Blockchain-enabled IoT framework with energy-efficient machine learning for scalable and secure smart cities

In a rapidly urbanizing environment, cities have changed into complicated ecosystems requiring sophisticated technological solutions to resolve excessive traffic, energy utilization, waste management, and public safety issues. This study discusses a single architecture for IoT-enabled smart cities through the use of blockchain enabled security, energy efficient machine learning, real-time analytics, and decision-making to overcome scalability, interoperability, and security issues generally present in a smart infrastructure. The framework utilizes lightweight algorithm-based cost-effective computation, integration of heterogeneous IoT devices, real-time decision making, transparency, and involvement of stakeholders. The simulation findings show substantial advantages over traditional methods: a 35 % decrease in processing latency; a 25 % decrease in energy consumption; and a 29 % increase in an index for data security. Also, predictive analytics exhibited over 90 % identification accuracy across the different urban contexts, including traffic control for improved public safety, and environmental monitoring/management scenarios that ensured reliable forecasted events and appropriate resource allocation. The blockchain module demonstrated median transaction validation times of less than 2 ms to validate IoT data streams enabling real-time secure operations even under demanding environmental conditions. Also, we achieved resource allocation optimization with efficiencies that exceeded 85 % for designated priority supplies, including food, energy, medical resources, and reduced waste and improved disaster resilience. This model is adaptable across different urban settings and is a scalable, secure, and energy efficient framework for the next generation of smart cities contributing to sustainable urbanization and improved quality of urban life.