Toward Greener Smart Cities - A Critical Review of Classic and Machine-Learning-Based Algorithms for Smart Bin Collection

Aug 6, 2026

Upload by Eliana

This study critically reviews research on machine-learning methods for optimizing smart bin collection in urban environments. The problem is typically formulated as a dynamic graph, where each smart bin’s changing fill level becomes a node over time. Reinforcement Learning, time-series forecasting, and Genetic Algorithms are analyzed together with Graph Neural Networks to improve routing and collection efficiency. Despite computational and adaptability constraints of individual approaches, the GNN-based RL, forecasting, and hybrid optimization perspectives support dynamic adaptation to real-time data and improved waste-management performance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages36

Reading Time1 h 31 min

AccessFree to download

Introduction

Smart bins and intelligent waste management

Classical routing algorithms for collection

From GIS and heuristics to data-driven optimization

AI and learning-based approaches in smart cities

electronics Review Toward Greener Smart Cities: A Critical Review of Classic and Machine-Learning-Based Algorithms for Smart Bin Collection Alice Gatti †, Enrico Barbierato *,† and Andrea Pozzi † Citation: Gatti, A.; Barbierato, E.; Pozzi, A. Toward Greener Smart Cities: A Critical Review of Classic and Machine-Learning-Based Algorithms for Smart Bin Collection. Electronics 2024, 13, 836 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)electronics13050836 Academic Editors: Jesús Ángel Román Gallego, María-Luisa Pérez-Delgado, María Concepción Vega Hernández, Alfonso Jose Lopez Rivero and Daniel Hernández De la Iglesia Received: 9 January 2024 Revised: 13 February 2024 Accepted: 16 February 2024 Published: 21 February 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . Department of Mathematics and Physics, Catholic University of the Sacred Heart, via Garzetta 48, 25133 Brescia, Italy; [alice.gatti@unicatt.it](alice.gatti@unicatt.it) (A.G.); [andrea.pozzi@unicatt.it](andrea.pozzi@unicatt.it) (A.P.) * Correspondence: [enrico.barbierato@unicatt.it](enrico.barbierato@unicatt.it)[ ](enrico.barbierato@unicatt.it)† These authors contributed equally to this work. Abstract: This study critically reviews the scientific literature regarding machine-learning approaches for optimizing smart bin collection in urban environments. Usually, the problem is modeled within a dynamic graph framework, where each smart bin’s changing waste level is represented as a node. Algorithms incorporating Reinforcement Learning (RL), time-series forecasting, and Genetic Algorithms (GA) alongside Graph Neural Networks (GNNs) are analyzed to enhance collection efficiency. While individual methodologies present limitations in computational demand and adaptability, their synergistic application offers a holistic solution. From a theoretical point of view, we expect that the GNN-RL model dynamically adapts to real-time data, the GNN-time series predicts future bin statuses, and the GNN-GA hybrid optimizes network configurations for accurate predictions, collectively enhancing waste management efficiency in smart cities. Keywords: smart bins; routing; graph neural networks; hybrid models 1. Introduction The role of AI in the domain of smart cities [1–4], especially in garbage collection, has recently emerged in the landscape of urban development and sustainable practices. In the contemporary landscape, the exploration of intelligent technologies—encompassing smart bins, robotic systems, predictive modeling, and optimized routing algorithms—and their pivotal role in optimizing waste collection processes has become essential. Thinking about city administration and garbage management, collecting, removing, and re-utilizing the produced garbage is an arduous task. The rapid accumulation of garbage necessitates a well-organized and efficient system, with a focus on minimizing the environmental impact wherever possible. The conventional garbage collection operation, with a rigid routine that encompasses the continuous reiteration of the combination of manual collection and removal with segregation and recycling tasks, can be both inefficient and resource-consuming. Consequently, integrating advanced technologies such as AI and smart waste management solutions is paramount. Pivotal components contributing to this transformation are smart bins, smart routing, smart segregation, and smart prediction. Hence, a novel approach to waste management is a characteristic of smart cities. Smart bins, also called intelligent dumpsters, are the starting point towards smartness in smart cities from a garbage co

FAQs

How is smart bin collection optimization commonly modeled in the reviewed research?

The task is usually formulated in a dynamic graph setting, where each smart bin’s evolving waste level is represented as a node.

Which algorithm families are reviewed to improve collection efficiency?

The review analyzes Reinforcement Learning, time-series forecasting, and Genetic Algorithms, including approaches that combine these with Graph Neural Networks.

What are the expected benefits of hybrid GNN-based models in smart bin management?

The GNN-RL perspective supports dynamic adaptation to real-time data, GNN time-series modeling predicts future bin statuses, and a GNN-GA hybrid optimizes network configurations to improve prediction accuracy and overall efficiency.

Aug 6, 2026

Upload by Eliana