<p>In edge-based serverless systems, scheduling strategies that consider residual energy across the network are key to achieving efficient resource allocation for IoT devices. In systems that utilize this technology, the resource allocation process is determined at runtime based on changes in the residual energy across the network of network nodes. The decision to allocate the necessary resources to execute requests is not straightforward and depends on the residual energy across the network of currently operational nodes. Therefore, to enhance resource efficiency, minimize access latency, and handle the highest possible number of network requests, it is essential to deploy pre-schedulers guided by forecasts of residual energy in operational network nodes. In this paper, we introduce an energy-conscious scheduler selection mechanism, referred to as an energy-conscious intelligently-driven scheduler, which is designed based on the residual energy across the network of currently operational nodes. Additionally, a framework is proposed to enhance system availability by increasing the time when the residual energy across the network of currently operational nodes decreases. Three distinct workload distribution models were employed to assess the effectiveness of the proposed method, and the results show that this approach has prevented energy waste and reduced energy consumption by an average of 1.54%. Also, the network availability has increased by an average of 6.4% compared to other approaches. Additionally, the proposed approach has demonstrated uninterrupted performance in terms of fault tolerance across all scenarios. Experimental outcomes demonstrate the proposed method’s strong capability in reducing energy usage while enhancing the stability and robustness of serverless environments.</p>