Accident analysis (8 articles)

An insight of World Health Organization (WHO) accident database by cluster analysis with self-organizing map (SOM)

Pal, C., Hirayama, S., Narahari, S., Jeyabharath, M., Prakash, G., & Kulothungan, V. (2018). Traffic injury prevention, 19(sup1), S15-S20.

This paper analyzes the types and causes of road accidents worldwide. Viscovery SOMine clustering shows that there are differences between low- and high-income countries, which should be addressed with different counter measures.

Comprehensive analysis of Indian road accident data to enrich road safety

Pal, C., Nobuhiko, T., Natarajasundaram, B., Manoharan, J., Sangolla, N., Kulothungan, V., & Padhy, S. (2018). FISITA World Automotive Congress 2018.

This paper analyzes data from the MORTH 2016 and ADAC, NATRIP databases to further understand the causes of road accidents in India. Viscovery SOMine is used to find unique profiles of factor combinations that lead to road accidents.

Identification of significant factors contributing to multi-attribute railway accidents dataset (MARA-D) using SOM data mining

Yu, G., Zheng, W., Wang, L., & Zhang, Z. (2018). In 2018 21st International Conference on Intelligent Transportation Systems (ITSC) (pp. 170-175). IEEE.

Data on 392 railway accidents in China is collected and analyzed. Viscovery SOMine is used to extract common key causes from the multi-attribute data.

Scientific analysis of road accidents in India by self-organizing map

Pal, C., Hirayama, S., Thierry, H., Vimalathithan, K., Rangari, N., & Jeyabharath, M. (2018). IRCOBI Asia 2018.

The types, severity and causes of Indian road accidents are the focus of this paper. Viscovery SOMine is used to cluster 1 779 sample cases and reveals several similarities between causes of different accident types.

Learning from accidents: human errors, preventive design and risk mitigation

Moura, R. (2017).

This dissertation introduces a systematic approach and data collection to analyze major high-technology accidents. Viscovery SOMine is used to cluster accidents according to underlying causes with special emphasis on deficiencies in the interaction of humans and technical systems.

Learning from accidents: interactions between human factors, technology and organisations as a central element to validate risk studies

Moura, R., Beer, M., Patelli, E., Lewis, J., & Knoll, F. (2017). Safety Science, 99, 196-214.

A risk management framework to prevent industrial accidents is developed. Viscovery SOMine is used on a data set of major industrial accidents to find common patterns and define safety measures.

Learning from accidents: analysis and representation of human errors in multi-attribute events

Moura, R., Beer, M., Patelli, E., Lewis, J., & Knoll, F. (2015, July). In Proceedings of the 12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12 (Vol. 15, p. 2015).

The analysis of major industrial accidents and the influence of human factors are the focus of this article. Viscovery SOMine is used to produce a map representing high-dimensional data about accidents, including technical data and human factors and analyze clusters of similar accident groups.

Detecting the impacts of socioeconomic factors on regional severity of work-related casualties in China

Zhou, L., Wei, J., & Zhao, D. (2014). Human and Ecological Risk Assessment: An International Journal, 20(6), 1469-1490.

Impacts of socioeconomic factors on the severity of work-related casualties are analyzed on casualty data from China. The Viscovery SOMine cluster model revealed that work-related casualties first increase with economic development, reach a peak at an intermediate stage and decline with further development. In addition, influences of industrial and employment structure, as well as education level, medical condition and insurance coverage are found.