7

1. Executive Summary

The transition from a linear “take–make–dispose” model to a circular economy has emerged as a priority in addressing global challenges related to resource depletion, waste generation, and climate change. Despite growing policy attention and industry commitments, the global economy remains largely linear, with material recirculation rates still low (Circularity Gap Report, 2024). A key constraint underpinning this gap is not only technological or economic, but informational—circular systems require high levels of data visibility, coordination, and optimization across complex value chains.

Artificial Intelligence (AI) and data science are increasingly recognized as enablers for scaling circularity. By transforming fragmented and opaque material flows into structured, actionable data, these technologies enable more efficient resource use, extend product lifecycles, and improve recovery and recycling outcomes. AI-driven capabilities—such as predictive analytics, computer vision, and optimization algorithms—support circular interventions across the entire lifecycle, from product design and material innovation to intelligent supply chains and advanced waste management systems (World Economic Forum, 2026).

One of the most transformative developments in this space is the emergence of digital infrastructures such as Digital Product Passports (DPPs), which enable traceability and data standardization at the product level. Combined with technologies such as IoT (Internet of Things) and distributed ledgers, DPPs provide the data backbone necessary for AI systems to operate effectively in circular ecosystems. These developments are aligned with broader global efforts toward a “twin transition” that integrates digital innovation with sustainability objectives (UNECE, n.d.).

However, the deployment of AI in circular economy systems also presents challenges. These include data silos, lack of interoperability standards, high initial investment costs, and the environmental footprint of digital infrastructure itself. Addressing these barriers requires coordinated action across governments, industry, and international organizations, including the development of regulatory frameworks, investment in digital infrastructure, and the establishment of global data standards.
This report examines how AI and data science can accelerate the transition toward circular systems by addressing core informational bottlenecks. It provides an analysis of key applications across value chains, highlights emerging global best practices, and assesses opportunities and challenges within the Arab region. The report concludes with strategic recommendations aimed at enabling scalable, data-driven circularity through policy innovation, technological adoption, and cross-sector collaboration.

2. Introduction: Circularity as an Information Problem

The global transition toward a circular economy is increasingly recognized as essential for addressing the interconnected crises of resource depletion, environmental degradation, and climate change. However, despite growing policy momentum and corporate commitments, the global economy remains overwhelmingly linear. Only a small fraction of extracted materials are cycled back into the economy, with the vast majority still following a “take–make–dispose” trajectory (UNDP, 2023). This persistent imbalance highlights a structural limitation in how modern economic systems manage and retain material value.

At the core of this challenge is not only a physical inefficiency in production and consumption systems, but a fundamental information gap. Circularity depends on the ability to track, understand, and optimize material flows across complex and fragmented global value chains. In practice, however, data about materials is often incomplete, inconsistent, or entirely unavailable once products move through multiple actors, geographies, and transformation stages (UNEP, 2019).
This lack of visibility creates what can be described as “opaque” supply chains, where neither origin, composition, nor end-of-life pathways are reliably known. This information asymmetry represents a key bottleneck to scaling circular systems. Without granular and interoperable data, it becomes difficult to enable reuse, remanufacturing, recycling, or high-value recovery of materials.

As a result, economic systems remain locked in linear inefficiencies, where valuable resources are prematurely discarded rather than reintegrated into production cycles. This reinforces the need to shift from traditional industrial models toward data-enabled circular intelligence systems, where material flows are continuously monitored, analyzed, and optimized.

Within this context, artificial intelligence (AI) and data science represent a structural shift rather than a marginal improvement. These technologies enable the transformation of fragmented and unstructured data into actionable insights, allowing for predictive, adaptive, and system-wide optimization of resource flows (UNEP, 2024). The transition from linear inefficiencies to circular intelligence is therefore fundamentally a transition from low-information systems to high-information systems, where decisions are driven by real-time data and system-level visibility.

The objective of this report is to examine how AI and data science can enable this transformation. Specifically, it explores how these technologies can address the core limitations of current circular systems by improving material traceability, enabling predictive resource management, and optimizing lifecycle decisions across supply chains. By doing so, the report positions AI not merely as an operational tool, but as a foundational enabler of system-level circular optimization.

3. The Data–Circularity Nexus (Analytical Framework)

The transition from linear to circular systems is not solely a shift in material flows, but fundamentally a transformation in how information is generated, shared, and utilized. As established in the preceding section, current economic systems are characterized by fragmented and low-quality data, limiting the ability to track and optimize resource use. In contrast, circular systems require high levels of data granularity, transparency, and interoperability to function effectively.

In this context, the transition from linear inefficiencies to circular intelligence is therefore fundamentally a transition from low-information systems to high-information systems, where decisions are increasingly driven by real-time data and system-level visibility. Artificial intelligence (AI) and data science play a central role in enabling this shift by converting complex and fragmented datasets into actionable insights that support circular decision-making.

3.1 Circular Systems as Data Systems

At their core, circular economy systems can be understood as data systems. Every physical flow of materials—whether raw inputs, intermediate goods, or end-of-life products—is accompanied by an underlying flow of data. This includes information on material composition, origin, ownership, usage patterns, and end-of-life pathways.

In linear systems, these data flows are typically incomplete or disconnected, resulting in limited visibility across the product lifecycle. However, circular systems depend on the ability to continuously capture, update, and share data across multiple actors within a value chain. This enables more informed decisions regarding reuse, repair, remanufacturing, and recycling.

The importance of data in enabling resource efficiency and lifecycle management is emphasized by the United Nations Environment Programme, which highlights that improved data availability and monitoring are critical for sustainable resource management and environmental decision-making (UNEP, 2024). Without such data infrastructure, circular strategies remain constrained by uncertainty and inefficiency.

3.2 Key Data Layers in Circular Economy

To operationalize circular systems, it is useful to distinguish between several key data layers that collectively enable system-level visibility and optimization:

• Product-Level Data:
  Includes detailed information on material composition, design specifications, durability, and repairability. This data is essential for enabling reuse, remanufacturing, and recycling, and forms the basis for emerging tools such as Digital Product Passports.
• Supply Chain Data:
  Tracks the movement of products and materials across different actors and geographies. This includes logistics data, ownership transfers, and transformation processes. Enhanced supply chain transparency is critical for identifying inefficiencies and enabling reverse logistics.
• Environmental Data:
  Captures metrics such as energy use, emissions, water consumption, and waste generation across the lifecycle. This data enables the assessment and optimization of environmental impacts associated with different circular strategies.
• Usage and Behavioral Data:
  Generated through connected devices and digital platforms, this data provides insights into how products are used, maintained, and discarded. It is particularly relevant for enabling product-as-a-service models and predictive maintenance.

The integration of these data layers remains a significant challenge due to issues such as data silos, lack of standardization, and limited interoperability. Addressing these challenges is essential for building the data ecosystems required for circular systems.

3.3 Role of AI in this System

AI and data science serve as the analytical engine within circular data systems. While data provides visibility, AI enables interpretation, prediction, and optimization at scale.

Specifically, AI technologies can:

• Process and structure large volumes of fragmented data from multiple sources
• Identify patterns and inefficiencies in material and resource flows
• Enable predictive analytics, such as forecasting demand, failure rates, and waste generation
• Support real-time optimization of supply chains, production systems, and resource recovery processes

According to the United Nations Environment Programme, AI has the potential to enhance environmental decision-making by enabling more efficient use of resources and improving system-wide monitoring and optimization (UNEP, 2024).

In this sense, AI does not operate independently but relies on robust data infrastructure. The effectiveness of AI-driven circular solutions is therefore directly dependent on the availability, quality, and interoperability of data across systems.

Ultimately, AI enables the shift toward circular intelligence, where decisions are informed by continuous data flows and system-wide insights rather than isolated or reactive interventions. This aligns directly with the objective of this report, which is to examine how AI and data science can enable system-level circular optimization by transforming fragmented information into actionable intelligence.

4. AI in Circular Product Design & Material Innovation

Product design sits at the upstream end of the value chain and has a disproportionate influence on circularity outcomes. Decisions made at the design stage—regarding material selection, product architecture, durability, and end-of-life pathways—largely determine whether products can be reused, repaired, remanufactured, or recycled. In this sense, embedding circularity into design processes is essential for shifting from linear inefficiencies toward circular intelligence systems.

Artificial intelligence (AI) and data science are increasingly transforming how products are designed and materials are developed. By enabling data-driven simulation, optimization, and discovery, these technologies allow designers and engineers to move beyond static, experience-based approaches toward dynamic, predictive, and system-aware design processes.

4.1 Generative & AI-Assisted Design

Generative and AI-assisted design tools use algorithms to automatically generate and evaluate multiple design configurations based on predefined parameters and constraints. These parameters may include material efficiency, structural performance, durability, and environmental impact.

In the context of circularity, AI-enabled design can:

• Optimize products for longevity and durability, reducing the need for frequent replacement
• Enable modular design, allowing components to be easily repaired or replaced
• Reduce material usage through lightweighting and structural optimization
• Simulate full lifecycle impacts, including reuse and recycling scenarios

By leveraging large datasets and simulation capabilities, AI allows designers to anticipate how products will perform across their entire lifecycle, rather than optimizing for a single stage such as production. This represents a shift from designing for efficiency to designing for circularity, where products are intentionally created to retain value over time.

The United Nations Environment Programme highlights that digital technologies, including AI, can support more resource-efficient production and consumption systems by enabling better-informed design and decision-making processes (UNEP, 2024).

4.2 AI in Material Discovery

Material selection is a critical determinant of circularity, as it influences recyclability, environmental impact, and the feasibility of recovery processes. Traditional material development processes are often time-intensive and resource-heavy, relying on trial-and-error experimentation.

AI is transforming this process by accelerating material discovery and innovation through data-driven approaches. Machine learning models can analyze vast datasets of material properties to:

• Identify alternative materials with improved recyclability or lower environmental impact
• Accelerate the development of biodegradable or bio-based materials
• Optimize material compositions for durability and reuse
• Predict material behavior under different conditions, reducing the need for physical testing

These capabilities significantly reduce the time and cost associated with material R&D while enabling the discovery of materials better suited for circular systems.

According to the United Nations Industrial Development Organization, innovation in materials and production processes—supported by digital technologies—is essential for advancing sustainable and resource-efficient industrial systems (UNIDO, 2022).

4.3 Design for Circular Business Models

AI and data science play a critical role not only in product design, but also in enabling circular business models that rely on continuous data flows and lifecycle management. Among the most prominent of these is the Product-as-a-Service (PaaS) model, where customers pay for access or performance rather than owning the product outright.

Under this model, manufacturers retain ownership of products and are responsible for their maintenance, performance, and end-of-life management. This fundamentally shifts incentives: instead of maximizing sales volume, firms are encouraged to design products that are durable, repairable, and resource-efficient, as these characteristics directly impact long-term profitability.

AI and data science enable the effective operation of such models by:

• Monitoring product usage through connected devices (IoT)
• Enabling predictive maintenance, reducing downtime and extending product lifespan
• Optimizing asset utilization and performance across users
• Supporting reverse logistics and end-of-life recovery decisions

These capabilities transform products into data-generating assets, allowing companies to continuously improve performance and resource efficiency over time.

According to the World Economic Forum, digital technologies such as AI and IoT are key enablers of new business models that decouple economic value creation from resource consumption, particularly by improving asset utilization and extending product lifecycles (World Economic Forum, 2026).

In this context, design priorities extend beyond physical attributes to include data integration, connectivity, and lifecycle traceability. Products must be designed not only for use, but for continuous monitoring, maintenance, and recovery within circular systems.

5. Intelligent Circular Supply Chains

AI-enabled circular supply chains use real-time data and predictive systems to improve how materials, products, and assets move across their lifecycle. Unlike traditional supply chains, which are linear and reactive, intelligent circular supply chains are adaptive systems that continuously optimize flows of resources, products, and information.

5.1 Predictive Maintenance & Lifecycle Extension

Predictive maintenance uses AI and IoT data to anticipate when equipment or products are likely to fail, allowing intervention before breakdown occurs.

Instead of fixing assets after failure, systems:

• Monitor usage and performance in real time
• Predict wear and degradation patterns
• Schedule maintenance proactively

This extends product lifespans and reduces premature replacement. The International Energy Agency highlights that digital monitoring and predictive systems significantly improve industrial efficiency by reducing downtime and extending asset life (IEA, 2017)

5.2 Demand Forecasting & Resource Efficiency

AI-driven demand forecasting improves circularity by reducing overproduction and material waste.

Traditional forecasting relies on historical averages, while AI models:

• Analyze real-time consumption patterns
• Incorporate external variables (seasonality, behavior shifts, supply shocks)
• Continuously update predictions

This enables firms to align production more closely with actual demand, reducing excess inventory and unnecessary resource extraction.

The Organisation for Economic Co-operation and Development emphasizes that data-driven forecasting and digital tools are key to improving resource efficiency in production systems (OECD, 2024).

5.3 Reverse Logistics Optimization

Reverse logistics refers to the movement of products and materials from consumers back into production systems (for reuse, repair, or recycling).

AI improves this process by:

• Optimizing collection routes
• Identifying economically viable recovery paths
• Matching returned products with refurbishment or recycling facilities

This reduces transport costs, emissions, and processing inefficiencies. The World Bank notes that digital logistics systems significantly enhance waste and resource recovery efficiency in urban and industrial systems (World Bank, 2026)

5.4 Dynamic Resource Mapping

Dynamic resource mapping uses AI and spatial data to continuously track where materials, waste streams, and secondary resources are located and how they can be reused.

It enables:

• Real-time identification of resource availability
• Matching waste outputs to industrial inputs (industrial symbiosis)
• Reduced dependency on virgin materials

This transforms waste streams into searchable, tradable resource pools, improving overall system efficiency.

The United Nations Industrial Development Organization highlights that digital tools enabling industrial symbiosis are essential for improving resource efficiency across sectors (UNIDO, 2026).

6. AI-Enabled Waste Management & Recycling Systems

AI is increasingly applied at the end-of-life stage to improve material recovery rates, sorting accuracy, and system efficiency. By combining sensor data, machine learning, and automation, these systems reduce contamination, increase recovery value, and enable more precise handling of complex waste streams.

6.1 Computer Vision in Waste Sorting

Computer vision uses cameras and AI models to identify and classify waste materials in real time on sorting lines.

This enables:

• Automated separation of plastics, metals, paper, and e-waste
• Detection of material composition, color, and shape
• Higher purity of recovered materials

Improved sorting accuracy is critical, as contamination significantly reduces recycling efficiency. The International Telecommunication Union highlights that advanced technologies, including AI, can enhance e-waste management by improving material identification and recovery processes (ITU, 2020).

6.2 Advanced Recycling Optimization

AI is used to optimize recycling processes—particularly in mechanical and chemical recycling systems—by improving yield, stability, and efficiency.

Applications include:

• Process control in recycling facilities
• Optimization of temperature, pressure, and chemical inputs
• Predicting output quality based on input waste composition

This is especially important for complex materials (e.g., multi-layer plastics), where process variability affects recovery outcomes.

6.3 Smart Waste Infrastructure

Smart waste systems use sensors and data platforms to monitor, manage, and optimize waste collection and infrastructure in real time.

Key functions:

• Fill-level monitoring of waste bins
• Dynamic routing of collection vehicles
• Data-driven planning of waste systems

These systems reduce operational costs and emissions while improving collection efficiency and service coverage.

7. Digital Product Passports (DPP) & Circular Data Infrastructure

Digital Product Passports (DPPs) and associated data infrastructures form the backbone of circular economies by enabling structured, traceable, and shareable product-level information across the value chain. They shift products from physical-only entities to data-rich assets, supporting transparency, recovery, and lifecycle optimization.

7.1 Digital Product Passports (DPP)

A Digital Product Passport is a standardized digital record attached to a product that contains key lifecycle information such as material composition, origin, repairability, and end-of-life options.

This enables:

• Full traceability across the product lifecycle
• Easier identification of reuse, repair, and recycling pathways
• Improved compliance with circular economy regulations

The European Commission identifies DPPs as a core instrument under its Circular Economy Action Plan, aimed at increasing transparency and enabling sustainable product design and recovery (European Commission, 2020).

7.2 Integration of AI, IoT, and Blockchain

DPPs become significantly more powerful when integrated with emerging digital technologies:

• IoT (Internet of Things): Generates real-time usage and condition data
• AI: Processes lifecycle data to optimize repair, reuse, and recovery decisions
• Blockchain: Ensures secure, tamper-proof recording of product history

Together, these technologies create a trusted, continuously updated digital representation of physical assets, enabling automated circular decision-making at scale.

The World Economic Forum highlights that combining digital technologies enhances transparency and enables more efficient and circular value chains through improved data sharing and traceability (World Economic Forum, 2026).

7.3 Data Governance & Interoperability

The effectiveness of circular data systems depends on how well data is governed, standardized, and shared across organizations.

Key requirements include:

• Data governance frameworks to define ownership, access, and usage rights
• Standardization protocols to ensure consistency across industries
• Interoperability mechanisms to enable seamless data exchange between platforms

Without these, even advanced digital systems remain fragmented and underutilized.

8. Case Studies: Applied AI in Circular Systems

8.1 AI in E-Waste Sorting & Recovery

AI-powered sorting systems use computer vision and robotics to identify and separate valuable components in e-waste streams (e.g., circuit boards, metals, plastics). This improves recovery rates and reduces contamination in high-value materials.

Given that only a small share of global e-waste is formally recycled, improving sorting efficiency is critical for recovering scarce materials such as rare earth elements (ITU, 2020).

8.2 Digital Product Passports in the EU

The European Union is implementing Digital Product Passports under its sustainable product policy framework to standardize product-level data across value chains. DPPs enable traceability of materials, repair history, and environmental performance.

This supports regulatory compliance, facilitates recycling, and improves transparency for consumers and recyclers (European Commission, 2020).

8.3 AI in EV Battery Second-Life Applications

AI is used to assess the remaining useful life of electric vehicle batteries, enabling their reuse in secondary applications (e.g. energy storage systems).

Machine learning models analyze battery performance data to:

• Predict degradation patterns
• Determine suitability for reuse
• Optimize allocation to second-life uses

This extends battery lifespans and reduces demand for new raw materials (IEA, 2023).

8.4 Smart Waste Management in Urban Systems

Cities are deploying AI-enabled systems to optimize waste collection, routing, and infrastructure planning.

These systems use:

• Sensor data (bin fill levels)
• Route optimization algorithms
• Real-time monitoring platforms

to reduce operational costs, improve collection efficiency, and lower emissions.

Urban waste challenges remain significant globally, requiring more efficient systems to improve collection and recovery rates.

9. Regional Perspective: Scaling AI-Driven Circularity in the Arab Region

The Arab region (Middle East and North Africa – MENA) presents both high urgency and high potential for scaling AI-driven circular systems. Rapid urbanization, rising consumption, and infrastructure gaps have led to growing waste volumes and low material recovery rates, making circular transformation a priority.

9.1 Structural Challenges

• The region generates over 155 million tons of waste annually, with volumes expected to nearly double by 2050 (World Bank, 2026)
• Only 10% of waste is recycled or reused, while a large share remains mismanaged (World Bank, 2026)
• Waste systems are often limited to collection and disposal, with weak recovery and recycling infrastructure

At the system level, key bottlenecks include:

• Limited data availability and monitoring systems
• Fragmented institutional coordination
• Underinvestment in advanced waste and recycling technologies

9.2 Role of AI & Data in the Region

AI and data science can directly address these constraints by:

• Enabling data-driven waste tracking and resource mapping
• Improving sorting, recovery, and logistics efficiency
• Supporting policy planning through predictive analytics

However, adoption remains uneven due to:

• Gaps in data infrastructure and governance
• Limited AI capacity and skilled workforce
• Unequal digital readiness across countries

9.3 Opportunities for Acceleration

Despite these barriers, the region has strong enabling conditions:

• High digital adoption in countries like United Arab Emirates and Saudi Arabia
• Emerging startup ecosystems and innovation hubs
• Large untapped circular value (up to 83% of waste could be recovered)

Key opportunity areas include:

• AI-enabled waste management systems in cities
• Industrial symbiosis platforms linking sectors
• Digital infrastructure such as material tracking systems and DPPs

9.4 Strategic Priorities

To scale AI-driven circularity, the region must focus on:

• Investing in data systems (collection, standardization, sharing)
• Strengthening policy frameworks and regional coordination
• Building technical capacity and workforce skills in AI
• Leveraging public–private partnerships to deploy solutions

10. Strategic & Policy Recommendations

Scaling AI-driven circularity requires coordinated action across policy, industry, and international systems, with data infrastructure and standardization as the central enablers.

10.1 For Governments

• Establish data standards and interoperability frameworks to enable cross-sector data exchange
• Mandate or incentivize Digital Product Passports (DPPs) for priority sectors
• Invest in digital and waste infrastructure (AI-ready systems, smart collection, recycling capacity)
• Introduce economic instruments (e.g., extended producer responsibility, circular incentives)
• Build national AI and data capabilities through education and public-sector adoption

10.2 For Industry

• Integrate AI across product lifecycle stages (design, use, recovery)
• Adopt data-sharing practices across supply chains to reduce information gaps
• Shift toward circular business models (e.g., product-as-a-service, remanufacturing)
• Invest in traceability systems to support compliance and recovery
• Collaborate across sectors to enable industrial symbiosis and resource exchange

10.3 For International Organizations

• Support development of global standards for circular data systems
• Provide technical assistance and capacity building in emerging markets
• Facilitate cross-border data and knowledge sharing platforms
• Mobilize financing for digital and circular infrastructure projects
• Align circular economy efforts with broader digital and climate agendas

11. Conclusion & Future Outlook: Toward Autonomous Circular Systems

The main constraint to scaling circular economy systems is no longer conceptual—it is operational. Circularity depends on the ability to track, manage, and optimize material flows, which in turn depends on data availability and system integration. AI and data science address this gap by enabling more accurate, real-time, and system-wide decision-making.

Across the lifecycle, these technologies improve how products are designed, used, and recovered. When combined with digital infrastructures such as connected devices and product-level data systems, they enable continuous visibility and coordination across value chains.

Looking ahead, circular systems are expected to become increasingly automated and data-driven. Products will generate data throughout their lifecycle, and AI systems will use this data to optimize maintenance, reuse, and recovery decisions with minimal manual intervention. This marks a shift toward autonomous circular systems, where resource flows are continuously optimized.

However, this transition depends on resolving key challenges, particularly:

• Data standardization and interoperability
• Alignment between digital systems and physical infrastructure
• Effective governance frameworks

In conclusion, AI and data science are not standalone solutions but core enablers of scalable circular systems. Their impact will depend on how effectively data ecosystems are developed and integrated across the economy.