A lightning strike is the main cause of fire accidents in chemical storage tanks, and the thermal radiation produced by a large storage tank fire may lead to a domino effect. A prediction method aiming to evaluate the probability of a domino effect at different levels triggered by lightning in the chemical tank farm is proposed. The developed method takes into account both the probability calculation model of fire triggered by lightning and the assessment method of the subsequent domino effect probability. Firstly, the accident scenarios and causes of fire triggered by lightning are analyzed by the event tree method, and a probability calculation model of fire accident triggered by lightning is developed. Secondly, the graph of chains of accidents is built considering synergistic effects and multi-level domino effects, and the Bayesian network is applied to calculate the probability of each accident chain. The most dangerous primary equipment is identified by comparison with the probabilities of the domino effect at different levels. By setting up the failure states of different tanks, the probabilities of events are updated under a given situation, and the most susceptible target equipment with respect to the domino effect are identified. Finally, the method is illustrated with two case studies in a chemical industry park. The results will be helpful for the prevention of domino effects based on the theory of chain-cutting disaster mitigation. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.