Internet-of-Things (IoT) has enormous interest from the research community in smart applications due to its ability to link any electronic object to a network. The IoT networks are centered over the Wireless Sensor Networks (WSN) and hence the IoT networks suffer from batter constraint problems similar to WSN. In the first contribution, Fractional Firefly Algorithm (FFA) and Chicken Swarm Optimization (CSO) algorithms are used to develop the FFA+CSO based multipath routing protocol. This model utilizes FFA for the optimal selection of cluster heads and CSO for optimal routing path selection based on different routing network metrics such as energy, delay, link quality, lifetime and hop count. The second contribution analyzes the fault tolerance limitations of FA+CSO and FFA+CSO to develop Fractional Gaussian Firefly Algorithm (FGFA) and Darwinian Chicken Swarm Optimization (DCSO) algorithms were developed to form FGFA+DCSO based multipath routing protocol by improving the FFA and CSO to adapt the energy efficiency in CH selection and fault tolerance in optimal route selection.