurrence Database System (RODS) on reportable accidents and incidents (which together are called occurrences) for use by industry and the public. The TSB gathers these data in the course of its investigations and uses them to analyze safety deficiencies and identify risks in the Canadian transportation system.

Accidents and incidents are reported in accordance with the Transportation Safety Board Regulations that were in effect at the time of the occurrence.

Accident and incident data prior to 1990 are provided on an as-is basis, as they were not collected by the TSB.

Content of the dataset

Some data collected by the TSB and recorded in RODS will be excluded from the public database to protect third party, personal, and privileged information, in accordance with the Canadian Transportation Accident Investigation and Safety Board Act, the Privacy Act and the Access to Information Act.

The data in text fields are provided in English and in French, while free-text fields contain data in the language in which they were entered. Some of the data in this dataset are from sources not subject to the Official Languages Act and are therefore available only in the language in which they were provided.

Data dictionary

The data are described in the data dictionary for this dataset, which is available in CSV [107 kb].

Release schedule

The RODS datasets are is released on or soon after the 15th of each month, and contains data from January 1983 to the last day of the month preceding its release.


The dataset is for information purposes only. The user of the data assumes all responsibilities for the data’s use and for verifying their completeness and accuracy for both critical and non-critical uses and applications.

The data in RODS have been collected by the TSB from third-party sources believed to be reliable. As many occurrences are not formally investigated by the TSB (see the Occurrence Classification Policy), information pertaining to some occurrences may not have been validated. Consequently, the dataset is provided on an as-is basis, and the TSB does not warrant its quality, accuracy, reliability, or completeness.

In no event will the TSB be liable for any direct, special, indirect, consequential or other damages, however caused, arising out of or in connection with the use of any data contained in the dataset. If you find any errors, please report them to

The dataset

The dataset comprises 5 tables in CSV format: