Efficiently managing the huge amount of data generated by IoT sensors in industrial environments is challenging. In this proposal, we explore the integration of active learning technique to optimize sensor usage by intelligently enabling and disabling sensors based on data patterns indicating uncertainties. In doing so, our approach ensures thorough inspection of products while conserving network resources. This intelligent and adaptive strategy reduces unnecessary energy consumption in battery-powered IoT environments, such as factories, where wearables and sensors play a critical role. Furthermore, the proposed strategy provides a strategic balance between network congestion, energy conservation, and quality assurance, marking a significant advancement in IoT sensor network management for industrial applications.