Wireless Sensor Networks (WSNs) are composed of collaborative nodes that perform environmental monitoring and control tasks, but their functionality is constrained due to the limited energy of each node. The structural design of WSNs include the arrangements of nodes into clusters, the appointment of a Cluster Head (CH) for each cluster, and the optimization of energy usage. The process of selecting CHs is influenced by a variety of factors, including the node's remaining energy, the cost of communication, the density of nodes, mobility, and the size of the cluster. Inadequate CH selection can result in inefficient energy use. Furthermore, in the two-step communication process from nodes to the base station (BS), a significant amount of energy is expended. To mitigate this, a novel strategy that integrates various input parameters with a method based on distance thresholds has been developed to improve the selection of CHs and relay nodes. This strategy considers factors such as the Received Signal Strength Indicator (RSSI), the remaining energy of nodes, and their centrality. It employs fuzzy logic for the selection of CHs, and relay nodes are chosen based on their proximity to the BS. The determination of the optimal number of relay nodes is achieved through the K-Optimal and K-Means methods, ensuring that every CH is connected to at least one relay node for efficient data transmission. The proposed protocol, named Energy Efficient Cluster Heads and Relay Nodes (EECR), surpasses both the Multi-Layer Protocol (MAP) and Stable Election Protocol (SEP) in performance by extending the lifespan of the network by 43% and 33%, respectively.