Delay Tolerant Networks (DTNs) are a type of Mobile Ad-Hoc Networks (MANETs) where nodes are mobile, resulting in the absence of an end-to-end path between the source and destination nodes. Due to frequent disruptions in communication links, nodes in DTNs rely on a store, carry, and forward pattern to transmit messages. This forwarding and carrying of messages is achieved through cooperation among relay nodes. However, certain nodes may exhibit selfish behavior by avoiding cooperation to conserve their own resources, such as buffer and energy. In this paper, we propose a method for detecting selfish nodes in DTNs based on fuzzy logic. The method considers parameters such as the number of packets sent and received by a node, centrality degree, and buffer capacity as fuzzy inputs. The fuzzy outputs categorize nodes as active, semi-selfish, or selfish, and appropriate treatment is applied based on these categories. Simulation results demonstrate that the proposed method enhances the delivery rate by 10% and reduces the average delay by 15% and hop count by 8% when compared to existing approaches.