ICAO USOAP CMA dataset issues 2022-2025
Description
This dataset includes International Civil Aviation Organization (ICAO) Universal Safety Oversight Audit Programme (USOAP) Continuous Monitoring Approach (CMA) public online data covering Effective implementation (EI) of ICAO Standards and Recommended Practices (SARPs) in relation to Accident Investigation (AIG). Because the USOAP database (see https://www.icao.int/safety/CMAForum/Pages/USOAP-Results.aspx) varies continuously, it is not possible to view historical data unless it has been downloaded and copied on a particular date and time in relation to one or more of ICAO's approximately 190 Member States. This also applies to the global average EI compliance for AIG. Initially, relevant ICAO Member State data was downloaded on 3 September 2022 to provide context for broader research in relation to those ICAO Member States that were members of the International Transportation Safety Association (ITSA). When an additional ITSA Accident Investigation Authority (AIA) had its first ICAO USOAP AIG audit, the USOAP online results for it and the other participants were reviewed on 15 April 2023. Some major discrepancies were found between the two dates without any change to the most recent (last) audit date underpinning the data. Ongoing liaison with ICAO and provision of the problematic data to them did not lead to an explanation for most of the issues. Additional ITSA research participants also had problematic data when the online database was accessed on 29 July 2024, 28 December 2024 and 19 March 2025. Further data has been added for 1 May 2025 and 12 May 2025. A draft manuscript with detailed analysis of the issues did not lead to a substantive response from ICAO. Additional ICAO subscription API datasets with USOAP data recorded conflicting results for the same last Mission audit year and Member States, compared with each other and with the accessed online data. After almost 2 years of raising these concerns, a manuscript has been completed that addresses the issues and their likely significance which may be much greater than for the 12 sampled ITSA AIAs. While the USOAP data is tabulated and discussed in this manuscript, it is not possible for other researchers to access the historical data unless they are separately made available by ICAO through restricted internal sources (if feasible). The files provided in the dataset provide this information. Because the issues were unexpected, the files include rough handwritten comments on the hard copies downloaded at the time.