Comparative analysis of two catastrophic hazardous chemical accidents in China

Abstract

On March 21, 2019, a serious explosion of hazardous chemicals occurred in Tianjiayi Company located in Xiangshui Chemical Industry Park in Yancheng City, Jiangsu Province, China. Seventy‐eight people were killed and 716 were injured. It is the most severe accident since the globally recognized fire and explosion at a hazardous chemical warehouse in Tianjin Port, Tianjin, China in 2015. According to the preliminary analysis of the Yancheng accident and the review of previous studies about the Tianjin Port accident, there are similarities regarding the process of the two accidents. Therefore, in addition to reporting the investigation results of the Yancheng accident, an important aim of this work is to focus on the brief comparative analysis between it and the Tianjin Port accident. The comparisons were mainly conducted according to the basic information of the events, triggers and processes, individual factors, organizational management factors, external regulatory factors, and then the differences and similarities between the two accidents are summarized. Thus, useful lessons in the aspects of the storage and management of hazardous chemicals, safety training and accident learning, implementation of safety management system, government supervision, and so on, can be drawn to avoid similar mistakes within the industry.

Investigation and Analysis of a Hazardous Chemical Accident in the Process Industry: Triggers, Roots, and Lessons Learned

“…The external causes mainly involve factors from natural events, defective design, poor supervision from the government, deficiencies in laws or regulations, etc. [3,6], which generally contribute to accidents by influencing the internal ones.…”

“…When accidental events occur in the chemicals' production, storage, transportation, or other processes, they often lead to multiple domino effects and may spawn serious consequences of mass casualties, property losses, and environmental pollution [1]. This kind of disaster is very important in process safety and risk management [2], and it occurs in China's process industry very frequently, and even shows an increasing trend year-by-year [3]. In the last few years, different measures in aspects of the integration of chemical industrial parks, the monitor and control of operator behaviors, the revision of related laws and regulations, etc., have been taken by the Chinese government to improve the situation, and people carried out extensive research on the occurrence of hazardous chemical accidents [4][5][6].…”

A New Text-Mining–Bayesian Network Approach for Identifying Chemical Safety Risk Factors

“…In this way, the highly unstructured incident reports can be transformed into structured numerical-type data. The correlation coefficients between feature terms are then calculated using TDM, as shown in Equation (8).…”

“…corr(i, j) = pearson(i, j) * log 2 1 mindist(i, j) + 1 (8) where pearson(i, j) refers to the Pearson correlation coefficient of feature words i and j; mindist(i, j) indicates the shortest distance between these two words in all documents. If two feature terms appear in a document at the same time but are located far apart, then it is clear that the two feature terms are not related when expressing a hidden problem.…”

“…Zhou et al [6] established an integrity chemical safety assessment system and, based on expert scores, made specific guidance recommendations for improving chemical park safety management and enhancing the overall safety capability of the park. Researchers have used tools such as brainstorming, Delphi method, questionnaires, interviews, cause-and-effect analysis, literature study, and their combination to encapsulate the domain knowledge [7][8][9]. These traditional data collection methods usually require a certain number of experienced experts and consume extensive time and cost [10].…”