Related Experiment Video
Updated: Apr 28, 2026
11:53
The Modular Design and Production of an Intelligent Robot Based on a Closed-Loop Control Strategy
Published on: October 14, 2017
11.0K
Development and simulation of an innovative autonomous knowledge-based smart waste collection system
Mohamed Abdallah1,2,3, Mariam Hosny4,5
1Department of Construction Engineering, The American University in Cairo, New Cairo, Egypt. Mohamed-Abdallah@aucegypt.edu.
Scientific Reports
|April 26, 2026
Summary
A new cloud-based waste collection system (KWC) uses machine learning to predict waste levels, outperforming hardware-intensive smart waste collection (SWC) systems. KWC significantly cuts costs and environmental impact by optimizing routes and reducing collections.
Area of Science:
- Environmental Science and Engineering
- Computer Science and Artificial Intelligence
- Operations Research
Background:
- Modernizing waste collection is crucial for decarbonization efforts.
- Smart sensor-based waste collection (SWC) systems aim to optimize routes but face practical and commercialization challenges due to hardware intensity.
- Existing systems often rely on fixed schedules rather than actual waste levels.
Purpose of the Study:
- To introduce and evaluate an innovative cloud-based Knowledge-based Waste Collection (KWC) system.
- To replace hardware-intensive bin sensors with historical data and machine learning for waste forecasting.
- To compare the operational and economic efficiency of KWC against conventional and SWC systems.
Main Methods:
- Developed a KWC system integrating machine learning (XGBoost, deep neural networks, etc.) for waste prediction, expert bin selection, and route optimization.
- Simulated KWC, SWC, and conventional collection scenarios in a residential district using historical data.
- Employed heuristic algorithms for bin selection based on actual (SWC) and predicted (KWC) waste quantities.
- Utilized connected and autonomous vehicles in simulations.
Main Results:
- XGBoost demonstrated high prediction accuracy (4.2% relative error) for daily waste generation.
- KWC significantly reduced travel distance (60.9%), collected bins (89%), and collection days (10%) compared to conventional methods.
- KWC achieved a 63% cost reduction through its cloud-based approach, contrasting with SWC's insufficient 5% travel expense reduction.
- Connected and autonomous vehicles further reduced total delay by 90%.
Conclusions:
- The cloud-based KWC system offers substantial economic and operational advantages over hardware-intensive SWC systems.
- KWC effectively optimizes waste collection by leveraging machine learning for accurate waste forecasting and intelligent route planning.
- This approach presents a more practical and cost-effective solution for modernizing and decarbonizing the waste collection industry.
