Abstract

The rapid adoption of Internet of Things (IoT) development has created the possibilities of smart environments that can maintain constant monitoring and smart control of the vital resources. Nevertheless, traditional rule based and fixed optimization methods tend to be rigid to dynamic and heterogeneous environments in the real world and result into inefficient use of resources and higher energy use. To offer solutions to these issues, the paper is proposing the implementation of an intelligent IoT-enabled smart environment framework to bring neural computing in conjunction with a Genetic Algorithm (GA)-based optimization plan to sustainable resource management. Under the suggested model, the IoT sensors would receive real-time data about the environment in which they are installed and the way they are used, which are then processed by a neural computing unit to forecast resource demand trends. A GA in turn uses these predictions to make adaptive optimization of the allocation of resources under a variety of sustainability and operating constraints. The fitness function used by the GA reflects a combination of energy efficiency, usage of resources, and comfort to the user and it allows making strong decisions in dynamic environments. The experimental testing done on a simulated smart environment shows that the suggested framework will considerably decrease the consumption of the resources and enhance the general sustainability of the process as compared to the traditional non-optimised and heuristically-driven techniques. The findings confirm the usefulness of the evolutionary optimization strategy along with the use of neural intelligence that structured the adaptive, energy-efficient IoT-based smart environments, and they demonstrate that the framework can be implemented in the next-generation sustainable architecture of infrastructure systems.