The accidental release of toxic gases leads to fire, explosion, and acute toxicity, and may result in severe problems for people and the environment. The risk analysis of hazardous chemicals using consequence modelling is essential to improve the process reliability and safety of the liquefied petroleum gas (LPG) terminal. The previous researchers focused on single-mode failure for risk assessment. No study exists on LPG plant multimode risk analysis and threat zone prediction using machine learning. This study aims to evaluate the fire and explosion hazard potential of one of Asia’s biggest LPG terminals in India. Areal locations of hazardous atmospheres (ALOHA) software simulations are used to generate threat zones for the worst scenarios. The same dataset is used to develop the artificial neural network (ANN) prediction model. The threats of flammable vapour cloud, thermal radiations from fire, and overpressure blast waves are estimated in two different weather conditions. A total of 14 LPG leak scenarios involving a 19 kg capacity cylinder, 21 tons capacity tank truck, 600 tons capacity mounded bullet, and 1350 tons capacity Horton sphere in the terminal are considered. Amongst all scenarios, the catastrophic rupture of the Horton sphere of 1350 MT capacity presented the most significant risk to life safety. Thermal flux of 37.5 kW/ m2 from flames will damage nearby structures and equipment and spread fire by the domino effect. A novel soft computing technique called a threat and risk analysis-based ANN model has been developed to predict threat zone distances for LPG leaks. Based on the significance of incidents in the LPG terminal, 160 attributes were collected for the ANN modelling. The developed ANN model predicted the threat zone distance with an accuracy of R2 value being 0.9958, and MSE being 202.9061 in testing. These results are evident in the reliability of the proposed framework for safety distance prediction. The LPG plant authorities can adopt this model to assess the safety distance from the hazardous chemical explosion based on the prior forecasted atmosphere conditions from the weather department.

Risk analysis and prediction is a primary monitoring strategy to identify abnormal events occurring in chemical processes. The accidental release of toxic gases may result in severe problems for people and the environment. Risk analysis of hazardous chemicals using consequence modeling is essential to improve the process reliability and safety of the refineries. In petroleum refineries: toluene, hydrogen, isooctane, kerosene, methanol, and naphtha are key process plants with toxic and flammable chemicals. The major process plants considered for risk assessment in the refinery are the gasoline hydrotreatment unit, crude distillation, aromatic recovery, continuous catalytic reformer, methyl–tert–butyl–ether, and kerosene merox units. Additionally, we propose a threat and risk analysis neural network for the chemical explosion (TRANCE) model for refinery incident scenarios. Significantly, 160 attributes were collected for the modeling on the basis of the significance of failure and hazardous chemical leaks in the refinery. Hazard analysis shows that the leakages of hydrogen and gasoline at the gasoline hydrotreatment unit, kerosene at the kerosene merox plant, and crude oil at crude-distillation units were areas of profound concern. The developed TRANCE model predicted the chemical explosion distance with an R2 accuracy value of 0.9994 and MSE of 679.5343.