Analysis on Causative Factors and Evolution Paths of Blast Furnace Gas Leak Accident

Although the interest in metallurgical accident investigation of blast furnace gas (BFG) leakage has increased to explore engineering failures, more effort is needed address individual and organizational causative factors clear determine weak links for improving safety management prevention achieve green manufacturing. This study aims examine evolution paths BFG by introducing a combined method, 24 model Bayesian network (BN), based on 50 cases fire, explosion suffocation accidents caused leakage. A BN was established identification 25 model. Results showed that eight nodes, including A1 (unsafe operation), A2 behavior), A4 condition), B1 (valve failure), B2 (improper X4 (use violates regulations), X5 (water not cut off before shutdown reduction) X6 (incomplete steam purging), were sensitive than others, posterior probability nodes A1, A2, A3 command), A4, B1, B2, B4 emergency B5 behaviors site) compared prior probability. Three main causal chains obtained which indicate control unsafe operations (A1) related (B2) valve (B1) are suggested be improved. Another important factor intrinsic conditions. Considering results, key points 3E strategy about suggested. provides useful insights understand their relative influence accidents, will support management.