Network densification has rapidly emerged as a critical aspect of the IoT since IoT networks are rapidly expanding, and there are immense difficulties in allocating resources, controlling latency, and improving energy efficiency. Accordingly, this research work presents a computational mathematics approach that seeks to improve the IoT network performance and resource utilization. Linear programming is used to control the use of resources, stochastic modeling is used to forecast loads, and calculus is used for optimization of power. The model’s implementation on IoT networks shows enhanced network performance in terms of the latency, dropped packet rates and energy utilization. According to quantitative results, there is an improvement on the latency with a value less than twenty percent and on the power that differs by less than thirty percent, which proves that the proposed model can fulfil the growing IoT system requirements.