Optimizing waste management in smart cities: An IoT-Based approach using dynamic bald eagle search optimization algorithm (DBESO) and machine learning

Waste management is a significantly challenging task in all countries, with the increasing amount of waste being produced by the proliferating population. Conventionally, the inefficient system is followed with irregular schedules, non-optimized routing, and failed prediction of waste generation which might increase the operational costs, inefficient resource utilization, and degradation of the environments. Manually managing waste takes enormous time hence internet of things-based smart management has been implemented in many smart cities. Meanwhile, it became a hot topic and probed the explicit link between the population and waste. The smart management of waste is predominant for disposal in big cities. Concerning these issues, this work proposes an innovative technique for smart bin monitoring and solid waste management. It is also associated with the machine learning approach for the status predictions of bins incorporated with waste forecasting prediction for the future. These are achieved with IoT-based devices. The IoT-based sensors are used to collect the information from the smart bins and forward to the storage module for further processing. The installation of bins in the required area is achieved with the statistical bin distribution approach followed by the Dynamic Bald Eagle search optimization algorithm (DBESO) based Kernel Soft Extreme Learning Machine (KSELM) technique is proposed for the status prediction of bins and future waste generation forecasting prediction. For effective disposal of waste, the levels of the bins are classified as Null, Partial, and Loaded levels and thus alert the users and avert the overflow of the waste. The information is passed to the successor stage through the wireless connection and saves the environment from the pollution. The integrated version of the proposed work provides better solutions for the theoretical advancements incorporated with real-time waste management solutions in smart cities. Based on the iteration of 20th to 100th our proposed system reached the proposed precision levels of 94.9 ​%, 95.4 ​%, 95.2 ​%, 96.1 ​% and 96.3 ​% that are better than the other approaches. The accuracy of predicting the waste attains 96.42 ​% which is higher than the existing works. • Utilizes IoT sensors and DBESO-KSELM technique for accurate waste status prediction and timely disposal. • Employs statistical distribution algorithm for efficient bin distribution and reduces waste accumulation. • Achieves better waste management in smart cities through simulations and analysis of statistical parameters. • Integrating deep learning models with optimization algorithms and IoT for enhanced waste management solutions.