Assessing the Implementation of Artificial Intelligence-Based Systems for Sustainable Solid Waste Management in Selected Countries in Asia, Europe and North America: A Systematic Review

Abstract

The efficient management of solid waste remains a persistent challenge worldwide, particularly in developing countries, due to increasing waste generation driven by rapid population growth and urbanisation. Artificial Intelligence (AI) has emerged as a transformative approach to improving solid waste management (SWM) by enhancing waste collection, sorting, recycling, and resource recovery processes. Several countries have integrated AI into their SWM systems; however, the extent of AI-driven automation across the SWM value chain remains insufficiently understood. This study examines the implementation of AI-powered systems across various areas of SWM in ten selected countries to identify progress, benefits, challenges, and opportunities for advancing sustainable SWM practices. A systematic literature review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 268 studies published between 2005 and 2025 were identified and analysed. The findings indicate that the selected countries have made significant progress in adopting AI-powered systems across different areas of the SWM value chain, contributing to improved efficiency, resource recovery, and sustainability. However, challenges related to data availability and quality, privacy, cost, infrastructure, and ethical considerations remain. Addressing these challenges requires collaboration among governments, waste management agencies, technology providers, private companies, and researchers. The study provides best-practice insights and recommendations for leveraging AI to optimise resource efficiency and support improved economic, environmental, and social outcomes in sustainable solid waste management.

1. Introduction

Globally, inappropriate solid waste management (SWM) practices have been documented as a major source of harmful air pollution, environmental dilapidation and severe health complications [1,2]. Swift urbanisation, technological transformation, intensifying population growth and economic and social advancement have considerably amplified the volume of solid waste generated in many countries [3,4]. The World Bank [5] reported that the volume of solid waste generation is projected to surge from around 2.01 billion tonnes in 2018 to 3.40 billion tonnes by 2050. Universally, around half of the solid waste generated is mishandled and deposited in illegal open waste dumps, incinerated through open burning or abandoned in uncontrolled landfills [6]. These common practices are frequently not regulated, predominantly in emerging countries, and have led to massive environmental and health hazards, including ground and surface water adulteration, poor air quality, escalating occurrences of cancer and birth defects, among others [7]. Past studies have identified several key factors contributing to ineffective solid waste management practices, including poor planning, the absence of sustainable management strategies, inefficient operations, limited knowledge and public awareness, weak regulatory and legislative frameworks, and inadequate funding [8].

Sustainable SWM aims to considerably reduce the quantities of natural resources presently being consumed by providing opportunities to reuse, recycle and recover valuable materials as many times as possible prior to reaching the end of their useful life. It also helps to reduce waste generation and guarantees that waste produced is disposed of in a manner that moderates adverse environmental and health consequences [9]. In recent years, AI technologies have gained popularity and reputation and have been applied in numerous research areas including SWM [6]. The implementation of AI and robotic technologies in SWM can potentially alter the design, processes and operations of waste treatment plants in municipalities, which could eventually lead to advanced operational effectiveness and enhanced sustainable management of solid waste [10]. AI-based systems improve the entire solid waste management value chain through the application of smart sensors, computer vision, and machine learning [4]. It transforms waste management operations from rigid schedules to data-driven systems, boosts sorting speeds to around 10 times faster than humans, and maximises material recovery. For example, in waste collection and transportation, sensors are used to track fill levels in real-time so that waste trucks can only visit bins that are full; monitor live traffic, weather, and bin data to shorten driving distances and reduce carbon emissions; and forecast areas with heavy waste generation in order to position waste trucks and staff more effectively. Likewise, in sorting recycling and resource recovery, high-speed cameras and multi-spectral sensors are used to scan fast-moving belts to identify plastics, metals, and paper by type. Machine learning algorithms are used to evaluate complex waste items to determine the most efficient methods to extract precious metals [2,4]. Thus, AI-driven systems in solid waste management help support economic growth through encouraging circular economy practices; protection of the environment through lowering emissions and pollution; support for cleaner and safer communities; and creation of job opportunities [9,11].

At the moment, nations around the world are gaining substantially from applying AI to develop and implement roadmaps, processes and procedures in their waste management systems. The implementation of AI in SWM in many countries will support processes such as collection, sorting, and recycling, optimised through smart decision-making abilities. AI technologies leverage machine learning algorithms and data analytics to enhance operational effectiveness and reduce issues emanating from human health and the environment arising from inappropriate SWM practices [2,9].

This study examines the implementation of AI-powered systems in various areas of SWM in the selected leading countries to enhance sustainable SWM practices. The study provides an overview of progress made, evaluates the level of AI automation and approaches, explores the benefits and challenges and presents best practice recommendations on how resource efficiency can be optimised to improve outcomes from economic, environmental and social aspects in other countries. In the context of this study, the level of automation in the selected countries implementing AI systems in SWM refers to the degree of independence and cognitive complexity with which AI systems perform tasks that were previously done by humans. The level of automation ranges from simple rule-following tools to fully autonomous AI systems that make and execute decisions. The specific AI algorithms used in SWM in the selected countries are beyond the scope of this study. To achieve the aims of this study, prior studies from 2005–2025 from various databases are collected and analysed. The authors used prior studies from 2005–2025 to provide a comprehensive overview of the literature on AI and SWM. Other reasons include (a) ensuring the research results are reliable and complete, (b) identifying additional relevant studies across databases, (c) reducing the risk of missing key data, (d) improving the accuracy of data extraction in the literature, and (e) ensuring the review results are trustworthy and reliable with strong conclusions. The findings of this study provide increased awareness of AI-powered systems in SWM and how the selected countries are implementing AI in SWM for enhancing sustainable waste management practices. From a real-world perspective, this study provides insights on AI as a support system for the sustainable management of solid waste to eradicate the use of manual labour, reduce operational costs, and boost efficiency to alter the management of solid waste.

This study identified three research gaps. Firstly, there is an absence of awareness of how AI technologies work in relation to their application to SWM. Secondly, researchers, practitioners, and policymakers have long commended AI technologies as a key enabler for a more efficient SWM system. While these advancements assure an escalating automated future for amassing, sorting, and recycling solid waste, very little is understood about the present degree of automation in countries implementing AI. Thirdly, the majority of studies reviewed [9,12] focused on governments’ automation intentions and fundamentally ignored the extent of actual implementation of AI technologies.

Although there are prior studies [9,13,14,15] on AI and SWM, no study has assessed how the selected countries (Austria, Germany, Denmark, USA, UK, Japan, Singapore, Switzerland, South Korea and Netherlands) are implementing AI-powered systems in resolving the varied SWM problems for accomplishing sustainable SWM practices. According to OECD [16], the selected countries are among the leading countries in AI deployment in SWM. The novelty/originality of this paper lies in the premise that it is the first comparative research on how the selected leading countries are implementing AI-driven systems in various areas of SWM to enhance sustainable SWM practices. To the best of the authors’ knowledge, this is the first study to examine the extent of AI implementation in the selected countries’ solid waste management value chain as well as the opportunities and challenges. As the selected countries’ solid waste generation is rapidly increasing, innovative systems are urgently needed to deal with the mounting solid waste stream to preserve the health of the residents and keep communities clean [17].

Thus, this study proposed two research questions:

• What is the extent of AI automation in the solid waste management value chain of the selected countries?
• What are the benefits and challenges of AI implementation in the selected countries?

The structure of this study is captured in five sections as follows: The existing literature on AI and SWM, the statement of the problem and gaps in the literature, and the research aims and justification for this study are presented in Section 1. The research methods employed and the rationale for undertaking a systematic literature review based on secondary data are outlined in Section 2. AI in SWM and sustainable development goals (SDGs) in the selected countries, AI implementation in SWM in the countries studied, and their application areas are discussed in Section 3. The analysis of the results is presented in Section 4. The results/outcomes of the study, limitations related to the present study, best practices and recommendations and opportunities for future research are provided in the final section.

2. Research Methods

2.1. Search Strategy and Selection Criteria

In evaluating the implementation of AI technologies in the selected countries’ SWM value chain, this study utilised secondary data gained through a systematic literature review (SLR) of articles from past research on AI and SWM. The systematic review was designed and reported in line with the PRISMA 2020 statement, which offers updated guidance for transparent identification, selection, appraisal, and synthesis of studies in systematic reviews (completed PRISMA 2020 checklist in Supplementary Materials). The reasons for conducting an SLR are to identify, assess and interpret prior studies on AI technologies and SWM. The purpose of systematically reviewing the literature is to ensure a fair evaluation, which leads to trustworthiness and dependability of the outcomes of the research [18]. The five-step procedure for carrying out an SLR by Wolfswinkel et al. [19] is espoused in this study. This five-step procedure “(a) defining the scope of the review, (b) searching the literature, (c) selecting the final samples, (d) analysing the samples using content analysis and (e) presenting the findings” is discussed below.

2.1.1. Defining the Scope of the Review

To begin, we clearly defined the scope of the review and clarified the explicit requirements for the inclusion and exclusion of pertinent studies and the requirements to identify and retrieve those studies in the literature. We then explored four major databases, including “Emerald”, “ScienceDirect”, “ProQuest” and “Web of Science”, used for the search. The databases were selected largely because of their far-reaching coverage and representativeness in publishing credible and trustworthy academic papers on AI and SWM. To achieve the necessary reportage on the existing studies in the databases, keywords such as “solid waste management” AND “artificial intelligence” OR “artificial intelligence implementation” OR “artificial intelligence systems” OR “artificial intelligence opportunities” OR “artificial intelligence challenges” AND “artificial intelligence in the selected countries” are utilised for the search. In setting the limitation, the researchers considered relevant criteria such as the nature and type of document, which includes book chapters, peer-reviewed conference papers, academic journals and other reputable reports from international institutions such as the United Nations (UN), World Health Organisation (WHO), government agencies and regulatory bodies, year of publication (2005–2025) and language in English. An explanation of the search strategy and the criteria for selection is provided in Figure 1.

2.1.2. Literature Search

In this stage, we first ran the search query in the selected databases to recover the search results. 32,822 articles were found based on the predetermined search strings explained above. This preliminary search conducted offers a wider coverage of the understanding of AI and SWM topics in the literature. Prior to applying the inclusion and exclusion criteria, the deduplication method in this systematic literature review was used to combine search results from multiple databases and then remove identical or overlapping articles. The authors conducted the deduplication method by comparing the titles, year of publication, authors and DOIs to avoid duplication and ensure each relevant study is counted only once. To control the results of the search, precise inclusion and exclusion criteria were adopted as shown in

Table 1. Inclusion and exclusion criteria adopted in this study.

Inclusion Criteria	Exclusion Criteria
- Articles that have information on solid waste management, artificial intelligence, artificial intelligence implementation, artificial intelligence systems, artificial intelligence opportunities and challenges in the selected countries and other related articles on AI and SWM	- Articles that do not emphasise solid waste management, artificial intelligence, artificial intelligence implementation, artificial intelligence systems, artificial intelligence opportunities and challenges in the selected countries and other related articles on AI and SWM
- The year of publication of the articles is between 2005–2025	- Articles that are not within the year of publication for this study.
- Articles are obtained from “academic journals, book chapters, peer-reviewed conference papers and other established reports from the United Nations, World health Organisation and other government agencies	- Articles arising from untrustworthy and unreliable sources
- Language of the article is in English	- Articles that are not written in English

.

Figure 1. Search strategy and criteria for selection.

Figure 1. Search strategy and criteria for selection.

2.1.3. Final Article Selection

We then selected the final articles for a detailed analysis. This selection is limited to the title and the abstract of the articles in order to enable the authors to focus on the results of the search. The titles and abstracts of all the final selected articles were examined for their applicability to AI and SWM. This resulted in the identification of 3257 articles. Those articles that did not meet the criteria were then removed, leading to a total of 268 articles to be reviewed further for analysis.

2.1.4. Analysing the Samples Using Qualitative Content Analysis

The identified 268 articles have been read in full for coding and analysis using qualitative content analysis. Qualitative content analysis can be used to analyse extensive data ranging from textual data to visual and audio data [20]. In this study, textual data or written text have been used as the data source for the qualitative content analysis. The aim of content analysis in this study is to: “(a) organise and draw meaning from the selected articles and to obtain realistic inferences from it; (b) determine and identify the keywords and themes in the articles and (c) develop a qualitative classification and categorisation of the selected articles” [7,20].

An explanation of the steps used in the qualitative content analysis is described below. Based on a three-step approach adopted by Srivastava and Thomson [21], and Ritchie and Lewis [22], the selected articles were carefully analysed, and emergent categories across the data were then identified. Firstly, we scrutinised and analysed the articles using an approach to qualitative content analysis [21]. Thereafter, based on the themes that emerged from the preliminary evaluation of the articles, a coding frame was created. These codes were complemented by the emergent codes identified through closer and careful reading and re-reading of the articles [23]. Secondly, the articles were coded following an iterative process of initial coding, refining of the coding and subsequent re-coding to ensure the robustness of the coding while checking for potential researcher bias and similarities between the coding. Thirdly, coded data were then summarised, and summaries of coded articles were compared to identify emerging themes and patterns. This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. The PRISMA framework was used to guide the identification, screening, eligibility assessment, and inclusion of studies. The study selection process is presented using a PRISMA flow diagram as shown in Figure 2.

3. Results of the Review

3.1. Overview of AI

In recent years, AI has emerged to become one of the most disruptive technologies and is transforming how individuals perceive and interact with the physical and digital worlds and remodelling the experiences, choices and values of humans [24]. AI technologies have a strong impact on all aspects of society and the economy, including sectors such as waste management, environmental management, transport, health, finance, education, manufacturing, welfare and government services. When suitably implemented, AI technologies provide coast-to-coast economic advantage and social value [7,9]. Presently, AI is implemented in countless areas globally and has progressively become a fundamental technology that individuals, organisations and governments need to expand their profitability and prosper. For example, the application of AI technologies in SWM is continually refining the methods to collect, transport, sort and recycle several categories of waste, including waste from medical, electronics, food and radioactive materials, among others [25]. Wilts et al. [2] defined AI as “machines or computers that mimic cognitive functions that are associated with the human mind, such as learning and problem solving”. According to Abdallah et al. [6] and Gao et al. [26], AI is defined as “a collection of many different technologies, computational methods and techniques working together to enable machines to sense, comprehend, act, and learn with human-like levels of intelligence”. AI has the ability to learn from prior knowledge, make logical decisions and reply quickly with positive outcomes [9].

Wilt et al. [2], and Chaudhry and Dhawan [27] reported that AI has four main components and these include: (a) expert system which deals with scenarios under examination as an expert and provides the expected outcomes, (b) heuristic problem solving which evaluates a small collection of solutions involving some guesswork to discover near-optimal solutions, (c) natural language processing which enables communication between humans and machine in natural language, and (d) computer vision which engenders the ability to recognise shapes and features automatically”. Numerous kinds of AI hardware (machines/robots) and software (algorithms) exist, and each has diverse competences and abilities and at varying heights of advancement [27].

3.2. Global AI Statistics in Solid Waste Management

In recent years, the emergence of the association between AI and SWM is meaningfully leading the sustainability research. AI technologies are progressively being used to deal with some of the most persistent global environmental challenges such as climate change, global warming, and waste minimisation. The application of AI in SWM is currently transforming how we collect, handle, process, use and recycle solid waste, thereby enhancing sustainable SWM practices. According to OECD [28], AI in SWM is predicted to increase from USD 1.6 billion in 2023 to about USD 18.2 billion by 2033, and the sector is expected to surge at an annual growth rate of 27% by 2033. Figure 3 summarises a 10-year forecast on the global trends of AI in SWM from 2023–2033.

Figure 3 shows that the implementation of AI in SWM is experiencing substantial growth, driven by the rising requests for forward-thinking solutions to deal with SWM challenges worldwide. The implementation of AI is characterised by the widespread acceptance of AI technologies across numerous segments such as municipal, residential, commercial, industrial, private organisations and governments. Major players in the AI and SWM sector are investing in AI-driven technologies to advance the sorting of waste efficiently and increase the capabilities of recycling facilities. These sectors’ expansion is supported by workable government policies that facilitate sustainable SWM practices and technological improvements that allow for the deployment of AI in complex and multifaceted waste management systems. The strategic incorporation of AI not only boosts operational efficiencies but also makes a contribution to the wider goals and objectives of environmental sustainability [11].

The application of AI in the SWM sector is segmented into three components (hardware, software and services) based on the category of waste, technology and the end user. The category of waste is segmented into industrial, hazardous, electronic, food, chemical, plastic, paper and biological wastes among others. By technology, the SWM sector is categorised into “predictive models, classification robots, intelligent garbage bins, sensor–based monitoring, waste facility software, waste wireless detection software and waste to energy systems”. The end user is segmented into governments, commercial and residential. The application of AI in SWM supports sustainable waste management practices by simplifying the recycling processes, increasing speed and lowering cost, reducing human error, and lowering carbon emissions and pollution across the main stages of the solid waste management value chain such as generation, collection, sorting and processing, disposal, treatment and recovery.

Table 2. Global emerging trends in the implementation of AI in the solid waste management sector.

AI Systems	Global Emerging Trends in Solid Waste Management	References
Smart bins and IoT integration	The solid waste management sector is increasingly utilising smart bins that leverage AI and Internet of Things (IoT) technologies. These bins can automatically sort solid waste and optimise collection schedules by analysing waste types, fill-levels and thereby enhancing efficiency in solid waste processing and recovery.	Zhang et al. [29]; Dubey et al. [30]; Ullah et al. [31]; Ahmed et al. [32]; Bijos et al. [33]; Olawade et al. [34]; Fuqaha and Nursetiawan [35]
AI-powered sorting systems	These AI powered sorting systems have become a key trend in SWM. The systems can recognise and separate diverse kinds of solid waste materials with high accuracy, consequently improving recycling rates and reducing contamination in recycling streams.	Gundupalli et al. [36]; Ma et al. [37]; Kutty and Abdella [38]; Yan et al. [39]; Mookkaiah et al. [40]; Dash and Sharma [41]
Route optimisation for solid waste collection systems	Solid waste collection via route optimisation improves waste collection and helps in making the process as efficient as possible for the waste truck. It reduces the number of miles made to accomplish a route of collection; reduces carbon emissions; reduces costs and improves fleet management. The systems uniquely support waste collections and ensure that waste collection processes in the field are captured in the systems.	Nwokediegwu et al. [42]; Bibri et al. [43]; Munir et al. [44]; Assef et al. [45]; Agrawal et al. [46]; Anh et al. [47]
Predictive analytics for operational efficiency systems	AI-driven predictive analytics systems are currently being used to forecast waste generation patterns and optimise logistics. This includes planning the routes for collection trucks and operational times, thereby reducing operational costs.	Bakh and Ahmed [48]; Liao and Wang [49]; Garre et al. [50]; Andeobu et al. [4]; Salam et al. [51]; Reza et al. [52];
Robotics in waste management systems	Robotic systems powered by AI are currently being implemented to handle and sort solid waste more efficiently. These robots can operate in environments that are unsafe for humans, improving productivity.	Sarc et al. [53]; Monzambe et al. [54]; Mao et al. [55]; Wang et al. [56]; Bhari et al. [57]; Nafiz et al. [58]; Bui et al. [59]
Enhanced waste data management systems	AI is improving data management in solid waste management through supporting the analysis of large datasets to track waste generation and management trends. This leads to better decision-making and regulatory compliance.	Rajasekaran et al. [60]; Ali et al. [61]; Nguyen et al. [62]; Ahmad et al. [63]; Andeobu et al. [4]; Yan et al. [64]
Waste to energy systems (WTE)	The procedure for producing energy that takes the form of electricity and/or heat from the primary processing of solid waste is known as waste-to-energy (WTE). As a procedure of recovering energy, WtE uses various AI powered technologies that can convert non-recyclable solid wastes into energy through several processes such as gasification, incineration, pyrolysis, anaerobic digestion and landfill gas recovery.	Kumar et al. [65]; Ge et al. [66]; Varjani et al. [67]; Suárez Valdés et al. [68]; Foong et al. [69]

below summarises the global emerging trends arising from the implementation of AI in the SWM sector.

3.3. AI in Solid Waste Management and SDGs in the Selected Countries

This section discusses how the selected countries are leveraging AI in SWM to achieve specific sustainable development goals (SDGs). The successful implementation of AI technologies in SWM has aided the selected countries in fast-tracking SDGs. The concept of sustainable development has been referred to as a revolutionary and forward-thinking paradigm [70]. The Brundtland Commission [71] reported that sustainable development is: “development that meets the needs of the present without compromising the ability of future generations to meet their own needs”.

The tension between innovation and equitable resource distribution, as well as between realising the needs of the environment, economy and society, is successfully demonstrated through the concept of sustainable development [70]. In the context of solid waste, sustainable waste management means sustainably reducing waste generation and encouraging the selected countries to adopt sustainable waste management practices and reporting. From the perspective of AI, sustainable AI is a drive to nurture change in the entire lifecycle of AI applications and systems towards superior environmental integrity and social justice. Sustainable AI entails how to advance AI that is harmonious and compatible with ensuring that the environment, economy and society are sustained [70,72]. Undoubtedly, a harmonised approach that reflects societal, economic and environmental considerations in the planning, development, and implementation of AI systems can create benefits that support SDGs [70]. AI-based technologies in SWM can support SDGs in the selected countries in the ways discussed below:

3.3.1. AI in Solid Waste Management for Social Sustainability

According to Vinuesa et al. [73], around 82% of society could perhaps gain from AI-based technologies and systems that support the realisation of SDGs. SDGs offer an exceptional structure and framework for evaluating the effects of AI on the SWM sector as well as for ascertaining potential interactions among them. For example, AI in SWM supports “SDG 1 (no poverty), SDG 4 (quality education), SDG 6 (clean water and sanitation), SDG 7 (affordable and clean energy), and SDG 11 (sustainable cities and communities)” in the selected countries. Accordingly, AI acts as an enabling tool for these SGDs through facilitating the delivery of food, water, health and clean energy services to communities in the selected countries. It can also facilitate low-carbon emissions through supporting the establishment of circular economies and smart cities that efficiently utilise their resources. However, achieving these SDGs depends on how the technological enhancements empowered by AI are applied in the various countries studied. These selected countries have different approaches, needs, cultural values and wealth in relation to sustainable development [73].

3.3.2. AI in Solid Waste Management for Economic Sustainability

The technological benefits provided by AI in SWM support the achievement of “SDG 8 (decent work and economic growth), SDG 9 (industry, innovation and infrastructure) and SDG 12 (responsible consumption and production)” in the circular economy of the selected countries [73]. Prior studies of Marzouki et al. [72] found that around 79% of the SDGs targets can be impacted by AI positively, while 35% may be impacted negatively by developments in AI. For instance, AI capabilities can support the building of a circular economy, at a much faster pace than would be possible without AI [72,74]. Across the selected countries, three potential circular economy opportunities can be unlocked with AI technologies to “(a) enhance and accelerate the development of new products, components and materials suitable for a circular economy through machine-learning, (b) magnify the competitive strength of circular economy business models by combining real-time and historical-data from products and users, and (c) build and improve the reverse logistics infrastructure required to close the loop on products and materials by improving the processes to sort and disassemble products, remanufacture components, and recycle materials” [73,75].

3.3.3. AI in Solid Waste Management for Environmental Sustainability

In achieving environmental outcomes in the selected countries, low-carbon energy systems can be supported with high integration of renewable energy and energy efficiency, which are required in addressing the climate crisis [72]. AI can also promote and enhance the health of the environment and entire ecosystems as it supports “SDG 7 (affordable and clean energy) and SDG 13 (climate action)” [73]. In addition to supporting the SDGs, AI acts as a catalyst for realising environmental goals of attaining sustainable energy and lower emissions in the countries studied [76]. Previous studies [9,77] demonstrate that AI has incredible prospects to fast-track combined efforts to protect and save the environment. Further, the preservation of resources is supported by AI through enabling seamless development and advancement of greener transportation, monitoring deforestation, removal of CO₂, and forecasting weather conditions. Similarly, in the selected countries, AI implementation has been used to overcome difficult challenges in the environment such as ocean health, biodiversity and conservation, water problems and clean air [73].

3.4. AI Implementation in Solid Waste Management in Selected Countries

The implementation of AI powered systems is evolving in the selected countries as a tool to accelerate SWM across the waste management and recycling industry. In recent years, there has been increasing implementation of AI technologies in SWM in the selected countries to provide solutions to the diverse challenges in SWM [72,78]. In this section, a framework of assessment of the extent of AI automation in the selected countries as well as a country-by-stage comparison matrix is presented. The section also provides a review of how the selected countries have successfully implemented AI technologies in solid waste recycling as a fundamental aspect of their waste management systems, along with tangible examples of how AI is shaping and modelling a more efficient and sustainable future in the selected countries. Although there are many countries where solid waste recycling is not visible, countries such as Austria, Germany, Denmark, the USA, the UK, Japan, Singapore, Switzerland, South Korea, and the Netherlands have been recognised by OECD [79] as leading the world in solid waste recycling. Japan, South Korea, and Germany lead OECD countries in solid waste recycling with high recycling rates of 79%, 72% and 69% respectively in 2024. Consequently, the criteria for selecting the 10 countries are based on their high recycling and recovery rates. According to the United Nations Environmental Program [80], Singapore is recognised as a global leader in solid waste management. The nation, through its National Environment Agency, reduced the volume of solid waste sent to landfills by up to 97% in 2024. Other selected countries such as the USA (63%), Austria (58%), South Korea (59%), the Netherlands (50%) and Switzerland (51%) are among the leading nations in solid waste recycling with high recycling and recovery rates. It is noteworthy to state that the majority of the selected countries are from Europe due to their high recycling and recovery rates. According to Laureti et al. [81], European Union countries lead in waste management because of strict, legally binding EU laws, clear waste reduction goals, and the “waste hierarchy” model. The EU Directives force member nations to stop using landfills, boost recycling, and make manufacturers responsible for the life cycle of their products through extended producer responsibility (EPR) schemes. Countries such as Turkey, Chile, Costa Rica, Romania, Cyprus, Malta, and some countries from Asia and Africa with recycling rates under 20–30% were excluded from this study [80,82].

3.4.1. Framework for Assessing AI Automation Levels in Selected Countries

In solid waste management, AI automation has become an essential component of streamlining processes and procedures, enabling waste management organisations to simplify their operations, reduce costs, and improve efficiency. However, the process of implementing AI automation is multifaceted and requires careful planning and execution [83]. Existing literature on the automation of solid waste management processes and procedures has focused on examining future automation technologies. Prior studies of Anagnostopoulos et al. [84] examined the concepts for digital waste management in sustainable cities, Ramos et al. [85] and Shah et al. [86] investigated simulations for digital dispatching and routing, Wagland et al. [87] focused on software-enabled image classification for waste sorting and Rovetta et al. [88] examined smart bin prototypes. Clearly, these studies focused on automation intentions but do not indicate the level of actual implementation of digital technologies. In this study, we have developed a framework to assess the level of AI automation in the selected countries based on value-chain stages, scope of deployment, technological maturity, operational status and demonstrated outcomes as illustrated in Figure 4.

Based on the review of the literature, an assessment of the level of AI automation in the selected countries was conducted. The authors considered five progress indicators (value-chain stages, scope of deployment, technological maturity, operational status, automation outcomes) to determine whether each indicator is “Progressing Speedily”, “Progressing” or “Progressing Slowly”.

Table 3. AI automation and implementation progress indicators in the selected countries.

Selected Countries	AI automation and Implementation Progress Indicators
	Value-Chain Stages	Scope of Deployment	Technological Maturity	Operational Status	Automation Outcomes
Austria	Progressing speedily	Progressing	Progressing slowly	Progressing	Progressing
Denmark	Progressing speedily	Progressing speedily	Progressing	Progressing	Progressing speedily
Germany	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily
United States	Progressing speedily	Progressing slowly	Progressing slowly	Progressing slowly	Progressing
United Kingdom	Progressing speedily	Progressing speedily	Progressing	Progressing	Progressing speedily
Japan	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily
Singapore	Progressing speedily	Progressing slowly	Progressing slowly	Progressing slowly	Progressing
Switzerland	Progressing speedily	Progressing speedily	Progressing slowly	Progressing slowly	Progressing
South Korea	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily	Progressing speedily
The Netherlands	Progressing speedily	Progressing speedily	Progressing slowly	Progressing slowly	Progressing

Data adapted from Shokrollahi [89]; BMUV [90]; Wales Audit Office (WAO) [91]; Yjlee [92]; Henam and Sambyal [93]; Stadlmann and Zehetner [94]; Ang [95]; Mansveld [96].

presents AI automation and implementation progress indicators in the selected countries.

3.4.2. Germany

Germany has been acknowledged as leading the way in the management and recycling of solid waste. The nation has led efforts across Europe in improving solid waste reduction through implementing various AI-based technologies and is supported by a hierarchy that facilitates preventing, reusing, recycling, recovering as energy, and landfilling for effectively managing and recycling solid waste [90,97]. By leveraging AI-powered technologies, waste sorting processes in Germany have become more efficient, and this has lowered costs and increased profit margins. German waste management organisations use AI-powered computer vision systems, which recognise each item of a waste stream using low-cost cameras and machine learning algorithms. For example, Recycleye, a robotic sorting company, has implemented its AI-powered computer vision system and robotic sorting technology in Germany’s waste management industry, supported by Veolia Germany’s U-Start programme, which supports innovative start-ups in solid waste disposal. The aim is to make solid waste recycling economically better and an attractive proposition and prevent additional valuable recyclable materials in Germany from being lost to landfill [97].

Currently, Germany recycles more than 70% of the solid waste generated, and this is the highest globally. The nation attained this through workable practices, policies and regulations in relation to efficient SWM. Apart from the implementation of AI technologies, the development of the German waste management and recycling industry has been very successful and is facilitated through handing out a number of waste specific directives including: “(a) efficient control of administrative bodies, (b) commitments of producers (product responsibility schemes), (c) environmentally sensitive citizens; (d) substantial saving of natural resources, minimisation of health and environmental vulnerabilities, (e) creation of qualified jobs in the recycling sector, (f) efficient controlling and monitoring, and (g) ongoing R&D” [98,99,100]. For example, to reduce waste generation, German policymakers made producers responsible for managing solid waste from the packaging they developed. Specifically, producers are also made to pay a fee when additional packaging material is used, and this has resulted in less packaging material and consequently less waste. In 2019, the German government introduced the “German Packaging Act 2019”. The aim of the Act is to significantly reduce the effect of solid waste on the environment, as well as ensure producers and retailers are made to take responsibility for promoting and upholding the use of eco-friendly products [99].

3.4.3. United Kingdom (UK)

Solid waste management and recycling in the UK is handled by local governments in collaboration with licensed recycling companies [91]. The UK government supports a circular economy which is founded on a notion that improves resource efficiency and can considerably contribute to a real increase in financial, environmental and societal benefits and payoffs. The implementation of AI in the solid waste and recycling industry in the UK requires a framework that is clear and supportive. As a result, the UK Department for Environment, Food & Rural Affairs (DEFRA) have released policy guidelines that allow materials recovery facilities (MRFs) to use AI technologies in their recycling and recovery operations. For example, the government of Leeds introduced the “Zero Waste Leeds Recycling Scheme” in 2021, which supports the implementation of AI on a number of solid waste recycling facilities. Similarly, the Welsh government introduced the “Zero Waste strategy” in 2020, which supports the implementation of AI-powered systems and banned a variety of single-use, tough-to-recycle and regularly littered plastic items [101]. Although SWM and recycling have been successful in the UK, the implementation of AI-powered systems in SWM has significantly improved recycling outcomes through enhancing the experience of consumers, offering valuable data insights, reforming and streamlining operations and improving the efficiency of the sorting process [91]. In addition to the implementation of AI to enhance resource recovery, the recycling system in the UK has been very effective due to efficient control and administration of waste management directives as well as environmentally sensitive citizens [101].

3.4.4. South Korea

In recent years, the South Korean government has strengthened its efforts in reducing solid waste generation and increasing the recycling rate. Currently, South Korea’s solid waste recycling rate is around 60%, which is among the highest in the world. Specifically, the nation plans to reduce its plastic waste stream by 50% and recycle 70% of it by 2030 [102]. Recently, the South Korean government introduced AI-powered intelligent waste management systems to deal with its solid waste challenges. This system uses predictive analytics and data modelling to forecast the trends in waste generation and optimise the various collection routes [92]. This significant innovation has led to efficiency in their operations and a reduction in cost, with favourable ecological impacts. In addition, the nation’s waste management companies use AI-driven tools to educate and create awareness while engaging residents in waste reduction and recycling practices.

This two-way approach of promoting public participation and encouraging waste collection demonstrates the country’s pledge and commitment to sustainable urban development and environmental sustainability [102]. Further, the South Korean government adopted the Polluter Pays Principle (PPP) scheme in 2020. This volume-based waste system imposes a differentiated treatment cost and is calculated and determined by the amount of waste that each resident generates [92]. This strategy is noteworthy because it works and provides an economic incentive that actualised and made sense of the PPP scheme [92,93]. Additionally, the AI-based volume waste fee system played a significant role in reducing the amount of waste generated and fostering the recycling rate. It also assists in reducing the waste processing cost and creates an environmentally friendly processing method, paving the way for a society with minimal waste generation [92]. In 2019, the Global Waste Index ranked South Korea in first place out of 36 OECD countries for the best waste management nation [103]. In the Eunomia [104] report, South Korea ranked third in the world for the highest recycling rates.

3.4.5. Denmark

The Danish SWM and recycling policies comprise both prevention and handling of solid waste [105,106]. Municipal and regional councils are responsible for the administration and implementation of SWM. In recent years, the Danish government implemented AI-powered robotic technologies developed by Zen Robotics. These intelligent robots sort materials and range from flammable waste, thick waste, metal and wood to plastics originating from municipalities, industry and businesses [106]. The construction of AI sorting plants eradicates virtually all work-related health risks linked with manual sorting, raises the degree of purity by up to 98% and reduces the accompanying costs. These new AI sorting plants not only increase the recycling percentage and the quality of the outputs, but also offer better flexibility in solid waste sorting and a greater degree of documentation. In addition to the implementation of AI, all municipal councils evaluate solid waste quantities and put together solid waste management plans, and these have been very effective in the management and disposal of solid waste in Denmark [107]. While Denmark has effectively managed its solid waste generation, the implementation of AI in SWM has transformed the way the nation manages solid waste [108].

3.4.6. Austria

Austria is among the countries that have one of the highest recycling rates across the world. In recent years, AI-powered systems have been implemented to make major contributions to improving solid waste recycling in Austria. For example, the Austrian AI-Waste project has been used to optimise the overall waste treatment processes [109]. Currently, food processing companies and retailers in Austria are using AI to minimise food waste. Specifically, SPAR Austria has implemented AI to help reduce food waste generation through developing AI-powered systems that permit more targeted order suggestions and demand predictions. The AI systems analyse data on volume of sales, climate conditions, advertising promotions and other factors, including seasonality, to produce a precise forecast of optimal product order quantities. The new AI-powered system provides a high accuracy prediction rate of more than 90%, which means that the precise amounts are available at the correct time, thereby further reducing the generation of food waste [94,110]. Further, Austria’s adoption of AI-based technologies and implementation of its efficient PPP scheme, where households and organisations pay for any solid waste produced that is recyclable or non-recyclable, have been very successful [94,111]. Approximately 96% of Austria’s population attempts to separate their solid waste into categories that are recyclable, and every household sorts an average of a million tons of solid waste annually [94]. Similar to Germany, Austria also runs a producer responsibility model. Since March 2020, the government banned manufacturers in Austria from importing and marketing certain categories of plastic bags, which mandates retailers not to distribute these categories of plastic bags in the country [94].

3.4.7. Switzerland

Switzerland is among the largest solid waste producers in Europe. SWM in Switzerland is based on the PPP scheme. As a result, the recycling rate has more than doubled in the past 20 years due to this strategy. In addition to the adoption of the PPP scheme to manage and recycle solid waste in Switzerland, AI-based technologies have been implemented in recycling household goods made from tin and aluminium materials, light bulbs, paper and electronic products [112]. The adoption of AI-driven waste management systems not only boosts operational effectiveness but also endorses and promotes sustainability and environmental conservation in Switzerland. As with Austria and Germany, recycling costs are incurred by the public as well as manufacturers, which implies that residents are encouraged to fund the cost of recycling. Alongside this policy, bin bags for solid waste are also taxed [112,113]. For example, across the City of Zurich, there are over 12,000 different waste recycling collection points. The Swiss government has now made solid waste recycling mandatory, and failure to comply can result in significant fines. Currently, more than 50% of the solid waste generated in Switzerland is recycled, and what is remaining is used for energy generation. None of the solid waste generated in the country is disposed of in landfills, which facilitates, supports, and lowers greenhouse gas emissions [113].

3.4.8. Singapore

Singapore has some of the lowest usage of landfills to manage solid waste in the world [114]. Singapore’s approach to SWM incorporates AI to support its reputation for efficiency and sanitation. The country has implemented AI-driven systems that employ a mix of innovative sensors and machine learning algorithms in the management and recycling of solid waste. These systems are capable of identifying and sorting numerous categories of solid waste and ensuring that they are sorted into the correct solid waste streams for recycling [95]. The application of AI has led to substantial improvements in operational activities and improved recycling rates. This operational method is not only beneficial to the entire ecosystem but also functions as a basis for urban sustainability. Further, manufacturers and retailers as well as residents in Singapore are made entirely responsible for the solid waste they produce and how they are disposed of [95,115]. These solid wastes are collected in nominated trucks and are taken to centres where they are sorted into different recycling streams powered by AI. There is only one landfill, which is mostly used for non -recyclable plastics in Singapore, and the rest of the solid waste which cannot be recycled is then used for energy production [95].

3.4.9. United States

Waste management (including solid waste) is a critical issue currently confronting communities across the United States, with substantial implications for public health, environmental sustainability, and socioeconomic equity. As a result, the majority of the states are now employing AI-driven waste management systems in the communities [42]. The implementation of AI-driven waste management systems in the USA has transformed how solid waste is handled and optimised. Leveraging AI technologies, including route recommendations based on IoT, deep learning models and machine learning algorithms in municipalities, has significantly enriched waste collection efforts and enhanced sustainability [116]. These AI-driven systems support the prediction of waste generation, optimise landfill site selection, and thereby help minimise the overall cost of solid waste disposal [117]. The implementation of AI for waste reduction and recycling has helped move cities in the US towards a more circular economy, substantially reducing environmental impact and promoting sustainability [42].

Furthermore, cities in the US utilise AI in the process of decision-making by providing insights into waste generation patterns and arrangements [117] and supporting the identification of opportunities and benefits of recycling and reducing solid waste [113,118]. For example, the recycling programme at the City of San Francisco has been enhanced with AI technologies at its facilities. These AI systems adopt sophisticated image recognition technology to sort recyclables faster and more effectively [113]. This technology ensures a purified and sorted material, decreasing contamination and enhancing the overall efficiency of the recycling process. Consequently, the volumes of waste moved to landfills have decreased significantly, aligning with the city’s ambitious environmental goals. The City’s use of AI in this context not only streamlines SWM processes and procedures but also contributes to a sustainable urban environment [118].

3.4.10. Netherlands

The Netherlands has led the way in SWM in recent years. Presently, the country recycles around 78% of its solid waste, incinerates about 19%, and merely 3% ends up in landfills when compared to the European Union (EU) with an average of 40% [119]. These figures demonstrate the country’s pledge to solid waste recycling and reflect good business sense adopted by the Dutch waste management organisations [119]. A significant reason for the country’s good track record can be ascribed to the close partnership between industry and the local, provincial and national governments [119]. Motivated and encouraged by the EU policy, the Dutch government has been recurrently developing and introducing innovative recycling techniques and methods [120]. With the advantage of a population recognised for its eco-consciousness and eco-friendly characteristics, the country has established innovative AI-powered systems that can efficiently deal with solid waste while leading in ecosystem conservation and optimisation of resources. From efficiently recycling and reducing waste to energy recovery and advanced techniques in sorting, the Netherlands has embraced AI-powered methodologies that are beneficial to both the environment and its citizens [121]. When compared to other countries, waste management and recycling in the Netherlands are subject to more rigorous rules and regulations than in other countries. A major factor that contributes to the Netherlands’ high rates of recycling is its powerful AI systems for efficient waste collection and sorting. In addition, the nation has a widespread network of recycling facilities and recycling centres, making it convenient and easy for its citizens to separate and correctly dispose of diverse categories of solid waste. For example, the implementation of AI-powered technologies and automated systems for sorting in organisations that manage solid waste has significantly improved the accuracy and efficacy of the recycling processes and recovery [121].

3.4.11. Japan

Japan is among the top generators of solid waste in the world, particularly food waste. The nation has a reputation for high solid waste separation compliance and collection rates. It is estimated that Japan generates more than 6 million tonnes of solid waste yearly and spends roughly 19 billion USD in managing and recycling solid waste [122]. To reduce solid waste generation and the resulting associated costs, the Japanese government has endorsed new policies and strategies to reduce solid waste. One of the key strategies is the introduction of the pay-as-you-throw (PAYT) system. While some of the selected countries adopt the PPP scheme and apply flat charging systems instead of PAYT systems, the Japanese government has used PAYT to successfully manage the huge amount of solid waste it generates [122].

Although the flat charging approach has been sufficient to secure the funding needed to operate and maintain municipal SWM in some of the selected countries, each individual paying the same amount regardless of how much solid waste they generate does not create any incentive for people to reduce solid waste generation [123]. Through the approach of employing a PAYT system in Japan, the cost of managing and recycling solid waste has been adjusted to how much waste people generate. This approach sends a message to the communities and reminds them that waste management is not for free but varies just like other utilities, such as water and electricity consumption [123,124]. Currently, solid waste management companies in Japan are implementing AI-powered systems to reduce solid waste and cut costs. The incorporation of AI in Japan’s SWM systems not only accelerates and facilitates operational efficiency but also drives innovation and collaboration among all stakeholders. Various stakeholders, including governments, businesses, and technology providers in Japan, are working together, advancing and implementing AI solutions specifically tailored to the unique needs of the country [124].

Thus, there are a variety of options for managing solid waste in the countries studied and the implementation of AI-powered systems which best meet the needs of each specific country. While most of the advanced countries described above have effectively implemented AI-powered solid waste recycling technologies, many emerging countries are yet to embrace AI technologies to support the achievement of sustainable waste management practices. Overall, the AI-powered technologies and other approaches adopted in the selected countries should sensibly match capital and operating costs to ensure that SWM and recycling are sustainable.

3.5. Practical Application Areas of AI in Solid Waste Management in the Selected Countries

Waste management is a critical aspect of maintaining ecological balance and sustainability. Solid waste management programs such as “extended producer responsibility”, “polluter pays principle”, and “takeback schemes”, among others, have enabled the governments of the selected countries to “reduce, re-use, recycle, and recover” valuable materials from solid waste. However, solid waste in various categories has been shown to be multifaceted, expensive to recycle as well as labour-intensive and creates major health risks to SWM workers directly engaged in collecting and disposing of solid waste on a regular basis [10,125]. Consequently, the implementation of AI-powered systems in the selected countries is changing how solid waste is managed through technological intervention, thus making traditional methods of recycling better restructured and more efficient. AI technology-led initiatives in SWM in the selected countries have also influenced the way solid waste is handled from generation, collection, sorting, and transportation through to recycling [126,127]. By leveraging AI, the selected countries successfully lower costs, enhance safety, and reduce the environmental and health implications associated with SWM [128,129]. The real-world AI application areas and the extent of automation in SWM in the selected countries are discussed in the following sections and illustrated in Figure 5.

3.5.1. Waste-to-Energy Systems

In recent years, the implementation of AI technologies has attracted substantial interest in SWM, particularly waste-to-energy (WTE) conversion [130]. WTE involves generating energy from waste materials, including solid waste. In addition to reuse and recycling, WTE is the last chance to recover valuable materials that would have otherwise ended up in landfills with no benefits [130]. Appropriately investing in WTE supports employment, economically benefits the communities and facilitates social development of countries worldwide [131]. As a result, the selected countries are implementing AI in their WTE conversion processes. It is noteworthy to state that conversion technology is the core infrastructure executing a fundamental change, such as converting data formats and energy types, while the AI-driven optimisation/automation layer is the cognitive superstructure that decides when, how, and what parameters the core technology runs with to maximise efficiency or yield. Thus, the AI-driven optimisation layer uses data models and machine learning to monitor sensors, adjust temperatures, and control feed rates in real-time to maximise product quality and energy efficiency in WTE conversion processes [132]. Currently, there are more than 100 WTE plants situated in the selected countries [16]. AI-powered WTE plants have a sophisticated gas cleaning system that conforms to the strictest air pollution standards.

Table 4. Criteria for selecting the appropriate WTE in the selected countries.

Criteria	Key Considerations	Description
AI technologies	Type, quantity and quality of waste	Compatibility of a specific AI technology with the composition, energy value, the volume of the solid waste generated in the country
	Volume of waste generated	Amount of energy produced from waste using a specific AI technique. Generally, the technique that generates the maximum energy is desirable
	Waste segregation technique	Sensitivity of a specific AI technology used in the waste segregation process. Generally, AI technology with a lower sensitivity is often preferable
	The potential for energy generation	Quantity of energy produced using a specific AI technology. The AI technology adopted is the technology that generates the maximum energy
	Consistency of supply	Sensitivity of the AI technology adopted to the consistency of supply
Social factors	Health and safety of the public	The use of specific AI technology that would be less detrimental to human health and safety during the plant’s operation
	Social acceptance	Resistance of the locals which can adversely impact on the establishment of waste to energy plants. Type of automation generally acknowledged by the locals is generally desirable
Economic factors	Cost of capital	Initial investment needed for the plant to commence operation.
	Operation and maintenance costs	Operational and maintenance costs such as repairs, plant consumables, purchases including insurance and replacements
Environmental Factors	Production of hazardous/toxic waste	Quantities of dioxins and furans during the operation of a specific AI technology. The lesser the emission, the more desirable the AI technology
	Greenhouse gas (GHG) emissions	Amount of GHG emission from a specific technology during the energy conversion process

Data adapted from Fuqaha & Nursetiawan [35]; Anagnostopoulos [84]; He et al. [128].

summarises the criteria for selecting the appropriate WTE in the selected countries.

Generally, WTE conversion using AI in the selected countries can be categorised into (a) “thermochemical conversion such as incineration, pyrolysis, gasification, and hydrothermal liquefaction, (b) biochemical conversion including waste volarisation, fermentation and anaerobic digestion, and (c) mechanical conversion such as landfilling equipped with biogas production” [66]. In recent years, thermochemical conversion have received more attention when compared to biochemical conversion and mechanical conversion owing to benefits such as higher (a) efficiency during the process of conversion, (b) the concept of zero-waste, (c) reduced period of resistance, (d) enhanced and increased economic performance, and (e) compatibility with several feedstocks either wet or dry [68]. Figure 6 illustrates the conversion of solid waste (feedstock) via the various WTE processes and resulting main products.

The frequency of WTE conversion varies significantly among the selected countries and depends upon the volume of solid waste generated, population and lifestyle. Generally, WTE plants require certain conditions to operate optimally and feedstock requirements. The operating conditions and feedstock requirements depend on certain factors such as the type of waste/feedstock used, its chemical composition, as well as the amount of moisture content, and these often have substantial influence on plant efficiency.

Table 5. Operating conditions and feedstock requirements for WTE technologies.

WTE Technologies	Thermal Process/Methodologies Adopted	Operating Temperatures	Feedstock Requirements
Thermochemical conversion	Incineration Pyrolysis Gasification Hydrothermal liquefaction	Incineration −850–1200 °C Pyrolysis −400–800 °C Gasification −800–1600 °C Hydrothermal liquefaction –300–350 °C	Dry waste
Mechanical conversion	Landfilling equipped with biogas production	900–1200 °C	Organic and dry waste

Data adapted from Foong et al. [69]; Ahmad et al. [130].

presents a summary of the operating conditions and feedstock requirements for WTE conversion technologies.

Thermochemical Conversion

• Incineration

Incineration has for many decades been the main technique employed in SWM as it helps in reducing the volume/weight of solid waste by 70–90% as well as saves land and space. Incineration is suitable for categories of solid waste with high calorific values and is effective in the eradication of bacteria, viruses and other particulate matter in solid waste, preventing these organisms and substances from entering the natural environment [133]. In recent years, AI-powered incinerators have been accompanied by energy and heat recovery units, which have largely improved their value and efficiency. These AI-powered incinerators are used in selected countries for the combustion of solid waste at high temperatures ranging from 850 °C to 1200 °C (Table 5) in the presence of excess oxygen or air, which converts the generated energy into electricity [133]. The entire process comprises three stages: “(a) Combustion of the solid waste in the presence of air in the range of 850 °C to 1200 °C (b) utilisation of hot gases generated from the combustion for heat energy recovery and electric energy, and (c) emission control”. While incineration helps recover the energy from solid waste, it also leads to the generation of greenhouse gases such as “Carbon dioxide (CO₂) and Nitrogen oxides (NO_x)” which are controlled by the AI powered incinerators ([134,135]).

Among the selected countries, Denmark and Japan are the leading countries in incineration technologies. In 2023, Japan and Denmark incinerated 75% and 67%, respectively, of their solid waste generated for energy recovery. According to Shan et al. [136], there are 1141 incinerators operating in Japan as of 2023 with a total capacity of 181,891 tons/day. Japan’s national average incineration rate is about 75%, which places Japan as a leading country worldwide in terms of the number of incinerators and the incineration rate. In Denmark, incineration generated around 67% of the entire household heat consumption in the national energy systems [106]. In 2022, U.S. power plants generated about 12% of electricity from incinerating about 26.6 million tons of combustible solid waste for electricity generation [42].

Figure 7 summarises the percentage share of solid waste incinerated based on WTE conversion for energy recovery in some of the selected countries.

• Pyrolysis

Pyrolysis is a thermochemical conversion technique for the treatment of solid waste. It often performs better in the absence of oxygen and needs higher operating temperatures that range from 400–800 °C [69]. In recent times, pyrolysis has been highly beneficial in solid waste treatment and has attracted significant attention for recycling various kinds of solid waste to recover char, condensable oil and gases [138]. Generating these main products from pyrolysis treatment of solid waste depends on several factors, such as the temperature of the pyrolysis conversion, rate of heating, residence time, type of waste, composition of feedstock, as well as particle size and operating conditions [139]. For example, when planning to produce huge amounts of oil, plastic waste would be most suitable rather than using a mixed bag of solid wastes. Although the char produced from solid waste is characterised by high temperature values [140], the existence of toxic organic contaminants and heavy metals can be a concern and requires further consideration [65].

Generally, higher reaction temperatures during pyrolysis promote “(a) volatile cracking, (b) dehydration, (c) decarboxylation reactions, and (d) secondary decomposition of char”, which results in increased gas production compared to oils [141]. Likewise, a longer residence time speeds up secondary decomposition, repolymerisation, and recondensation reactions, leading to increased char yield [45,142,143]. While pyrolysis has shown strong viability in the treatment of solid waste, correctly predicting pyrolysis processes is challenging due to multi-scale complex reactions. As a result, the selected countries are implementing AI-powered thermochemical pyrolysis processes in the treatment of solid waste, as it produces value-added products [144,145]. The applications of AI in pyrolysis enhance the optimisation process, boost product yield and quality, as well as save costs, leading to lower budgets accompanying the pyrolysis process [146,147].

• Gasification

Gasification is another significant thermochemical conversion technique for the treatment of solid waste, often at heating temperatures between 800–1600 °C in a controlled environment with oxygen, steam, and air. The major end product that results from this process is syngas, which is composed of CO, H₂, CH₄, CO₂, and insignificant quantities of ethane and ethylene [143]. This syngas can then be utilised to manufacture other products such as liquid fuels, speciality chemicals, and energy recovery. Gasification is considered a favourable WTE technique because it can produce hydrogen, a clean energy source with a high heating value of around 141.7 MJ/kg [143]. There are four phases in the gasification process, and these are “(a) dehydration/drying, (b) pyrolysis/devolatilisation, (c) oxidation/combustion, and (d) reduction”. These phases involve numerous endothermic and exothermic reactions. The drying zone naturally eradicates the moisture content of the biomass to below 10% at temperatures lower than 200 °C [51,143].

To optimise the efficiency of these techniques and to minimise their negative impact on the environment in the selected countries, AI-powered systems are implemented in the optimisation of gasifier design in order to improve the efficiency of gasification and enhance efficient syngas production. AI-powered systems are also adopted in the selection of the most appropriate feedstock for the generation of biogas for more efficient and sustainable SWM. For example, in 2023, a total of 9.3 million USD was invested by the US in federal funding to develop cutting-edge technology solutions using gasification techniques to make clean hydrogen a more accessible and inexpensive fuel for the generation of electricity, transportation and industrial decarbonisation [148]. As the selected countries move towards a low-carbon future, AI-powered gasification technologies are poised to play a pivotal role in shaping the energy landscape. Gasification offers a workable and sustainable pathway towards meeting energy demands in the selected countries while mitigating environmental impact due to its ability to convert diverse feedstocks into clean energy and valuable products. By embracing evolving trends and advancements in gasification technologies, the selected countries have paved the way for a cleaner, greener and more resilient future for their communities [143].

• Hydrothermal liquefaction

In recent years, hydrothermal liquefaction (HTL) has attracted more attention in WTE systems design and implementation due to its ability to handle various wet biomass feedstocks and reshape WTE systems [149]. This transformative technology provides sustainable solutions for addressing inadequacies, increased costs and conservation concerns associated with traditional SWM practices. According to Gulec et al. [149], hydrothermal liquefaction is defined as a thermochemical procedure used for realising bio-oil from biomass as the primary target, along with solids, gases, and aqueous phases as by-products, in the presence of water as a solvent at a reaction temperature and high pressure. The quality and quantity of products in the hydrothermal liquefaction process are considerably influenced by several key factors such as “(a) temperature, (b) pressure, (c) heating rate, (d) preloaded pressure, (e) residence time, (f) feedstock characteristics, (g) catalysts, (h) solvent-to-feedstock ratio, (i) particle size and pH” [150]. Hydrothermal liquefaction has the potential to create lower-oxygen bio-oil faster than other thermo-chemical methods. During the hydrothermal process, the oxygen content of the biomass is reduced from about 40% to around 10–15% [151]. When operating at high temperature and pressure conditions, hydrothermal liquefaction has the ability to effectively reduce the volume of waste, moderate the discharge of detrimental pollutants and extract valuable energy from organic waste materials [149]. Hydrothermal processes are generally endothermic at low temperatures but become exothermic at high temperatures.

As a major condition, the reaction temperature is viewed as the most noteworthy parameter throughout the process of hydrothermal liquefaction and should be raised to an optimal point to realise the highest bio-oil yield. In the initial process, increasing reaction temperature would lead to achieving the highest bio-oil yield and promote the disintegration of the biomass [152]. Characteristically, bio-oil yield surges with the reaction temperature up to a level where additional upsurge subdues liquefaction and moves to the gasification phase with secondary decomposition. Conversely, at reaction temperatures below 275 °C, bio-oil yield appears to show a decline owing to the incomplete breakdown of biomass components; hence, a reaction temperature alternating between 300–350 °C is considered suitable for superior bio-oil yields and lesser production of solids and gases [17]. An advantage of the hydrothermal liquefaction process is that it merely consumes roughly 10–15% of the energy in the biomass feedstock and produces an energy efficiency of approximately 85–90%. A drawback of this technique lies in its increased cost of manufacture owing to the use of specialised equipment needed and severe corrosion that necessitates continuous maintenance due to extreme reaction conditions [29].

In recent years, the selected countries have implemented AI-powered systems in various biofuel production systems to promote hydrothermal conversions to predict and optimise products from char, oil, and gas. Machine learning tools have exhibited favourable characteristics to optimise hydrothermal liquefaction operating parameters in the selected countries to enhance their SWM capabilities and applicability [153]. Recent advances include: (a) development of catalysts, (b) continually developing hydrothermal liquefaction processes, (c) unified process configurations, and (d) upgrading/elevating existing techniques. Collectively, these developments position hydrothermal liquefaction as an adaptable and favourable technology for the sustenance of biomass conversion and managing resources and reflect the ongoing commitment in the selected countries to explore and harness its full potential to create a more sustainable and environmentally responsible future [154].

Biochemical Conversion

• Anaerobic digestion

Traditional Anaerobic digestion (AD) has long been an extensive technique employed in SWM, mostly in rural regions, because the biogas obtained by this process requires less clean-up, as it is naturally high in methane (CH₄) [155]. AD has contributed substantially to biogas production and has been generally acknowledged and recommended as an alternative method that is flexible and ecologically friendly. As a result, AD is sometimes symbolised as “biomethanation”—a technique that produces biogas in the absence of oxygen from other organic waste materials [156]. Anaerobic digestion of organic materials, including wastes from agriculture, food, industries, sewage sludge and other wet materials, produces biogas that can be used for either generating power or for heating and cooling systems [156]. In addition to preventing organic waste ending up in landfills, the AD process also generates useful renewable energy [157]. It is a viable source of energy that supports numerous SDGs. For example, by offering a sustainable energy source and minimising methane emissions from organic waste, biogas satisfies “SDG-7 (Affordable and Clean Energy)” and “SDG-13 (Climate Action)” [158,159]. Additionally, it promotes circular economy principles through waste-to-energy conversion, supporting “SDG-12 (Responsible Consumption and Production)” [160]. Biomethane is acknowledged as a practical solution to the difficulties and challenges linked with crop residues and is primarily used for the generation of power and steam, heating, cooking, as well as automobile fuel [161]. According to Helander et al. [162], 1 m³ of biogas obtained through biomethanation can generate around 2.14 KW of electricity with an efficiency of around 35%. Generally, the procedure for accomplishing AD can be classified into a four-stage procedure “(a) hydrolysis, (b) acidogenesis, (c) acetogenesis, and (d) methanogenesis” [67]. Of these four procedures, hydrolysis is the main challenging step in the technique for AD and is contingent on the category of feedstock. The biomass is often pre-treated to enhance the production of methane and reduce the digestion time [163]. While AD has been widely recognised, there are unique challenges in relation to the operations of biogas plants [164]. Factors such as operational instability, difficulty in controlling uncertain AD parameters and the conditions for real-time monitoring are some of the major challenges that are often faced, and this affects optimum biogas manufacture and inefficiency in product delivery [160]. The implementation of AI in the operations of biogas plants in the selected countries has considerably improved biogas production by guaranteeing a process that is efficient and stable [165]. In upholding the most ideal conditions for microbial activity, machine learning algorithms have been used to forecast AD parameters including pH, temperature, and organic loading rate.

AI-powered sensors applied in real-time monitoring can be used in the identification of any deviations, and this allows rapid responses to halt process interruptions [166]. AI-powered prediction boosts the pre-treatment and feedstock selection processes and improves the biogas yield [64]. Additionally, data analytics driven by AI also offer insights into patterns of performance and efficiency standards, thereby making decision-making simplified and streamlined [167]. A few examples of the AI technologies that provide a multifaceted solution in resolving the problems of SWM in the selected countries include machine learning, deep learning, and data analytics. More specifically, the AD-powered bio plants adopted in the selected countries deliver “de-fossilisation and decarbonisation”, which averts greenhouse gas emissions by converting organic wastes to: (a) renewable energy and (b) organic fertilisers, thereby decreasing the need for chemical fertilisers [168]. Furthermore, in comparison to other AI-powered WTE technologies such as incineration that emits CO_2, AD technology’s capability to change waste into valuable energy and organic nutrients without the risk of environmental pollution makes it the preferred option and a brilliant tool for realising a circular economy [168].

• Waste valorisation

Waste valorisation is another biochemical conversion procedure used to reduce the harmful impacts of waste (including solid waste) and is acknowledged to be one of the most effective strategies for sustainably managing solid waste. The practice of reusing, recycling or composting waste materials and transforming them into more useful products such as chemicals, fuels or other sources of energy is referred to as waste valorisation. Waste valorisation techniques substantially assist in eliminating the hazardous and detrimental effects of waste as well as support the creation of high-value-adding products from it [169]. Subramanian et al. [169] surveyed the main techniques of waste valorisation by employing life cycle assessment and techno-economic assessment of solid waste. The study found that valorisation of solid waste materials provides a potential solution for treating solid wastes and recovering energy as well as supporting environmental and economic sustainability.

Waste valorisation is crucial to the notion of circular economy and is now gaining considerable policy relevance within the selected countries’ waste management value chain. The concept of the circular economy is an alternative flow model for many economies that is recurring rather than linear. The policy and business-focused circular economy approach adopted by the countries studied emphasises products, components and material reuse, remanufacturing, repair and upgrading in the SWM value chain [170]. Innovative technologies in waste-to-energy, including waste valorisation, can offer alternate sources of energy that are both economically and environmentally sustainable [65]. Currently, the selected countries have implemented AI-powered systems in the treatment and valorisation of solid waste, which includes waste from food, paper and other packaging materials, construction, plastics, wood and biomass materials and has received the highest and utmost consideration [171]. For example, Japan has developed a food waste management system that tracks and predicts food waste using AI technologies. This innovative smart food-waste monitor has the capability to envisage and forecast what, when, and how much food is thrown away, provides actionable insights, assists in preventing food waste and averts the emission of CO₂, GHG emissions and water and environmental pollution issues [172].

Mechanical Conversion

• Landfilling equipped with biogas production

The International Solid Waste Association (ISWA) reported that landfilling continues to be a broad and the most widely used method for handling and disposing of solid waste, particularly waste from municipalities. This view is based on the fact that municipal solid waste (MSW) contains around 30–50% of organic matter, which is beneficial for power generation [173]. Generally, landfilling of MSW engenders biogas and leachate production, which can hypothetically resolve the problems arising from SWM. Besides, residuals from solid waste emanating from anaerobic fermentation can be reused as fertilisers [174]. Biogas from landfills contains mostly CH₄ and CO₂ produced via anaerobic fermentation of biomass materials such as manure, sewage sludge and municipal solid waste [173]. The whole process of manufacturing biogas can be categorised into three stages: (a) hydrolysis which is a procedure that breakdowns organic matter into smaller products facilitated by bacteria degradation, (b) acidification which is when an acid-producing bacteria changes biogas into acetic acid (CH₃COOH), hydrogen (H₂) and carbon dioxide (CO₂) and these bacteria are facultatively anaerobic and subsist under acid settings, and (c) methane formation which is when methane-producing bacteria decompose and decay compounds with a lower molecular weight [175].

Furthermore, the degradation of organic materials in landfills produces liquefied petroleum gas (LPG). LPG production is contingent on the degradation status of the waste material, moisture content and temperature, and can vary considerably among the different portions of the landfill. LPG also contains unpredictable amounts of other contaminants including nitrogen, oxygen, water vapour and sulphur [173]. According to ISWA [173], landfills are projected to discharge between 30 and 70 million tons of methane (CH₄) yearly, and this significantly contributes to global warming and the climate change crisis.

Consequently, the selected countries have implemented AI-powered systems to resolve operational parameters, the unpredictable and inconsistent composition of solid waste, the complexities and difficulties around the design of efficient and workable landfills, and specific social, economic, and environmental issues that create limitations in the applicability of sanitary landfill technologies in a cost-effective and ecologically sound manner [173]. For example, as part of its climate policy, Germany promotes the production of biogas via its Renewable Energy Act. In Denmark, LPG is well-managed to moderate emissions, while some of its landfills have installed and connected gas collection systems that facilitate energy recovery and mitigate the emission of methane [176].

Table 6. Summary of WTE technologies, applications, advantages and disadvantages.

WTE Technology	Application	Advantages	Disadvantages
Incineration	- Utilised in processes where energy is produced from the heat generated	- Currently established in industrial facilities and mature technologies for solid waste management	- The process can generate perilous and harmful substances that impact the environment - Incineration techniques involve high cost of capital
Pyrolysis	- Power generation is obtained and bio-oil is often utilised as a base material in the production of chemicals and other solvents	- Operating cost is low and is commercially viable - High fuel quality is generated - Ideal for carbonated waste materials - Reduction in fuel gas treatment	- Involves high cost of capital
Gasification	- Generates energy - Syngas realised can be utilised as fuel and raw materials to manufacture other specialty chemicals	- The fuel gas/oil generated can be utilised for several purposes	- Gasification systems are currently underdeveloped and lack flexibility. - The process is less competitive, and failure rates are high
Hydrothermal liquefaction	- The bio oil generated can be utilised in manufacturing chemicals	- The quality of products from liquefaction is high	- The operation processes vary considerably
Anaerobic digestion	- The production of gas through this process can be used for power generation - The digestate, which is the material left after anaerobic digestion, is rich in nutrients and can be used as fertilizers for crops.	- Higher quantities of methane and lower CO2 are produced when compared to landfilling	- The process is not suitable for solid waste with low organic matter contents
Waste valorisation	- Helpful in the conversion of waste materials into further beneficial products. - These include the production of chemicals, fuels and other sources of energy	- Provides alternative sources of energy that are both commercially achievable and environmentally sustainable	- Involves high capital and operating cost
Landfilling	- Landfill gas generated can be used for power generation	- There is a lower operational cost, and natural resources are re-instated into the soil	- Expensive to maintain and high operational cost - High land requirement - Can lead to contamination of soil and groundwater

Data adapted from Vyas et al. [12]; Shafizadeh et al. [150]; Hayashi et al. [172]; ISWA [173].

presents a summary of WTE technologies, applications, advantages and disadvantages.

3.5.2. Field Application Systems

• Automated waste collection operations

Waste collection is a key aspect of managing waste efficiently to attain a clean and sustainable environment [45]. The introduction of AI has substantially transformed solid waste collection processes through integration of algorithms that support the models for (a) smart bin system, (b) optimisation of waste routes, (c) dynamic procedures and scheduling, and (d) demand prediction and forecasting [45,177]. These AI-driven technologies are shaping the future of solid waste collection by improving the efficiency of operational activities, lowering budgets, and reducing impacts on the environment in the selected countries. AI-driven SWM systems have the capability to deal with challenges relating to scalability through leveraging capabilities to predict, automate, and analyse data [43,178]. AI technologies support the selected countries in efficiently assigning resources, improving the processes for solid waste collection, sorting and recycling. Through the process of automation, tasks such as route planning and collection operations, AI systems assist in enhancing the efficiency of operations and allowing waste management facilities to deal with higher quantities of waste without an equivalent upsurge in resources [177]. In addition, AI implementation effectively anticipates current and future patterns of waste generation, simplifying proactive and hands-on decision-making processes and allocation of resources to accommodate the ever-increasing volumes of solid waste. Most importantly, AI driven systems have been successfully implemented in the selected countries in addressing (a) traffic analysis in the transportation of solid waste where AI algorithms are used to analyse real-time traffic data to enable flexible modifications to routes which assists in averting delays and reducing idle time for waste collection vehicles, (b) fill level forecasting through the use installed sensors in bins whereby AI can correctly predict how filled they are and based on this information, and alter the scheduling of pickups so that they only happen when needed. This eradicates redundant trips and improves allocation of resources, and (c) pattern recognition through the use of sensors installed in bins where AI can recognise patterns, optimise the planning and scheduling of collections, which consequently saves time and improves allocation of resources [42].

• Optimisation of logistics and transportation

Route optimisation is another major phase in the solid waste value chain, and its objective is to reduce the time it takes to travel, fuel intake, and emissions from vehicles [46,63]. Route optimisation algorithms support the tracking of GPS, traffic conditions, current and past data collection, dynamic re-routing capabilities and weather predictions to tactically modify routes based on the latest requests and circumstances [44,179]. This flexibility empowers the selected countries to respond as quickly as possible to changes and optimise their transportation and logistics operations. Moreover, by incorporating AI algorithms such as patterns of traffic, population size and density, rates of waste generation, and proximity to the sources of waste generation, SWM systems can dynamically modify routes of transportation and collection schedules [44]. Further, integration of AI-driven systems in the selected countries not only ensures that each vehicle is on the most efficient and fastest route but also anticipates possible changes in the quantities of waste produced across different areas. This forethought permits practical and workable adjustments that support the smooth running of operations notwithstanding the intrinsic risks and volatility around urban environments. This proactive approach helps in maintaining the quality and efficiency of services even when exposed to risks and uncertainties in urban environments [44]. Thus, the selected countries are leveraging AI-driven systems, while reducing travel time, consumption of fuel, and emissions from waste vehicles to efficiently and timely deliver waste collection services in their respective communities [47]. AI-driven systems have also enabled the selected countries to improve efficiency of operations and activities, lower costs, and boost quality of service provided, thereby contributing to a better and more sustainable and resilient waste management system that makes a positive environmental impact [180].

• Predictive analytics for operational efficiency and cost saving

Predictive analytics employs AI technologies to analyse past and real-time data, facilitating the forecast of future patterns of solid waste generation and trends [50]. Predicting the trends of solid waste generation is critical for improving solid waste management systems’ effectiveness and sustainability. Through the prediction of future solid waste outputs, municipalities and companies involved in solid waste management in the selected countries can allocate resources more effectively, ensuring that facilities and personnel are prepared to handle the anticipated volume of solid waste [181]. Similar to the use of AI in the optimisation of logistics and transportation, factors such as population size and density, weather conditions, historical and current waste data and models for predictive analytics can estimate rates of solid waste generation, optimise and adjust collection routes, and assign resources efficiently. These AI technologies constantly learn from new data and enhance their accuracy over time [182].

Further, the implementation of predictive analytics helps solid waste management authorities in the selected countries to actively plan strategies for efficient solid waste management, optimise schedules for collection, and assign resources efficiently, thereby ultimately lowering costs and improving overall solid waste management effectiveness [142]. Likewise, AI-driven models for predictive analytics offer valuable and hands-on tools to solid waste management authorities in the selected countries to anticipate and mitigate the detrimental consequences of the climate change crisis on patterns of waste generation and rates of recycling [49]. Through the assessment of past data and ecological factors, these models can forecast changes in the trends of solid waste generation, such as variations in patterns of consumption and population growth affected by climate change [52,142].

In addition, predictive analytics can enhance solid waste management strategies and allocation of resources and allow authorities to adapt to shifts in waste streams and prioritise efforts in recycling in order to respond to climate change-related issues and challenges. For example, predictive models can identify areas that are vulnerable to increased solid waste generation owing to extreme weather conditions and events or natural disasters, permitting the selected countries to implement planned and targeted initiatives for solid waste reduction and recovery. Furthermore, AI-powered predictive analytics accelerate practical decision-making processes by identifying evolving trends and likely risks and uncertainties, which allows solid waste management authorities in the selected countries to come up with resilient and sustainable strategies to ease the climate change impacts on solid waste management systems [182,183].

• Real-time waste monitoring systems

Real-time monitoring is a critical aspect of SWM practices, ensuring that bins are not overfilled, which can result in increased contamination and spillage due to ineffective processes of waste collection [184]. Real-time systems for solid waste monitoring adopt AI technologies to gather and analyse data, and offer quick insights into solid waste generation, collection and disposal patterns in the selected countries [32]. These technologies use numerous sensors such as ultrasonic sensors, load sensors and GPS trackers, to capture data on the levels of waste in bins and containers, routes for collection trucks, and points of disposal. The integration of AI with real-time monitoring systems permits hands-on and immediate response to challenges relating to solid waste management, such as bins that are overfull, uncertain routes of collection, and optimised allocation of resources [184].

Through providing timely information, AI-driven monitoring systems reduce unnecessary truck journeys, optimise allocation of resources, and enhance overall operational efficiency of the entire solid waste management process [185]. However, the implementation of such systems at scale can lead to a number of technical challenges: (a) amalgamating varied sources of data and sensors into an integrated platform that requires a robust arrangement and interoperability standards [186], and (b) ensuring data integrity, reliability, correctness, trustworthiness, and security for effective data management protocols and privacy safeguards [187]. Dealing with these procedural issues and challenges is vital to unravelling the full potential of monitoring systems driven by AI in SWM and achieving considerable savings in costs and benefits in efficiency in the selected countries [188].

• Tracing and tracking illegal dumping

Detecting solid wastes that are illegally dumped has become an integral part of the processes involved in handling illegal dumping. The challenges around illegal dumping have long been a persistent issue in SWM and detrimentally affect communities and their surrounding environment, lead to social problems, and pose threats to human health [189]. As the volumes of solid waste generated are intensifying year after year, criminalities leading to illegal dumping are also growing [190]. Hence, governments of the selected countries have implemented AI-driven systems that can deal with the problem of illegal dumping as a critical solid waste management issue [190]. For example, the South Korean government engaged in the installation of AI-driven cameras in areas where there are high concentrations of illegal dumping to monitor and assign waste officers to patrol and guard. In addition, South Korean municipalities and other waste management organisations have also adopted various waste monitoring approaches to identify illegal dumping and penalise lawbreakers and offenders [191].

3.5.3. Municipal/End User Application Systems

• Smart bin systems

Prior to the development of smart bin systems, traditional garbage bins were mostly used for collecting solid waste. The waste workers often manually check to assess the level of trash in the bins. However, in recent years, the introduction of smart bin systems has received substantial consideration owing to their capability to optimise and support the processes of waste collection [32]. These systems have sensors fitted that have the capability to monitor the fill levels of bins in real-time and record the collected data for onward analysis through the use of AI algorithms to establish the ideal collection routes and schedules.

Recent advancements in smart bin systems include the integration of innovative sensors such as ultrasonic sensors and weight sensors, which offer additional precise data relating to fill levels [30]. Furthermore, smart bins are often equipped with connectivity features which allow for seamless communication and interaction between the waste bins and the organisation authorised to manage waste [192]. This connectivity creates opportunities for real-time monitoring, remote /distant management, and proactive maintenance and care of the bins [193]. Thus, the novelty in the implementation of smart bins in the selected countries lies in the ability to quickly and automatically monitor garbage fill level and ensure users are alerted in a timely manner. The information is received by sensors and conveyed through the network. Smart bin systems have the potential to raise the effectiveness of waste collection, reduce the spread of diseases and contamination, and enhance environmental sustainability [194].

• Automated Sorting Systems

AI-powered automated sorting systems are employed to recognise and sort solid waste items, and this depends on the nature and composition of the solid waste material [37]. These sorting systems utilise advanced sensors that examine both the physical and chemical characteristics of the solid waste items [195]. The data collected is analysed to determine the composition of the materials and is sorted into the correct categories of solid waste. The implementation of sensor-based sorting techniques significantly improves the accuracy, reliability and efficiency of the processes of sorting solid waste through automation and optimisation to identify and separate the various types of solid waste materials [38]. Through the automation of the sorting process, AI-driven systems can rapidly and correctly distinguish between recyclable materials and non-recyclable materials, organic waste, and contaminants, which ensures that recycled materials are sanitised [40,196].

This process leads to reduced contaminants in recycling streams and recyclable materials of higher quality, which increases the overall effectiveness of the recycling process. Nevertheless, a number of key challenges need to be resolved in order to effectively achieve this process. One major challenge is the complex nature of various streams of solid waste, which often contain numerous materials that are problematic and difficult to accurately categorise [39]. Additionally, the presence of contaminants and impurities in the various streams of solid waste can hinder the process of sorting and reduce the quality of materials that are recycled [36]. Navigating these challenges necessitates constant efforts in research and development to advance the accuracy and efficiency of AI technologies as well as investments in solid waste infrastructure and staff training to effectively adopt these technologies in SWM activities and operations. With the continual advancement in technologies, sorting systems driven by AI technologies play a crucial role in supporting the circular economy and enhancing sustainable SWM practices [41,197].

• Predicting/estimating consumers’ solid waste generation

In recent years, research on solid waste generation prediction has gained increasing attention; as a result, a number of AI technologies have been used to better predict the amount of solid waste generated by consumers. AI-driven systems are considered the best advanced systems to predict and estimate the generation of solid waste data since they have capabilities to uniquely input data, learn, and predict [4,198]. AI technologies regularly used in SWM include “(a) artificial neural networks (ANN), (b) support vector machines (SVM), (c) linear regression (LR), (d) decision trees (DT), and (e) genetic algorithms (GA)” [4]. Of these AI technologies, artificial neural networks have been extensively applied in the prediction of solid waste generation applications, owing to their valuable characteristics such as robustness, fault tolerance, and ability to describe and handle the complex associations between variables in multi-variable systems and are followed closely by support vector machines [4,199]. Through the analysis of past data and environmental factors, these technologies can predict changes in rates of solid waste generation and trends, such as variations in patterns of consumption and population growth, affected by the climate change crisis [4]. By correctly forecasting consumer demands in the selected countries, AI-powered SWM can optimise collection resources, efficiently assign staff and vehicles, and accordingly modify collection schedules [200]. Current records and past historical data in the selected countries were used to establish a gradient boosting regression model. For example, Johnson et al. [201] developed a prediction model for short-term solid waste generation in New York (USA) and attained an average accuracy of 88%.

• Robotics in solid waste management

In the solid waste recycling industry, robotics technologies are altering how we manage and deal with solid waste through the automation of labour-intensive tasks, which increase output [58]. AI-powered robots equipped with sensors, cameras and intelligent algorithms have the capability to handle tasks such as sorting, dismantling, and processing in the recycling process [57,202]. These robots can adapt to various shapes, sizes, and weights of solid waste items and use AI algorithms in the identification and sorting of recyclable materials [55]. In many instances in the selected countries, collaborative robots, also referred to as “cobots”, are engaged to work together with human operators, supporting and enhancing the safety and effectiveness of recycling operations [203]. The introduction of robotics and automation technologies in the recycling of solid wastes streamlines activities and operations, increases productivity, reduces the cost of labour, and enhances the overall rate of resource recovery [203]. The recent trends in solid waste recycling driven by AI focus on improving accuracy in identification and sorting, optimising the entire process of recycling, enhancing inspection and quality control, and implementing robotics and automation. These advancements have contributed to an industry that is thriving, more efficient and sustainable [203].

In the UK, Veolia waste management organisation installed a robotic arm equipped with AI at its waste facility in London, which has meaningfully upgraded the process of sorting efficiency and accuracy. Likewise, Cireco’s investment in its facility in London in sorting and processing solid waste using robots led to a considerable surge in the operational performance of its waste facility. Other Robotic sorting systems, such as those developed by The Recycleye Company, exemplified a remarkable advancement in solid waste sorting technologies. These systems adopted AI in detecting and sorting various materials with high precision. They can be distinguished from the traditional optical sorters, which frequently rely on near-infrared (NIR) technologies to identify solid waste materials on the basis of their spectral signatures [54]. Thus, ongoing research and development needs to focus on areas that require the integration of AI technologies with current infrastructure for solid waste recycling and data standardisation that ensures the scalability and cost-effectiveness of AI solutions [59]. Collaboration among researchers, organisations involved in the management of solid waste, technology providers, and regulatory authorities in the selected countries is essential to addressing these challenges to drive and support the widespread implementation of AI in solid waste recycling in the selected countries [56].

• Demand prediction

Demand prediction systems employ AI technologies in forecasting the rates of solid waste generation in different parts of the selected countries [204]. Demand prediction technologies play a crucial role in SWM by offering insights and awareness into future patterns of impending solid waste generation, allowing recycling facilities to optimise the allocation of resources and operational planning and forecasting [205]. The key benefits of these technologies include improved operational efficiency, better waste collection planning and scheduling, reduced waste of resources, and enhanced openness to changing patterns of demand [204]. Additionally, by anticipating fluctuations in waste generation, waste management authorities can confidently assign resources, adopt initiatives that specifically target waste reduction, and optimise the processes of recycling and disposal, which ultimately leads to solid waste management practices that are more cost-effective and sustainable.

Recent trends in demand prediction include the incorporation of numerous data sources and the adoption of innovative AI technologies [158,206]. Waste management organisations in the selected countries are combining sources of data from social media, other online platforms, and IoT sensors to capture real-time raw data and information on patterns of solid waste generation [206]. Accurate predictions to analyse the collected data are achieved using AI algorithms and deep learning models [207]. These models consider factors such as information on population density, demographic information, and past rates of solid waste generation to forecast current solid waste generation [206]. Through accurately forecasting demand, waste management organisations in the selected countries can optimise resources collection, efficiently and reliably assign staff and vehicles, and accordingly adjust collection schedules [207].

• Enhanced data management and decision-making

AI-powered systems expedite data-driven decision-making through the evaluation and analysis of huge amounts of collected data from several sources of monitoring solid waste [61]. Actionable insights are acquired and developed through AI technologies that process and analyse data that enhance decision-making in SWM processes [63]. For instance, data on rates of solid waste generation, rates of recycling and cost of disposal can assist in informing policy decisions, investments in infrastructure, and strategies for resource allocation. AI technologies adopted in the selected countries identify patterns, relationships, and trends in solid waste data that may not be clearly apparent to human operators, thus allowing for decision-making that is based on visible evidence for improved solid waste management practices that are efficient and sustainable [62].

• Maintaining quality control and inspection

Sustaining the quality and value of solid waste materials that are recycled is crucial in safeguarding their marketability and reuse. Quality control and inspection systems powered by AI make use of cutting-edge technologies such as sensors, computer vision, and machine learning to sense and eliminate environmental impurities and contaminants emanating from the recycled materials [208]. Combining AI systems that are filled with advanced sensor technologies such as high-resolution cameras and spectroscopic analysis ensures precise and reliable identification of contaminants present in the recyclable solid waste materials [209]. These AI-driven systems categorise materials based on criteria and benchmarks with predefined quality, guaranteeing precise sorting and improved efficiency. By continuously monitoring solid waste streams using near-infrared spectroscopy and computer vision, these systems can detect impurities and contaminants with high precision. This practical and real-time analysis permits instant discovery and removal of items that are non-recyclable, guaranteeing the cleanliness of solid waste materials for recycling [210]. Recent progressions in quality control systems powered by AI include the development and advancement of AI technologies that are sophisticated and capable of dealing with huge volumes of data and complex composition of materials. These technologies can recognise understated variations in materials and differentiate between items that are recyclable and non-recyclable with superior accuracy in solid waste when compared to traditional sorting techniques [211]. Moreover, improvements in relation to sensor technologies have led to the development of more compact, cost-effective, and versatile inspection systems that can easily integrate into existing recycling facilities. Overall, technologies for quality control and inspection driven by AI offer substantial potential in enhancing the cleanliness and purity of recycled materials, improving rates of recycling, and facilitating the selected countries’ transition towards a circular economy [209].

• Improving public health and quality of life

AI-powered systems used to improve sustainable SWM can help reduce solid waste without harming the environment and destroying wildlife and compromising quality of life, standard of living and human health [9]. In implementing AI technologies for intelligent and smart recycling, waste categorisation, and disposal in the selected countries, the process of managing municipal solid waste is reinforced and strengthened, leading to recycling approaches that are more viable and sustainable [4,212]. Hazardous solid waste such as that originating from solar panels and other electrical and electronic equipment (EEE), as well as medical waste, must be cautiously handled to reduce its detrimental impact on the quality of life and living conditions of residents in the communities. Intelligent bins exemplify how AI-powered systems can help reduce the impact of inappropriate disposal of solid waste to achieve sustainable SWM practices [4]. Through embracing diverse SWM approaches in the selected countries, AI-driven technologies can appropriately accommodate various demographic groups, improve environmental planning and development, and optimise the efficiency, accuracy and performance of solid waste management systems [4].

This section has discussed practical application areas of AI in solid waste management in the selected countries. Artificial intelligence is applied in solid waste management across the global cities studied to increase recycling and recovery rates, optimise collection routes, power robotic sorting facilities, monitor bin fill levels, and maximise waste-to-energy plant efficiency as noted above. The selected countries utilise these AI technologies to increase operational efficiency, lower carbon footprints, reduce manual labour costs, achieve circular economy goals and national recycling targets.

Table 7. Comparison of AI application areas, representative projects, deployment scale, benefits and limitations in the selected countries.

Country	AI Application Areas	Representative Projects	Deployment Scale	Benefits	Limitations
Austria	Austria applies AI in SWM in waste through computer vision on collection trucks for real-time contamination tracking, AI-driven fire prevention and material sorting in recycling facilities, track municipal refuse, waste-to-energy as well as smart sensor networks for dynamic route optimisation.	Key initiatives of Austrian AI-Waste projects include: PreZero’s AI sorting plant, RecAL aluminium recycling project, and Austria’s smart waste collection pilots	Overall deployment is progressing. For example, the PreZero facility in Sollenau stands as one of Europe’s largest and most complex deployments, utilising automated process control for up to 20 material fractions	- Improves and advances operational efficiency - Increases productivity, - Amplifies recycling rates - lowers costs - Sustainable waste management - Contamination reduction	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
Germany	Germany integrates AI across its waste management sector to optimise waste sorting, enforce stringent household recycling policies, guidelines and laws, as well as to streamline municipal logistics. Key application areas include computer-vision recycling plants, sensor-based waste monitoring, automated contamination tracking, and smart collection routes.	Key initiatives include robotic sorting plants, smart garbage trucks project, AI-powered smart scanners project, AI-powered industrial and construction recycling plants	Deployment is progressing and includes municipal smart-bin logistics, deep-learning sorting systems, AI-powered material recovery facilities and predictive route optimisation	- Improves and advances operational efficiency - Increases productivity, - Amplifies recycling rates - lowers costs - Sustainable waste management - Optimises collection logistics - Supports strict national circular economy goals	- Infrastructure requirements and high costs - Availability and quality of data - Ethical concerns
Denmark	Denmark uses AI and advanced robotics technologies to optimise waste sorting, streamline collection logistics, waste-to-energy, censored based waste monitoring and advance its circular economy objectives	Key initiatives include NOMI4S and Solum’s construction recycling plants, AI-driven E-waste recycling plants led by Danish Technological Institute (DTI), Smart municipal collection routing project, Copenhagen RGS Nordic Robotic Facility and Ento municipal energy waste projects	Denmark deploys AI in waste management at a targeted, growing, commercial and early-stage municipal scale routing and administrative automation	- Improves and advances operational efficiency and logistics - Lowers costs - Sustainable waste management - Higher recovery rates - Less landfill waste - Climate goals-supports Denmark’s strict green transition targets	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
United States	AI in SWM in the United States is used in material recovery facilities, optimisation of municipal truck routes, waste-to-energy and in reducing contamination in recycling streams through the use of computer vision, robotics, and predictive analytics	United States AI-waste projects mostly focus on automated recycling sorting, municipal solid waste diversion, waste-to-energy, commercial food waste reduction using computer vision and machine learning. Examples of key projects include: Recology Recycling Plant in Seattle, Washington, AMP ONE in Portsmouth Virginia, and Alameda County Industries (California) Winnow Solutions (US commercial food tracking sites)	Key deployment areas include Material recovery facilities (MRFs), Route optimisation, Smart monitoring and Waste-to-energy	- Operational efficiency and logistics - Smarter collection and logistics - Lower labour costs - Fewer injuries - Data tracking - Sustainable waste management - Meeting sustainability goals and ESG Requirements	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
United Kingdom	AI in SWM in the UK uses computer vision, machine learning, and robotics technologies to streamline operations from sorting facilities to street-level collection. Major players include: Renewi Plc, Viridor Limited, FCC Environment, Mick George, DS Smith, and Reconomy, among others	UK AI waste management projects focus on automated sorting, regulatory compliance, and food redistribution. These projects leverage computer vision and cloud analytics to optimise recycling and curb surplus waste. Examples include Greyparrot AI Analyzers deployed by major UK waste operators like Biffa and FCC Environment; Veolia UK AI Sorting Lines and Innovate UK BridgeAI Food Redistribution Pilot plants	The deployment scale of AI in United Kingdom waste management is expanding rapidly from facility-level sorting to official regulatory compliance. Deployment areas include: Material Recovery Facilities (MRFs), Regulatory Reporting; Smart Bin and Fleet Logistics, and Recycleye Computer Vision	- Data-Driven decision making and reporting - Cost reduction and operational efficiency - Meeting sustainability goals and ESG requirements - Cost reduction - Sustainable waste management - Higher recycling and recovery rates	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
Japan	Japan applies AI to waste management to manage stringent recycling rules and labour shortages, strict circular economy goals, and dense urban centres. Key areas include waste-to-energy, automated sorting with computer vision, optimisation of collection logistics, forecasting maintenance of processing plants, Environmental Monitoring, and in reducing commercial food waste	Japan combines government funding, robotics, and AI to tackle municipal waste, marine debris, and food loss. Key initiatives include: EII recycling infrastructure projects; advanced municipal sorting lines, Takanome by Pirika, Inc. projects, NEC & Convenience Store Demand Forecasting project	AI applications in SWM are deployed across municipalities and major transit hubs in Japan. Major operators, such as those in Tokyo and Central Japan Railway, have implemented AI-enabled sorting plants, IoT sensors, and robotic arms to optimise recycling and waste-processing workflows.	- Data-driven decision making and reporting - Cost reduction and operational efficiency - Meeting sustainability goals and ESG requirements - Increase recycling and recovery rates - Tackles acute labour shortages and strict recycling mandates. - Optimises municipal operations through automated robotic sorting, real-time IoT bin monitoring, and dynamic collection route planning, - Boosting recycling accuracy - Lowering fuel emissions	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
Singapore	Singapore applies AI in SWM to optimise collection routes, automate material sorting, enhance recycling purity, and forecast food waste. These smart systems reduce fuel use, lower carbon emissions, and support the nation’s zero-waste goals under the Smart Nation initiative	Singapore uses AI and robotics to optimise waste management, sorting, and recycling. Key projects include multi-spectral AI plastic-sorting project, AI-powered robotic sorting lines, automated e-waste processing plants, and IoT & Sensor-optimised collection plants	Singapore is deploying AI-driven SWM at a city-wide scale under its smart nation initiative by integrating IoT fill-level sensors, route optimisation algorithms, and automated recycling facilities to manage dense urban sanitation, reduce collection trips, and achieve sorting accuracy. Core areas of deployment include smart bins and IoT sensors, route optimisation, AI-enhanced sorting facilities, carbon and efficiency targets and national recycling goals	- Advanced material sorting - Contamination reduction - Smart bin monitoring - Route optimisation - Contamination control - Extended landfill life	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
Switzerland	Switzerland applies AI to enhance recycling accuracy, reduce waste recycling costs, and prevent food loss through the use of computer vision sorting systems and machine learning platforms tracking disposal streams in real-time	Swiss waste projects use AI to optimise waste sorting, waste-to-energy, track food loss, and streamline municipal collection. Notable examples include real-time material recovery systems such as SENS eRecycling project, Burgdorf Smart Waste Collection plant, and EbiMIK high capacity recycling plant	Switzerland deploys AI in waste management at a targeted, growing municipal and industrial scale. Key AI deployment areas include: urban cleanliness and litter mapping, facility sorting and recycling copilots plants	- Higher sorting accuracy - Reduced incineration volume - Lower contamination - Optimised efficiency and logistics - Improve recycling precision - Increase recycling and recovery rates	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
South Korea	South Korea applies AI heavily to municipal solid waste and plastic recycling, utilising deep-learning sorting robots, smart infrastructure, and predictive facility management to meet zero-waste and circular economy goals	Key projects include SuperBin (Nephron) project, AETECH (Airo-MRF) project, Government Textile AI Project and Nuvilab Food Scanners project	South Korea is scaling up AI in waste management through municipal robotic sorting centres, smart automated collection networks, and vertically integrated digital tracking. Key areas of deployment include: AI sorting & recycling robots, smart collection & bins, predictive plant operations, facility monitoring & safety and waste generation forecasting & logistics	- Higher sorting accuracy - optimised collection logistics - Reduce landfill and emissions - Smart infrastructure support	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns
The Netherlands	In the Netherlands, AI is transforming waste management to meet ambitious national circular economy goals. Key application areas include waste-to-energy, AI-powered high-purity recycling and material sorting, computer vision for public cleanliness monitoring, logistics and route optimisation, and food waste prevention analytics	The Netherlands uses AI and smart data systems to tackle municipal waste, plastic sorting, food waste, and wastewater treatment. These projects boost recycling precision, reduce carbon output, and optimise resource recovery. Key projects include High-Capacity PET Recycling plant, Circular Economy Research project and RWZI Tilburg DARROW Project	The deployment scale of AI in SWM across the Netherlands is progressing speedily from localised municipal pilots to high-capacity industrial implementations. Dutch recycling and engineering enterprises are deploying AI-driven computer vision, sensor fusion, and analytics software to automate complex sorting and optimise municipal collection networks.	- Advanced material sorting - Contamination reduction - Smart bin monitoring - Route optimisation - Contamination control	- Infrastructure requirements and costs - Availability and quality of data - Ethical concerns

Data adapted from Nwokediegwu & Ugwuanyi [42]; Shokrollahi [89]; Andersen et al. [108]; Alzamora & Barros [123]; Uche et al. [213]; Waste management world [214]; Brunn [215]; Magazzino et al. [216]; Danish technological institute [217]; Aberger et al. [218]; MVSC Research Reports [219]; Voice of America [220]; Veolia [221].

presents a comparison of AI application areas, representative projects, deployment scale, benefits and limitations in the selected countries.

4. Analysis of the Results

4.1. Summary of Results, Benefits and Challenges of Implementing AI in the Selected Countries

4.1.1. Summary of Results

The analysis of the results shows that the implementation of AI in SWM and recycling in the selected countries is not only driven and motivated by concerns for environmental sustainability but also by factors relating to economic and social sustainability as described in Section 3.3. AI-powered systems in SWM have the potential to lower costs, increase the rates of resource recovery, predict the generation of solid waste and establish new and innovative recycling opportunities in the countries studied. As the mandate to sustainably manage solid waste and adopt circular economy practices and models continues to grow and mature in the selected countries, recycling solutions driven by AI are poised to play a fundamental role in the achievement of these objectives [204]. Further, the implementation of AI in the selected countries enables valued resources from solid waste streams to be extracted with better precision, thereby accelerating the expansion and growth of innovative recycling techniques and approaches and the manufacture of recycled products of higher quality [222]. The results of the review show that the selected countries have made significant progress in the implementation of AI in their SWM systems. Whereas the selected countries have made remarkable advancement in various areas of the SWM value chain and systems, the extent of automation can be categorised into: (a) waste to energy systems, (b) field application systems and (c) municipal /end user application systems to efficiently and sustainably manage solid waste, while leveraging on the benefits and mitigating challenges and risks encountered in AI implementation in SWM as discussed in the following sections.

4.1.2. Benefits

As noted earlier, implementing AI in SWM systems improves and advances operational efficiency, increases productivity, amplifies recycling rates and lowers costs through smart and data-driven automation. According to OECD [16], the application of robotic technologies improves the sorting of solid waste by up to 99% accuracy, optimises the collection of solid waste, enhances solid waste tracking and illegal dumping while improving sustainability and lowering the emission of carbon when compared to the traditional methods of SWM.

Consequently, the selected countries have taken advantage of the rising demand for sustainable products and services and explored the introduction of innovative business models and market opportunities in recycling solid waste. Furthermore, the implementation of AI-driven technologies generates new streams of revenue through data-driven awareness and predictive analytics that empower the selected countries to offer personalised solutions and value-added services to their consumers in their communities. Generally, integrating AI into solid waste recycling processes not only progresses economic competitiveness in the selected countries but also facilitates innovation and development while placing these countries as key global players and performers in the quest to transition and switch towards a circular economy [16]. AI-driven robotics and automation technologies deliver noteworthy capabilities to simplify the processes of recycling solid waste in numerous ways.

For example, AI-driven robotics can systematise and automate repetitive and labour-intensive tasks such as sorting, segregating and processing diverse categories of solid waste materials. Using AI algorithms and sensor-based technologies, robotic systems pinpoint and classify numerous recyclable solid waste materials with high accuracy and efficiency, thereby decreasing the need for physical and manual labour and minimising errors associated with humans. It can optimise the workflow and productivity of recycling facilities by improving the speediness and accuracy of the processes of material handling. This amplified efficiency and effectiveness transforms to greater productivity, lower operational budgets and costs, and improved performance of the entire recycling operations [223]. Further, the selected countries have encouraged the development of solid waste recycling platforms driven by AI technologies that support unified integration, collaboration and partnerships among various stakeholder groups in the solid waste recycling value chain. These platforms accelerate the sharing of data, awareness, consciousness, knowledge, and best practices within SWM organisations, thereby promoting and upholding a cleaner and more efficient environment for solid waste recycling. The platforms utilise AI technologies to provide insights on predictive analytics, optimise allocation of resources, and allow recycled solid waste materials to be traceable and visible [16].

Consequently, the countries studied have successfully implemented AI-powered systems in accomplishing sustainable SWM practices. Some of the key benefits of implementing AI in SWM in the selected countries include: (a) data-driven waste collection, which analyses data from several sources such as sensors and GPS trackers available on waste trucks, and historical patterns of collection to optimise collection routes, (b) robotic technology and image recognition utilised for sorting which locates and classifies various forms of solid waste disposed of into bins. Robotic technology lets diverse types of garbage at the recycling facilities to be methodically and automatically sorted, increasing recovery of recyclable materials and decreasing the risk of contamination and pollution, (c) waste-to-energy conversion which encompasses the application of AI powered systems to transform waste into energy by spontaneously analysing the solid waste category and dealing with factors such as temperature, pressure, and structure/composition of the solid waste materials”. Examination and analysis of these factors lead to efficient generation of energy and minimisation and/or elimination of physical environmental impacts; (d) smart bin technology which integrates the concept of AI into solid waste bins, permitting easy detection of fill levels and notifying the vehicles used for garbage collection. This process guarantees that bins are only emptied when required, thereby reducing redundant collections and improving operational efficiency; (e) behavioural analytics, which examines and scrutinises the patterns and behaviours of users who dispose of their waste and those who are involved in illegal dumping. These behavioural patterns assist the selected countries in understanding specific communities’ habits for disposing of their garbage. The behavioural pattern of the individual communities provides insight into incentive programs available and educational workshops and events needed for appropriate practices of waste disposal; (f) predictive analytics which screen and monitor the real condition of solid waste management equipment, including machines used for sorting and smart waste collector bins, in real-time. This maintenance forecast is required prior to equipment breakdown, moderates interruption and downtime and decreases operational budgets and costs, and (g) integration of AI with IoT devices and blockchain technology, which integrates AI with IoT devices and blockchain technologies to successfully preserve transparency and traceability in the overall processes of solid waste management [16,89].

4.1.3. Challenges

While AI has vast and infinite possibilities implemented in transforming and advancing SWM practices in the selected countries, challenges still exist that need to be addressed to further enhance the implementation of AI in SWM [155]. These challenges are discussed below:

• Infrastructure requirements and costs

The implementation of AI technologies in SWM requires significant initial costs and infrastructure requirements [224]. AI systems frequently need hardware, software, and computational resources that are specialised in dealing with huge datasets and multifaceted and complex algorithms. Additionally, integrating AI technologies with current SWM infrastructure breeds issues and challenges in relation to compatibility, scalability, and cost-effectiveness. Overcoming these challenges in the selected countries requires AI solutions that are cost-effective and easily integrated into the prevailing SWM systems [56,225]. Furthermore, collective and collaborative efforts as well as teamwork between providers of technology, waste management agencies, regulators and policymakers in the selected countries can assist in identifying strategies and approaches to reduce the costs of implementation and streamline requirements for infrastructure [226].

• Availability and quality of data

The accessibility, integrity and quality of data are among the key challenges faced in the implementation of AI-driven SWM practices in the selected countries [186]. This is because AI algorithms rely greatly on huge and varied datasets in order to train and make estimates and predictions that are accurate and reliable [227]. Nonetheless, data gained from SWM are often inconsistent, disjointed, or inadequate, which can considerably have an impact on performance outcomes [195]. Owing to the numerous stakeholder groups that are involved in the entire management of solid waste, the lack of data integration among diverse stakeholders in the waste management value chain can hamper the development, progress and implementation of effective AI solutions [228]. In addition, data collection and monitoring systems must essentially be robust and trustworthy to guarantee precise and high-quality data to allow AI algorithms to produce insights that are meaningful and significant. To overcome these challenges, collaborative and partnership relationships between researchers, waste management agencies, providers of technology and other stakeholders are necessary to standardise data formats, improve data platforms, and encourage data-sharing practices [229].

• Concerns about privacy and security

The implementation of AI-powered systems in SWM raises a number of concerns relating to privacy and security. AI-driven systems frequently necessitate access to sensitive data, such as patterns of solid waste generation, collection routes, and consumer behaviour [230]. The protection of the privacy and security of information is crucial to upholding public trust and compliance with regulations on data protection. To manage and deal with these issues, there is a need for the selected countries to strike a balance between the benefits relating to AI-powered SWM systems and privacy and security considerations [231]. In addition, techniques that boost privacy, such as data anonymisation and encryption, can be adopted. Again, striking the right balance between accessibility and quality of data in AI algorithms and safeguarding the privacy and security of data is a significant challenge that necessitates constant attention and consideration [232].

Another main concern is the integration of systems driven by AI and sensor networks in SWM systems. When it is not appropriately done, there is an increased threat of cyberattacks and data breaches, which can potentially compromise data security, confidentiality, and availability [233]. Robust data protection and security protocols are the two most vital measures that must be applied to successfully mitigate and minimise these concerns without compromising the efficiency and effectiveness of AI systems. These measures include “(a) encrypting sensitive data, (b) implementing access controls and authentication mechanisms and (c) regular updates to reduce exposures and vulnerabilities” [23]. Thus, by making concerns for privacy and security measures a priority while designing and deploying AI systems, organisations with responsibility to manage solid waste in the selected countries can encourage and promote trust among stakeholders and achieve responsible SWM practices deeply rooted in the ethical use of AI technologies [230].

• Ethical concerns

The implementation of SWM systems powered by AI in the selected countries presents potential for algorithm biases, data privacy issues and other ethical concerns that require cautious and careful consideration [234]. For example, AI algorithms could potentially introduce biases during the process of decision-making, which could unintentionally propagate inconsistencies and discrepancies or produce unintended outcomes. These concerns are typically relevant and applicable in SWM, where equitable and justifiable allocation of resources and fair treatment of communities are essential. Overcoming these algorithm biases involves transparency and accountability, which are crucial in SWM systems driven by AI [230]. Transparency permits stakeholders to carefully and prudently assess all processes for decision-making and recognise and resolve algorithm biases as necessary. In addition, guidelines and frameworks for ethical processes and procedures can be established in the selected countries to deal with how AI is responsibly used in SWM [235]. These frameworks maintain advocacy for fairness, equity, and inclusivity in the deployment of AI systems, and ensure that algorithms are operated within ethical limits and boundaries. By observing these transparent practices and ethical guidelines, the selected countries and other countries can promote and uphold the fairness and transparency principles and frameworks while harnessing the potential and capabilities of AI in SWM practices.

4.1.4. Future Directions of AI in Solid Waste Management in the Selected Countries

The future of AI in SWM in the selected countries is expected to continually present significant modifications and broad changes and create novel ideas that alter how solid waste is managed and recycled. AI-powered systems support productivity, efficiency, sustainability and optimisation of resources in SWM. Four major areas where AI is anticipated to make significant imminent contributions in SWM are “(a) advancements in AI technologies in SWM, and (b) AI integration into several SWM processes, (c) collaboration and knowledge sharing, and (d) policy and regulatory frameworks”. Other areas include integrating AI with IoT/blockchain and advancing deep learning/natural language processing. As AI implementation in SWM continues to evolve, several future directions and opportunities are emerging.

• Integration and consolidation of AI and Internet of Things (IoT)

The integration and consolidation of AI with Internet of Things (IoT) is anticipated to play a fundamental role and significantly contribute to the future of SWM [236]. Devices from IoT such as sensors and smart bins produce enormous quantities of real-time data that are utilised by AI algorithms to make decisions and optimise SWM processes [237]. The association between AI and IoT technologies facilitates and advances the establishment of intelligent waste management systems that can autonomously and independently plan, monitor, assess, and optimise the processes of waste collection, sorting, recycling, and disposal [238]. This integration supports SWM operations that are smarter, with predictive maintenance of infrastructure for SWM, and enhanced monitoring and control in real-time.

• Improvements and expansion in machine learning and deep learning

Machine learning and deep learning algorithms are leading in the advancement and expansion of AI and are predicted to constantly evolve in the SWM context and environment [44]. Machine learning algorithms, such as (a) “decision trees, (b) random forests, and (c) support vector machines”, are currently improved to boost the categorisation of waste, as well as for sorting and predictive modelling capabilities [239]. Deep learning algorithms, on the other hand, which are mostly (a) “convolutional neural networks (CNNs) and (b) recurrent neural networks (RNNs)” are revolutionising SWM processes such as “(a) image recognition, (b) material identification, and (c) quality control” [240]. The future of AI in SWM is expected to witness advancements and expansions in machine learning and deep learning algorithms that can deal with the multifaceted and complex solid waste streams, adjust to SWM conditions that are flexible and dynamic, and take advantage of large-scale data to facilitate improved capabilities for decision-making and optimisation [241].

• Collaboration, teamwork and knowledge sharing

The non-existence of collaboration and knowledge-sharing among diverse stakeholders in the solid waste recycling value chain is exposed to numerous key obstacles that deter advancement. A major challenge is the absence of unified conversation and interchange/exchange of information among the various stakeholders such as: “(a) waste policy makers, (b) waste regulators, (c) technology providers, (d) manufacturers, (e) retailers, (f) local governments and (g) consumers” [242]. Besides, fears regarding data privacy, integrity, ownership, and security frequently discourage stakeholders from sharing and exchanging useful data and insights. Moreover, the absence of centralised platforms for data-sharing and knowledge conversations among the stakeholders further intensifies and deepens fragmentation within the solid waste recycling value chain [243]. While AI-driven systems provide favourable resolutions to overcome these obstacles, leveraging capabilities for cutting-edge technologies and data analytics can accelerate safe and well-organised data sharing, particularly when blockchain technologies are employed, which ensures confidentiality and compliance with regulatory requirements [244].

When stakeholders embrace platforms driven by AI for exchanging knowledge and sharing data, they can easily: (a) reach out and connect with one another and build and establish collective intelligence, (b) fast-track the development and growth of transformative solid waste management solutions, and (c) drive continuous improvement across the solid waste recycling value chain. According to Kurniawan et al. [245], responsible and accountable implementation of AI in SWM facilitates collaboration among stakeholders, mostly waste management agencies, providers of technology and policymakers. Waste management agencies can offer substantial insights into their exact problems/challenges and the needs and requirements of the industry, while technology providers can provide up-to-date outcomes for leveraging AI. Policymakers, on the other hand, can create a regulatory environment that is supportive and promotes ethical standards, compliance with regulations and sustainable implementation of AI in SWM. By partnering together, stakeholders can share their expertise, resources, and best practices while encouraging innovation and driving favourable outcomes for the environment and society [246]. Cooperation and teamwork among the various stakeholders can simplify the adoption of responsible, transparent and accountable AI-powered systems that promote trust among stakeholders and the general public, which leads to more efficient and environmentally sustainable waste management practices.

• Policy and regulatory frameworks

Developing regulatory frameworks, industry guidelines, and standards is crucial to ensure responsible and ethical use of AI in SWM. These frameworks and standards are obligated to address concerns in relation to privacy, security and transparency of data. Additionally, regulatory frameworks can support strategies for responsible solid waste management practices and ensure that AI algorithms appropriately rank environmental sustainability and social equity [247]. Governments and regulatory agencies have a fundamental contribution to make in modelling and shaping these frameworks to create a supportive environment that favours innovation while protecting the interests of the communities [248]. Further, by joining forces with industry stakeholders and policymakers to come up with regulations that balance the “need for innovation with the protection of ethical principles and societal values”. Furthermore, an emergent opportunity in AI-powered solid waste management systems includes implementation of “(a) natural language processing (NLP) for solid waste data analysis, (b) AI-driven robotics for waste sorting and processing, and (c) transportation and logistics” [4]. These innovative technologies have the potential to transform SWM by enhancing efficiency, lowering operational costs, and reducing detrimental impacts on the environment. As AI endures and continually advances, governments and regulatory agencies in the selected countries must consistently remain proactive, ensuring that policies and standards are updated to address evolving uncertainties, risks and opportunities in AI-driven SWM practices.

5. Conclusions and Recommendations

5.1. Conclusions

The implementation of AI technologies in several areas of SWM in these selected countries to enhance sustainable SWM practices was examined. The study provides a review of progress made, assesses the level of AI automation in the various areas of the SWM value chain, explores the benefits and challenges and offers best practice recommendations on how the efficiency of resources can be optimised to facilitate economic, environmental and social outcomes in other countries. The study employed secondary data gained through a review of articles from past and current studies on SWM and AI implementation from 2005–2025. Based on the literature review and content analysis, the selected countries have made considerable progress in implementing AI-powered systems across the SWM value chain. These applications demonstrate AI’s potential to improve efficiency, sustainability, recycling, waste collection, and resource recovery, contributing to cleaner and smarter cities.

However, challenges remain, including data availability and quality, privacy, high implementation costs, inadequate infrastructure, and ethical concerns. Addressing these barriers requires collaboration among governments, waste management agencies, private companies, technology providers, and researchers.

Overall, AI has significant potential to support more sustainable and resource-efficient SWM through applications such as smart bins, waste-sorting robots, predictive waste tracking, route optimisation, and waste generation forecasting. This study focused on ten leading countries in SWM. Future research could examine AI implementation in a wider range of developed and emerging economies using quantitative or mixed-method approaches to further understand its contribution to sustainable SWM practices. Other limitations of this study include the inability to include articles from non-English language sources, geographic representation imbalances, the lack of AI algorithmic details and the inability to measure national implementation based solely on the literature.

5.2. Recommendations

This study recommends and suggests the way forward as follows:

-

Strengthen SWM data availability and quality.

• Governments and waste management organisations should establish standardised systems for collecting, storing, sharing, and validating solid waste data to improve the accuracy of AI predictions and decision-making.

-

Improve data security and privacy.

• Organisations should implement encryption, access controls, authentication, data anonymisation, and regular security assessments to protect sensitive information and build stakeholder trust in AI-powered SWM systems.

-

Invest in AI infrastructure and R&D.

• Governments, industry, and academia should increase investment in smart bins, IoT sensors, automated sorting systems, digital connectivity, and R&D to develop AI solutions suited to local SWM challenges.

-

Develop skills and capacity.

• Waste management organisations should provide continuous training and capacity-building programmes for SWM professionals to improve their understanding and effective use of AI technologies.

-

Promote responsible and inclusive AI implementation.

• Policymakers and SWM stakeholders should address ethical issues, algorithmic bias, privacy, employment impacts, and environmental fairness while involving waste workers and communities in AI implementation.

-

Establish pilot projects, regulations, and performance standards.

• Governments and waste management organisations should implement pilot projects before large-scale adoption and establish clear regulatory frameworks and measurable indicators covering costs, recycling rates, collection efficiency, resource recovery, and environmental outcomes.