An AIoT-enabled framework for real-time predictive analytics and sustainable optimization in urban waste management systems
Abstract
The high rate of urbanization and population growth has led to an increased generation of municipal waste, which poses challenges for environmental sustainability and efficient waste management. Most waste management systems are still manual, inefficient, non-scalable and can’t support intelligent real-time decision making, which are increasing costs, fuel consumption and environmental pollution. Despite recent developments in Artificial Intelligence (AI) and the Internet of Things (IoT), current waste management systems have yet to achieve intelligent levels of predictive ability, interpretability and adaptability to complex urban conditions. The study suggests an AIoT based smart waste management system that fuses the features of the IoT based real-time sensing, cloud-edge computing, deep learning based predictive analytics, explainable artificial intelligence (XAI), and sustainable route optimization approaches. The proposed framework uses smart bin sensors to constantly record waste levels and environmental conditions, and a stacked Long Short-Term Memory (LSTM) model is used to predict waste level and collection need in real time. Moreover, explainability techniques are integrated with SHAP analysis to enhance the transparency and trustworthiness of the model. Experimental evaluation was performed on those publicly available datasets of smart waste management, and synthetically generated IoT sensor data to create realistic urban waste scenarios 97.1% prediction accuracy, 96.8% precision, 97.0% recall, 96.9% F1-score, and a ROC-AUC score of 0.978, which outperforms the conventional machine learning methods like Random Forest, Decision Tree, Logistic Regression, KNN, and XGBoost. Further sustainability analysis shows that the fuel consumption has been reduced by 32% and that there were 35% fewer unnecessary collection trips and 72% fewer overflow cases. By combining real-time IoT sensing, explainable Artificial Intelligence and intelligent route optimization, the proposed AIoT-based framework enables smart city waste management with high accuracy, interpretability, scalability and eco-friendly operation for next-generation smart cities.
Related
- A System-Level Framework for Sustainable Packaging Design and Waste Management Integration
- A coupled surface water-groundwater multi-objective optimization framework for coordinated water-ecosystem-agriculture management in arid inland river basin
- Artificial Intelligence for Blue Transformation: A Review of Predictive Modeling and Decision Support Systems in Sustainable Aquaculture
- Editorial for the Second EJP SOIL Special Issue: “Climate‐Smart Sustainable Agricultural Soil Management for the Future II”
- How sustainable land management practices may impact water resources in the Sahel: Insights from an integrated critical zone modelling approach
BibTeX
@article{ullah2026aiotenabled,
title = {{An AIoT-enabled framework for real-time predictive analytics and sustainable optimization in urban waste management systems}},
author = {Haseeb Ullah and Asia Kanwal and Noreen Ashraf and Saira Andleeb Gillani and Meshari Alazmi and Ibrahim Abdullah Zalah and Faleh Alqahtani and Saddam Hussain Khan},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-69874-9},
url = {https://www.nature.com/articles/s41598-026-69874-9},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - An AIoT-enabled framework for real-time predictive analytics and sustainable optimization in urban waste management systems
AU - Haseeb Ullah
AU - Asia Kanwal
AU - Noreen Ashraf
AU - Saira Andleeb Gillani
AU - Meshari Alazmi
AU - Ibrahim Abdullah Zalah
AU - Faleh Alqahtani
AU - Saddam Hussain Khan
PY - 2026
DA - 2026/10/21
JO - Scientific Reports
DO - 10.1038/s41598-026-69874-9
UR - https://www.nature.com/articles/s41598-026-69874-9
AB - The high rate of urbanization and population growth has led to an increased generation of municipal waste, which poses challenges for environmental sustainability and efficient waste management. Most waste management systems are still manual, inefficient, non-scalable and can’t support intelligent real-time decision making, which are increasing costs, fuel consumption and environmental pollution. Despite recent developments in Artificial Intelligence (AI) and the Internet of Things (IoT), current waste management systems have yet to achieve intelligent levels of predictive ability, interpretability and adaptability to complex urban conditions. The study suggests an AIoT based smart waste management system that fuses the features of the IoT based real-time sensing, cloud-edge computing, deep learning based predictive analytics, explainable artificial intelligence (XAI), and sustainable route optimization approaches. The proposed framework uses smart bin sensors to constantly record waste levels and environmental conditions, and a stacked Long Short-Term Memory (LSTM) model is used to predict waste level and collection need in real time. Moreover, explainability techniques are integrated with SHAP analysis to enhance the transparency and trustworthiness of the model. Experimental evaluation was performed on those publicly available datasets of smart waste management, and synthetically generated IoT sensor data to create realistic urban waste scenarios 97.1% prediction accuracy, 96.8% precision, 97.0% recall, 96.9% F1-score, and a ROC-AUC score of 0.978, which outperforms the conventional machine learning methods like Random Forest, Decision Tree, Logistic Regression, KNN, and XGBoost. Further sustainability analysis shows that the fuel consumption has been reduced by 32% and that there were 35% fewer unnecessary collection trips and 72% fewer overflow cases. By combining real-time IoT sensing, explainable Artificial Intelligence and intelligent route optimization, the proposed AIoT-based framework enables smart city waste management with high accuracy, interpretability, scalability and eco-friendly operation for next-generation smart cities.
SN - 2045-2322
ER -