TY - GEN

T1 - Road Safety

T2 - 7th International Conference on System Reliability and Safety, ICSRS 2023

AU - Maschke, Jan Enno

AU - Putter, Roman

AU - Schoenawa, Stefan

AU - Gaas, Thorsten

AU - Leschke, Andre

AU - Lachmayer, Roland

N1 - Funding Information: ACKNOWLEDGMENT This research project is funded and supported by Volkswagen AG. The authors possess a valid data use license for the GIDAS (221231_GIDAS2022, 230131_GIDAS_PCM_5.0_2022_2) and DESTATIS 50% dataset (©Statistisches Bundesamt, Wiesbaden). The results, opinions and conclusions expressed in this publication are those of the authors and do not necessarily represent the views of Volkswagen AG.

PY - 2023

Y1 - 2023

N2 - In this paper, the authors examine car-to-car accidents for similarities of the three accident data sources German In-Depth Accident Study (GIDAS), GIDAS-Pre-Crash-Matrix (PCM) and the German Federal Statistical Office (DESTATIS). The data from the Federal Statistical Office are compared with the GIDAS data in a descriptive analysis. Among other variables, accident type, accident locations and injury severity distributions are investigated. In the process, systematic deviations between GIDAS and DESTATIS are presented. In addition, a previously published methodology for the extrapolation of GIDAS accidents to the national statistics is applied on car-to-car accidents and the weighting factors for 36 different sections are calculated. When comparing GIDAS and GIDAS-PCM, statistical tests are presented and used to check distributions of accident types and collision speeds for similarities. One of the results of the analysis is the dissimilarity of collision velocities between the GIDAS database and the GIDAS-PCM simulation database in car-to-car accidents. The presented analysis method can be part of a robust simulative effectiveness assessment to evaluate the validity of the simulation results in relation to the overall considered accident dataset.

AB - In this paper, the authors examine car-to-car accidents for similarities of the three accident data sources German In-Depth Accident Study (GIDAS), GIDAS-Pre-Crash-Matrix (PCM) and the German Federal Statistical Office (DESTATIS). The data from the Federal Statistical Office are compared with the GIDAS data in a descriptive analysis. Among other variables, accident type, accident locations and injury severity distributions are investigated. In the process, systematic deviations between GIDAS and DESTATIS are presented. In addition, a previously published methodology for the extrapolation of GIDAS accidents to the national statistics is applied on car-to-car accidents and the weighting factors for 36 different sections are calculated. When comparing GIDAS and GIDAS-PCM, statistical tests are presented and used to check distributions of accident types and collision speeds for similarities. One of the results of the analysis is the dissimilarity of collision velocities between the GIDAS database and the GIDAS-PCM simulation database in car-to-car accidents. The presented analysis method can be part of a robust simulative effectiveness assessment to evaluate the validity of the simulation results in relation to the overall considered accident dataset.

KW - car-to-car accidents

KW - GIDAS

KW - road safety

KW - similarity analysis

KW - virtual testing

UR - http://www.scopus.com/inward/record.url?scp=85183460221&partnerID=8YFLogxK

U2 - 10.1109/ICSRS59833.2023.10381120

DO - 10.1109/ICSRS59833.2023.10381120

M3 - Conference contribution

AN - SCOPUS:85183460221

T3 - 2023 7th International Conference on System Reliability and Safety, ICSRS 2023

SP - 229

EP - 236

BT - Proceedings 2023 7th International Conference on System Reliability and Safety, ICSRS 2023

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 22 November 2023 through 24 November 2023

ER -