In a distributed system, fog computing (FC) is an emerging computer technique. The goal of FC is to position cloud-based services in close proximity to endpoints. The method is meant to meet the minimal latency requirement for Iot - based healthcare equipment. There is a wide range of healthcare data volumes produced by Iot - based healthcare equipment. Network congestion and increased delay are the direct effects of this massive influx of data. Patient information is rendered useless and insufficient for end-users when round-trip time delays rise due to huge data transfer and increasing hop counts between IoTs and cloud servers. In the healthcare industry, real-time data is essential for time-sensitive applications. The medical IoT devices and their users have stringent requirements for latency, and traditional cloud servers just can’t provide them. Therefore, it is important to decrease network delay, compute latency, and communications latency while transmitting data through the Internet of Things. FC allows data to be stored, processed, and analyzed in the cloud and at the network’s edge, where the latency is lower. This article proposes an innovative approach to solving the aforementioned issue. It combines an FC-based analytical model with a hybrid fuzzy-based RL algorithm. High latency in healthcare IoTs, between users and cloud servers, is something that has to be mitigated. Allocation and selection of data packets in an IoT-FC setting are handled with the help of a fuzzy inference system, reinforcement learning, and neural network evolution techniques provided by the suggested smart FC model parameters and algorithms. The method is put through its paces on the iFogSim (Net-Beans) as well as Spyder simulations (Python). The acquired findings demonstrated that the suggested strategy outperformed the state-of-the-art techniques.