Process facilities are vulnerable to catastrophic accidents due to the storage, transportation, and processing of large amounts of flammable/explosive materials. Among a variety of accident scenarios, fire and explosion are the most frequent ones. Fire and explosion are interactive events and may cause a "chain of accidents" (also known as the "domino effect"). Especially in processing facilities where units are located within a limited distance, fire or explosion occurring in one unit is likely to spread to other units. Currently, there is a lack of proper methodology that considers the effect of fire and explosion interaction. Ignoring this interaction provides uncertainty in the domino effect risk analysis. High complexity and uncertainty, due to the interaction of fire and explosion, thus make it challenging to analyze the domino effect propagation. The fuzzy inference system (FIS) is known to be an efficient tool for handling uncertainty and imprecision. The current study has developed a new methodology by adopting the FIS method to handle the data uncertainties in the dynamic Bayesian network (DBN) to conduct a robust domino effect analysis considering interactions of fire and explosion. Application of the proposed methodology demonstrates that the FIS acts as a quick semiquantitative method involved in the domino effect analysis. Results obtained from FIS are consistent with those obtained using the DBN. Moreover, it illustrates that DBN is an effective technique to analyze the combination of a fire and explosion accident.