No abstract

No abstract

No abstract

The Process Safety Management System (PSMS) of an industrial asset relies on multiple and independent barriers for preventing the occurrence of major accidents and/or mitigating their consequences on people, environment, asset and company reputation. It is, then, fundamental to assess the performance of the barriers with respect to the occurrence of Process Safety Events (PSEs), i.e. unplanned or uncontrolled events during which a Loss Of Primary Containment (LOPC) of any material, including non-toxic and non-flammable material, occurs. An essential aspect of PSMS is learning from incidents and taking corrective actions to prevent their recurrence. For this, a procedure for timely and consistently reporting and investigating PSEs is generally implemented. After the occurrence of a PSE, a report containing free-text and multiple-choice fields is filed to describe the PSE, its causes and consequences, and to provide a quantification of its the level of severity with reference to predefined Tier levels, as per API RP 754 guidelines. This work investigates the possibility of text-mining and structuring the knowledge on the performance of the PSMS from an electronic repository of PSE reports. The methodology developed falls within the framework of Natural Language Processing (NLP), combining Term Frequency Inverse Document Frequency (TFIDF) and Normalized Pointwise Mutual Information (NPMI) for the automatic extraction of keywords from the PSE reports. Then, a taxonomy is built to organize the vocabulary in a top-down structure of homogeneous categories, such that semantic and functional relations between and within them can be defined. Based on these relations, a Bayesian Network (BN) is developed for modeling the PSEs consequences. The proposed methodology is applied to a repository of real reports concerning the PSEs of hydrocarbon facilities of an Oil and Gas (O&G) company.

Although road safety has improved in the last decades, the rate of accidents with severe and fatal consequences is still exceeding the safety objectives (European Commission 2019; World Health Organization 2018).This work explores the possibility of using Natural Language Processing (NLP) techniques for the automatic extraction of knowledge from road accidents reports, with the objective of supporting the safety management of the road infrastructure system (Persia et al. 2016).To this aim, we consider databases of textual reports on road accidents, provided by the local public authorities. These reports contain the descriptions of the accidents and the results of the post-accident investigations. The aim is to analyze the reports by NLP to extract the features that most influence the accidents, for informing road safety management.For the analysis of the reports, we develop a method that combines Hierarchical Dirichlet Processes (HDPs) (Teh et al. 2006), Artificial Neural Networks (ANNs) and a feature selection technique based on the Sequential Forward Selection (SFS) strategy (Marcano-Cedeño et al. 2010). HDPs allow representing each report as a mixture of topics, i.e. distributions of words co-occurring in the reports. In practice, each report is transformed into a vector whose elements are the degrees of membership to each topic, i.e. a measure of the contribution of each topic to the description of the report. ANNs are then used to classify the reports, represented by the extracted vectors, into classes characterizing the severity of the accident consequences. Finally, the SFS technique is used for identifying those topics which most influence the reports classification. In this way, the factors causing the accidents and influencing its evolution are automatically extracted. The developed method is validated considering a database of real accident reports. ReferencesEuropean Commission. 2019. “EU Road Safety Policy Framework 2021-2030 - Next Steps towards ‘Vision Zero.’” Brussels,19.6.2019 SWD(2019) 283 final.Marcano-Cedeño, A, J Quintanilla-Domínguez, M G Cortina-Januchs, and D Andina. 2010. “Feature Selection Using Sequential Forward Selection and Classification Applying Artificial Metaplasticity Neural Network.” In IECON 2010 - 36th Annual Conference on IEEE Industrial Electronics Society, Glendale, AZ, 2845–50. https://doi.org/10.1109/IECON.2010.5675075.Persia, Luca, Davide Shingo, Flavia De Simone, Véronique Feypell, De La Beaumelle, George Yannis, Alexandra Laiou, et al. 2016. “Management of Road Infrastructure Safety.” Transportation Research Procedia 14: 3436–45. https://doi.org/10.1016/j.trpro.2016.05.303.Teh, Yee Whye, Michael I Jordan, Matthew J Beal, and David M Blei. 2006. “Hierarchical Dirichlet Processes.” Journal of the American Statistical Association, 1566–81. https://doi.org/10.1198/016214506000000302.World Health Organization. 2018. “Global Status Report on Road Safety.” Geneva: World Health Organization.