The automatic and intelligent diagnosis of faults in marine equipment is a task that is considered to be of great importance considering the numerous tasks that are associated with professionals working on ships. The possibility of including automatic and intelligent processes on a ship makes it possible to monitor equipment more effectively and make more informed decisions. This approach has received a lot of attention in the academic and industrial fields as it can offer considerable economic and safety advantages. Some fault diagnosis approaches can be found in the literature, where mathematical and control theory models are taken into account. However, in complex processes not all their characteristics are always known exactly, so mathematical modelling of processes is an extremely difficult task. Fault diagnosis can therefore be based mainly on data or heuristic information. The inherent characteristics of fuzzy logic theory make it suitable for processing this type of information, which is why it will be used to model processes and diagnose faults in a marine equipment valve. The fault diagnosis architecture proposed in this paper is based on analysing the discrepancy signals obtained between the outputs of the fuzzy models and the process data under study. These discrepancies, the residuals, are indicative of equipment fault. The proposed fault diagnosis architecture uses an intelligent decision-making approach to indicate the occurrence of faults. In this paper, this architecture will be used to diagnose abrupt faults in a marine equipment valve.