Explainer: Digital twins, simulation & synthetic data — the concepts, the economics, and the decision checklist

By 2025, the global digital twin market reached USD 16.7 billion, with Asia-Pacific commanding a 37% share and growing at a compound annual rate exceeding 35%—faster than any other region. For sustainability practitioners navigating Scope 3 emissions, supply chain traceability, and decarbonization pathways, digital twins have evolved from futuristic concepts into operational necessities. Yet beneath the market enthusiasm lies a more complex reality: implementation failures remain stubbornly high, stakeholder misalignment derails promising pilots, and hidden bottlenecks—from data quality to organizational inertia—consume resources without delivering promised returns. This explainer unpacks the core concepts, examines what distinguishes successful deployments from costly experiments, and provides a decision framework for teams evaluating whether digital twins belong in their sustainability toolkit.

Why It Matters

The convergence of three forces makes digital twins, simulation, and synthetic data indispensable for sustainability transformation in the Asia-Pacific region. First, regulatory pressure has intensified dramatically: the European Union's Carbon Border Adjustment Mechanism (CBAM) requires granular emissions data that Asian exporters must now provide, while Singapore's mandatory climate reporting for listed companies (effective 2025) and Japan's GX League covering 680+ corporations demand unprecedented visibility into value chain emissions. Second, physical climate risks are materializing faster than anticipated—the Asian Development Bank estimates that without adaptation, climate change could reduce Asia-Pacific GDP by 24% by 2100, creating urgent demand for predictive modeling of infrastructure resilience. Third, the economics of sustainability data have fundamentally shifted: manual carbon accounting costs enterprises USD 50,000–500,000 annually in professional services fees, while digital twin-enabled automation can reduce these costs by 60–80% once operational.

The 2024-2025 period has witnessed several pivotal developments. MarketsandMarkets projects that the Asia-Pacific digital twin market will grow from USD 6.2 billion in 2024 to USD 23.8 billion by 2029. Manufacturing leads adoption, with 47% of large Asia-Pacific manufacturers now operating at least one digital twin pilot, according to a 2024 McKinsey survey. However, only 18% have achieved enterprise-scale deployment—revealing the implementation gap that separates proof-of-concept success from organizational transformation.

For sustainability specifically, digital twins address three critical pain points: (1) Scope 3 emissions, which typically constitute 70–90% of a company's carbon footprint yet remain notoriously difficult to measure; (2) supply chain traceability across the region's fragmented manufacturing ecosystems; and (3) scenario modeling for transition planning under multiple decarbonization pathways. The Asia-Pacific context presents unique characteristics: complex multi-tier supply chains spanning diverse regulatory environments, significant infrastructure for renewable energy deployment, and a manufacturing base responsible for approximately 40% of global industrial emissions.

Key Concepts

Digital Twins are dynamic virtual representations of physical assets, processes, or systems that are continuously updated with real-time data. Unlike static 3D models, digital twins maintain bidirectional data flows—sensor inputs update the virtual model, while simulations in the virtual environment can inform physical interventions. For sustainability applications, this means modeling energy consumption patterns across a factory floor, simulating emissions under different production scenarios, or predicting equipment failures before they occur. The fidelity of a digital twin ranges from descriptive (representing current state) to predictive (forecasting future behavior) to prescriptive (recommending optimal actions).

Benchmark KPIs refer to the standardized metrics against which digital twin performance is measured. In sustainability contexts, critical KPIs include energy intensity per unit of production (kWh/unit), carbon intensity (kgCO2e/unit), water consumption efficiency, and waste diversion rates. The challenge lies in establishing baselines: without accurate historical data, digital twins cannot demonstrate improvement. Leading implementations establish KPI hierarchies—operational metrics (real-time process efficiency), tactical metrics (monthly resource consumption), and strategic metrics (annual emissions trajectories aligned with science-based targets).

Traceability encompasses the ability to track materials, components, and products throughout supply chains with verified provenance data. Digital twins enable traceability by creating immutable records of material flows, transformation processes, and custody transfers. For Scope 3 accounting, traceability answers the fundamental question: what is the embodied carbon of each input to my production process? Blockchain-enabled digital twins add cryptographic verification to traceability claims, though implementation complexity and interoperability challenges remain significant barriers.

AI Agents are autonomous software systems that can perceive their environment, make decisions, and take actions without continuous human supervision. Within digital twin ecosystems, AI agents perform functions including anomaly detection (identifying deviations from normal operating patterns), predictive maintenance (scheduling interventions before failures), and optimization (adjusting process parameters to minimize energy consumption or emissions). The integration of large language models (LLMs) with digital twins—emerging strongly in 2024-2025—enables natural language interfaces for querying complex simulation environments.

Scope 3 Emissions comprise all indirect emissions occurring in a company's value chain, both upstream (purchased goods, transportation, business travel) and downstream (product use, end-of-life treatment). For most organizations, Scope 3 represents the largest and most challenging emissions category to measure. Digital twins address Scope 3 by creating virtual representations of supply chain processes, enabling estimation based on activity data, spend data, or supplier-specific emissions factors. The accuracy of Scope 3 estimates from digital twins depends critically on the quality and granularity of input data from value chain partners.

What's Working and What Isn't

What's Working

Integrated building management systems in commercial real estate have demonstrated consistent success. Singapore's CapitaLand deployed digital twins across 49 properties in 2024, achieving verified energy reductions of 15–20% through AI-optimized HVAC scheduling and predictive equipment maintenance. The key success factor was integration with existing building management systems rather than parallel deployment—reducing change management resistance and enabling incremental optimization. Similar results have emerged from Hong Kong's Swire Properties and Japan's Mitsui Fudosan, suggesting that commercial real estate represents a mature use case with established ROI patterns.

Port operations and maritime logistics have emerged as high-impact applications. The Port of Shanghai's digital twin, operational since 2023 and expanded in 2024, processes data from 43,000+ sensors to optimize container routing, crane operations, and vessel berth assignments. Reported outcomes include a 23% reduction in vessel turnaround time and 12% decrease in port-related emissions per TEU (twenty-foot equivalent unit). The Port of Busan in South Korea and PSA Singapore have implemented comparable systems, with PSA reporting 8% energy efficiency gains in terminal operations.

Renewable energy asset management demonstrates strong alignment between digital twin capabilities and sustainability outcomes. China's Goldwind and India's Suzlon have deployed digital twins across wind farm portfolios, using predictive analytics to optimize maintenance scheduling and reduce unplanned downtime. Goldwind's 2024 sustainability report documented a 6% improvement in capacity factor across digital twin-enabled turbines, translating directly to increased clean energy generation and improved project economics.

What Isn't Working

Multi-tier supply chain visibility remains elusive despite significant investment. The fundamental challenge is data sharing: Tier 2 and Tier 3 suppliers—often small and medium enterprises with limited digital capabilities—cannot provide the granular data required for accurate digital twin modeling. A 2024 study by the World Economic Forum found that 67% of Asia-Pacific supply chain digital twin initiatives failed to extend beyond Tier 1 suppliers. The incentive structure is misaligned: suppliers bear implementation costs while buyers capture value, creating rational resistance to participation.

Cross-organizational interoperability continues to frustrate ecosystem-level applications. Digital twins from different vendors typically cannot exchange data seamlessly, forcing organizations to choose between vendor lock-in and costly custom integrations. The absence of universally adopted standards for sustainability data exchange (despite initiatives like the Partnership for Carbon Transparency) means that each supply chain participant may define emissions factors, allocation methodologies, and reporting boundaries differently—undermining the consistency that digital twins require.

Synthetic data quality for edge cases presents challenges for training AI models within digital twins. While synthetic data can address privacy concerns and augment limited training datasets, generated data often fails to capture rare but consequential events—equipment failures, extreme weather impacts, or supply chain disruptions. Models trained primarily on synthetic data may perform well under normal operating conditions but fail precisely when accurate predictions matter most. Validation approaches for synthetic data in sustainability applications remain immature.

Key Players

Established Leaders

Siemens operates one of the most comprehensive digital twin portfolios through its Xcelerator platform, with particularly strong capabilities in manufacturing and infrastructure. The company's 2024 acquisition of Altair strengthened simulation and AI capabilities relevant to sustainability applications.

AVEVA (a Schneider Electric company) focuses on industrial digital twins for process industries including chemicals, oil and gas, and power generation. Their Industrial AI capabilities specifically address energy optimization and emissions monitoring.

Dassault Systèmes offers the 3DEXPERIENCE platform with strong applications in product lifecycle management and sustainable innovation. Their Virtual Twin Experience extends to supply chain modeling and circular economy applications.

Bentley Systems specializes in infrastructure digital twins, with particular relevance for sustainable infrastructure planning, construction, and operations. Their iTwin platform has gained traction in Asia-Pacific infrastructure projects.

PTC provides digital twin solutions through its ThingWorx and Vuforia platforms, with manufacturing applications including energy monitoring and predictive maintenance relevant to industrial decarbonization.

Emerging Startups

Akselos (Singapore-based) specializes in digital twins for offshore energy assets, using patented reduced-order modeling to simulate structural integrity and extend asset lifecycles—directly relevant for both conventional and renewable offshore installations.

Cognite (Norway-based with significant Asia-Pacific presence) offers an industrial DataOps platform that enables digital twin development without extensive custom engineering, lowering barriers for sustainability data integration.

Willow (Australia-based) focuses on built environment digital twins with strong sustainability capabilities, including carbon tracking and energy optimization across building portfolios.

Pratiti Technologies (India-based) develops digital twin solutions for utilities and smart city applications, addressing grid optimization and urban sustainability in emerging market contexts.

AspenTech (though established, their sustainability-focused digital twin capabilities represent newer developments) has expanded into emissions management and ESG data integration within process industry digital twins.

Key Investors & Funders

Temasek Holdings (Singapore) has invested significantly in digital infrastructure and sustainability technology, including companies developing digital twin capabilities for decarbonization.

SoftBank Vision Fund has backed multiple digital twin and industrial AI companies, signaling continued interest in the intersection of digitalization and sustainability.

Asian Development Bank (ADB) provides financing for smart infrastructure projects across developing Asia, often incorporating digital twin requirements into project specifications for climate-resilient infrastructure.

Energy Catalyst (UK-based but active in Asia-Pacific) funds clean energy innovations including digital solutions for energy system optimization and grid management.

Breakthrough Energy Ventures has invested in companies developing simulation and digital twin capabilities for hard-to-abate sectors, with portfolio companies operating across Asia-Pacific markets.

Examples

Toyota's Multi-Plant Digital Twin Network in Japan and Thailand exemplifies successful cross-border implementation. Beginning in 2023 and expanding through 2024-2025, Toyota deployed integrated digital twins across 12 manufacturing facilities, creating unified visibility into energy consumption, emissions, and production efficiency. The system processes data from over 2 million sensors and uses AI-driven optimization to reduce energy consumption. Documented outcomes include: 11% reduction in energy intensity per vehicle produced, 8% decrease in Scope 1 and 2 emissions across connected facilities, and JPY 4.2 billion (approximately USD 28 million) annual cost savings. Critical success factors included executive sponsorship, standardized data architectures across plants, and phased implementation that demonstrated value before scaling.

Singapore's Tuas Nexus Integrated Waste Management Facility represents a public infrastructure application. This first-of-its-kind facility combines a water reclamation plant with an integrated waste management facility, with a comprehensive digital twin enabling real-time optimization of energy recovery from waste. The digital twin integrates data from over 15,000 sensors monitoring waste processing, water treatment, and energy generation. Performance metrics from initial operations (2024-2025) show: 42% of facility energy needs met through waste-to-energy processes, 3% improvement in water recycling efficiency versus design specifications, and 18% reduction in unplanned maintenance downtime. The project demonstrates how digital twins can optimize complex multi-process facilities for circular economy outcomes.

Tata Steel's Jamshedpur Works Decarbonization Initiative in India applies digital twins to heavy industry emissions reduction. Tata Steel deployed simulation capabilities across its integrated steel manufacturing complex to identify decarbonization pathways without disrupting production. The digital twin models blast furnace operations, coke plant efficiency, and energy flows across the facility. Results from the 2024 implementation phase include: identification of 340,000 tonnes annual CO2 reduction potential through process optimization, 7% improvement in blast furnace fuel efficiency, and validated pathways for hydrogen injection that inform capital investment decisions. The initiative illustrates how digital twins enable scenario planning for industrial transition where physical experimentation is prohibitively expensive or risky.

Action Checklist

• Conduct a data readiness assessment: inventory existing sensor infrastructure, data quality, and gaps before committing to digital twin implementation
• Define clear sustainability KPIs with measurable baselines—digital twins cannot demonstrate improvement without accurate starting points
• Identify a bounded pilot scope with strong executive sponsorship; enterprise-wide implementations without proven value typically fail
• Evaluate vendor lock-in risks and data portability; ensure contracts address data ownership and interoperability requirements
• Engage Tier 1 suppliers early on data-sharing expectations; address incentive alignment before assuming supply chain visibility is achievable
• Establish governance structures for AI agent decision-making, particularly for autonomous optimization that affects emissions or energy consumption
• Plan for organizational change management; digital twins require new skills, workflows, and potentially new roles that must be resourced
• Build validation protocols for synthetic data if used for model training; document limitations and edge case performance
• Align digital twin outputs with regulatory reporting requirements (CBAM, ISSB, local mandates) to maximize compliance value
• Create feedback mechanisms to continuously improve model accuracy based on divergence between simulated and actual outcomes

FAQ

Q: What is the typical ROI timeline for sustainability-focused digital twins? A: Implementation timelines vary significantly by use case and organizational readiness. Building energy optimization typically achieves positive ROI within 18–24 months, driven by energy cost savings and relatively straightforward sensor integration. Manufacturing digital twins for emissions reduction often require 24–36 months to demonstrate enterprise-scale ROI, given the complexity of production environments and the need for organizational learning. Supply chain digital twins targeting Scope 3 visibility have the longest and most uncertain payback periods—often exceeding 36 months—due to dependency on supplier participation and data quality challenges. Organizations should expect higher upfront investment and longer payback for applications addressing harder sustainability problems.

Q: How do digital twins integrate with existing carbon accounting and ESG reporting systems? A: Integration approaches range from manual data extraction to API-based automated flows. Leading implementations use digital twins as data sources that feed into carbon accounting platforms (such as Persefoni, Watershed, or Sphera), with standardized emissions factors and allocation methodologies applied at the reporting layer. The challenge is maintaining consistency between operational digital twin data and financial/ESG reporting boundaries—production data may need transformation to align with organizational reporting structures. Some organizations are consolidating digital twin and carbon accounting functions into unified platforms, though this introduces vendor concentration risk.

Q: What distinguishes high-fidelity digital twins from simpler simulation approaches? A: Fidelity refers to how accurately the virtual model represents physical reality. High-fidelity digital twins incorporate physics-based modeling (computational fluid dynamics, finite element analysis), real-time sensor data at high resolution, and validated predictive accuracy. Simpler approaches may use rule-based heuristics, historical averages, or statistical models without continuous real-time updates. For sustainability applications, the required fidelity depends on the decision being informed: strategic scenario planning may tolerate lower fidelity, while process optimization for emissions reduction typically requires high-fidelity representations to avoid unintended consequences. Higher fidelity correlates with higher implementation cost, longer development timelines, and greater computational requirements.

Q: How should organizations address data privacy and competitive sensitivity when sharing supply chain data for digital twins? A: Several approaches have emerged: aggregated data sharing (where individual supplier data is anonymized within broader benchmarks), trusted third-party intermediaries that process sensitive data without exposing it to buyers, and blockchain-based systems that verify claims without revealing underlying data. Differential privacy techniques can add mathematical noise to datasets while preserving analytical utility. However, no approach fully resolves the tension between data granularity required for accurate Scope 3 modeling and legitimate supplier concerns about competitive intelligence. Successful implementations typically combine technical safeguards with contractual protections and governance frameworks that give suppliers meaningful input into data usage.

Q: What role does synthetic data play in sustainability digital twins, and what are its limitations? A: Synthetic data serves three primary functions: augmenting limited training datasets for AI models, enabling scenario simulation without requiring actual (and potentially dangerous) conditions, and protecting privacy when using sensitive operational data. For sustainability applications, synthetic data can model extreme weather impacts, equipment failure cascades, or demand scenarios that have not yet occurred. Limitations include: difficulty capturing correlations between variables that only emerge under specific real-world conditions, potential for model bias if synthetic data generation reflects assumptions rather than physical reality, and validation challenges—there is no ground truth for events that have not happened. Best practice combines synthetic and real data, with explicit documentation of synthetic data provenance and ongoing validation against observed outcomes.

Sources

• MarketsandMarkets. "Digital Twin Market - Global Forecast to 2029." Published 2024.
• McKinsey & Company. "Digital Twins: The Art of the Possible in Operations." McKinsey Digital, 2024.
• Asian Development Bank. "Climate Change and the Future of Asia: Regional Modeling and Analysis." ADB Publications, 2024.
• World Economic Forum. "Digital Twins in Supply Chain: State of Adoption." WEF Industry Reports, 2024.
• Singapore Exchange Regulation. "Climate-Related Disclosures: Implementation Guidance for Listed Companies." SGX RegCo, 2024.
• International Energy Agency. "Digitalization and Energy Systems: Technology and Policy." IEA Publications, 2024.
• Partnership for Carbon Transparency. "Pathfinder Framework: Guidance for the Calculation and Exchange of Product Carbon Footprints." PACT, Version 2.1, 2024.