Trend watch: Digital twins, simulation & synthetic data in 2026 — signals, winners, and red flags

The global digital twin market surged from $17.7 billion in 2024 to an estimated $24.5 billion in 2025—a 38% year-over-year increase that outpaced nearly every other enterprise technology category (Fortune Business Insights, 2025). For sustainability leaders navigating decarbonization mandates, supply chain complexity, and Scope 3 reporting requirements, this trajectory signals a fundamental shift: digital twins are no longer experimental pilot projects but mission-critical infrastructure for climate action. The convergence of digital twins, physics-based simulation, and synthetic data generation creates capabilities that were impossible just two years ago—virtual replicas that predict emissions impacts, stress-test decarbonization pathways, and generate the training data needed to optimize systems at scale. Yet beneath the market enthusiasm lies significant variance in deployment success, unit economics that challenge conventional ROI models, and adoption barriers that derail even well-funded initiatives.

Why It Matters

Digital twins represent the operational backbone of corporate sustainability strategy. According to a 2025 Siemens survey of 263 senior sustainability leaders, 59% of organizations using industrial AI and digital twins reported measurable carbon reductions averaging 24%, while 65% achieved energy savings averaging 23%. These aren't marginal improvements—they represent the difference between incremental progress and genuine trajectory change toward net-zero commitments.

The urgency intensifies when examining regulatory timelines. The SEC's climate disclosure rules, the EU's Corporate Sustainability Reporting Directive (CSRD), and California's Climate Corporate Data Accountability Act collectively mandate granular emissions reporting across value chains. Scope 3 emissions—those occurring in supplier and customer networks—typically constitute 70-90% of a company's carbon footprint yet remain notoriously difficult to measure. Digital twins, when integrated with supplier data platforms, transform Scope 3 from a guessing game into a measurable, optimizable system.

Synthetic data compounds this value proposition. Gartner projects that 80% of AI-developing organizations will use synthetic data by 2025, up from 23% in 2019. For sustainability applications, synthetic data enables scenario modeling without waiting for years of operational history—testing the emissions impact of switching suppliers, reconfiguring logistics networks, or adopting new materials before committing capital.

The strategic question for 2026 isn't whether to adopt these technologies, but how to deploy them without falling into the traps that have stalled 67% of digital twin initiatives before reaching production scale.

Key Concepts

Digital Twins for Sustainability

A digital twin is a dynamic virtual representation of a physical asset, process, or system that continuously updates based on real-world data. Unlike static models or simulations, digital twins maintain bidirectional data flows—operational data informs the virtual model, which then provides insights back to operators. For sustainability applications, this means tracking energy consumption, emissions generation, material flows, and waste streams in near-real-time.

The most valuable implementations extend beyond individual assets to system-level twins. A factory digital twin might model equipment efficiency, but a supply chain digital twin models material sourcing, transportation emissions, manufacturing processes, and distribution networks as an integrated system. This systems-level view proves essential for Scope 3 accounting, where emissions reductions in one area can trigger increases elsewhere.

Simulation and Scenario Modeling

Physics-based simulation enables predictive capability within digital twins. Rather than simply mirroring current operations, simulation allows sustainability teams to test alternative scenarios: What happens to our carbon footprint if we shift 30% of procurement to regional suppliers? How do different energy sourcing strategies affect our Science Based Targets trajectory?

The computational intensity of high-fidelity simulation previously limited these capabilities to specialized engineering applications. Cloud computing and AI-accelerated simulation have democratized access, enabling scenario analysis that would have required supercomputer access just five years ago.

Synthetic Data Generation

Synthetic data—artificially generated information that preserves the statistical properties of real data without exposing sensitive details—addresses two critical gaps in sustainability analytics. First, it solves privacy and confidentiality constraints that prevent sharing supplier emissions data across organizational boundaries. Second, it generates edge cases and failure scenarios that rarely occur in operational data but prove essential for robust AI model training.

The synthetic data market reached approximately $510 million in 2024 and is projected to grow at 35-46% CAGR through 2029 (Mordor Intelligence, 2025). For sustainability applications, synthetic data enables training emissions prediction models on scenarios like facility closures, extreme weather disruptions, or material supply shocks—events too rare to appear in historical data but critical for resilience planning.

What's Working

Integrated Energy-Water Optimization

The Siemens-Ecolab Climate Intelligence partnership demonstrates the value of integrated digital twins for resource optimization. Deployed across refineries in Latin America, Asia, Middle East, Europe, and the United States, the platform combines Siemens gPROMS process modeling with Ecolab's water management expertise. Results from a mid-size Latin American refinery showed CO₂ reductions of up to 38,000 metric tons annually, water savings of 1 million cubic meters per year, and cost savings of $1.8 million—all while maintaining or increasing production output.

The key insight: water carries 33-75% of industrial plant energy through cooling and steam systems. Optimizing water efficiency directly reduces emissions in ways that equipment-focused approaches miss entirely. The partnership estimates potential to reduce 25 million tons of carbon annually if deployed across global industry.

Product-Level Carbon Footprint Tracking

Unilever's Supplier Climate Programme illustrates how digital twins enable product-level emissions transparency. By 2024, the company engaged 291 suppliers representing approximately 44% of Scope 3 emissions from raw materials, ingredients, and packaging. Using the PACT (Partnership for Carbon Transparency) methodology, Unilever collected over 250 Product Carbon Footprint data points, transitioning from generic emission factors to supplier-specific measurements.

The digital twin infrastructure supporting this initiative spans 124 factories and 2,100 manufacturing lines, covering 75% of production capacity. Results include 3% increases in overall equipment effectiveness, 5% higher labor productivity, and 8% cost reductions—demonstrating that sustainability data infrastructure delivers operational benefits beyond environmental compliance.

Building and District Energy Management

Severyn Trent's Net Zero Hub project in Stoke-on-Trent, UK, represents the first application of digital twins to achieve carbon-neutral wastewater treatment. The £38 million initiative uses Siemens gPROMS digital twins to optimize bacterial digestion processes, targeting elimination of the 57% of wastewater emissions arising from CO₂, methane, and nitrous oxide in treatment operations. The project provides a replicable blueprint for net-zero water utilities globally.

What's Not Working

Pilot Purgatory

The most common failure mode isn't technical—it's organizational. According to Deloitte's 2024 digital twin research, 67% of scaled deployments underperformed their pilots by at least 20% on key metrics. Pilot conditions differ fundamentally from production: curated task types, motivated users, extra attention from developers, and simplified integration requirements. Organizations frequently scale based on pilot success without recognizing these artificial conditions.

The unit economics often collapse at scale. A pilot serving one facility might cost $150,000 annually, which executives approve as an innovation investment. Scaling to 50 facilities doesn't cost $7.5 million—it costs $12-15 million after accounting for data integration complexity, change management, and infrastructure requirements that pilots avoid.

Data Integration Nightmares

Building robust multimodal data architecture remains the primary technical barrier. Most manufacturing environments run 15-30 disparate systems—MES, SCADA, ERP, quality management, maintenance scheduling—with minimal interoperability. Digital twins require data from all of them, synchronized in near-real-time.

Legacy systems particularly complicate emissions tracking. Equipment installed before 2010 typically lacks the sensors needed for granular energy monitoring. Retrofitting sensors across a global manufacturing footprint can exceed the cost of the digital twin platform itself.

Model Drift and Validation Gaps

Digital twins lose accuracy as physical systems age and operating conditions change. A model calibrated to 2024 performance may provide misleading guidance by 2026 if equipment degrades, processes evolve, or external conditions shift. Organizations often underinvest in continuous validation—the ongoing comparison of twin predictions against actual outcomes.

Synthetic data compounds this risk when used without proper validation. Models trained on synthetic scenarios may perform excellently on synthetic test cases while failing on real-world edge cases that differ in subtle but critical ways.

Key Players

Established Leaders

Siemens AG — The dominant force in industrial digital twins, with the Xcelerator platform integrating Teamcenter, NX, and Simcenter for design, simulation, and manufacturing applications. Their gPROMS process modeling platform leads sustainability applications in chemicals, refining, and water treatment.

Microsoft — Azure Digital Twins provides the cloud infrastructure underlying many enterprise deployments. Their partnership with NVIDIA for industrial metaverse applications and integration with Dynamics 365 for supply chain visibility positions them as the horizontal platform layer.

NVIDIA — The Omniverse platform enables physically accurate simulation at scale, with January 2025 integration with Siemens Teamcenter for "Digital Reality Viewer." Their partnership with Siemens and BMW demonstrates manufacturing applications achieving 30% planning efficiency improvements.

Dassault Systèmes — The 3DEXPERIENCE platform serves aerospace, automotive, and life sciences with emphasis on product lifecycle sustainability. Their Virtual Twin Experience enables end-to-end carbon footprint modeling.

PTC — ThingWorx and Windchill platforms focus on connected products and service lifecycle management, with particular strength in industrial equipment and automotive applications.

Emerging Startups

OroraTech (Germany) — Raised €37 million Series B in 2024 for satellite-based wildfire monitoring using digital twin technology. Their platform provides early warning for infrastructure operators and insurers managing climate risk.

Neara — Network infrastructure digital twins for utilities, helping energy companies model grid resilience and renewable integration scenarios. Backed by BNP Paribas Solar Impulse among sustainability-focused investors.

SIMCEL (Singapore) — AI-powered supply chain digital twins with integrated carbon modeling, enabling scenario analysis for procurement decisions and emissions optimization across complex supplier networks.

Gradyent — Energy and industrial process twins focused on heating networks and district energy systems, critical infrastructure for European decarbonization.

ANNEA — AI-powered wind turbine digital twins for predictive maintenance and performance optimization, reducing downtime while maximizing renewable energy generation.

Key Investors

Breakthrough Energy Ventures — Bill Gates-backed fund with $555 million raised in January 2024 focuses on decarbonization technologies including digital infrastructure.

World Fund — European climate tech investor with €300 million+ (closed April 2024) backing enterprise sustainability solutions.

Lowercarbon Capital — $550 million fund (2024) specifically targeting carbon reduction technologies with track record in industrial decarbonization.

Blue Earth Capital — Led OroraTech's Series B, demonstrating appetite for climate intelligence applications of digital twin technology.

Sector-Specific KPIs

Sector	Primary KPI	Benchmark Range	Red Flag Threshold
Manufacturing	Energy per unit output	15-25% reduction in Year 1	<5% after 18 months
Supply Chain	Scope 3 data coverage	40-60% of emissions tracked	<20% coverage
Buildings	HVAC efficiency gain	20-35% energy reduction	<10% reduction
Utilities	Grid utilization increase	20-30% capacity gain	<10% improvement
Logistics	Route emissions reduction	12-18% per shipment	<5% reduction
Water/Wastewater	Treatment energy intensity	25-40% reduction	<10% reduction

Examples

1. Siemens-Ecolab Climate Intelligence (Refining): Latin American refinery pilot achieved 38,000 metric tons CO₂ reduction annually with $1.8 million cost savings. The digital twin combined water management with process optimization, demonstrating that resource efficiency drives both environmental and financial returns. Scale potential: 25 million tons CO₂ reduction across global industry.

2. Unilever Manufacturing Digital Twins (CPG): 124 factories with digital twin coverage achieved 8% cost reduction alongside sustainability data infrastructure supporting Scope 3 transparency. The Brazil Indaiatuba facility (World Economic Forum Lighthouse site) delivered 20% capacity increase with €3 million savings while enabling 84% faster sustainable packaging trials.

3. Severn Trent Net Zero Hub (Utilities): £38 million project creating world's first carbon-neutral wastewater treatment facility using Siemens digital twin technology. Targets 57% reduction in treatment process emissions, providing replicable model for water utility decarbonization globally.

Action Checklist

• Audit current data architecture: Identify all systems generating sustainability-relevant data and assess integration requirements before platform selection
• Define Scope 3 boundaries: Map supplier network and prioritize digital twin coverage for the 20% of suppliers typically representing 80% of emissions
• Establish baseline metrics: Document current state across energy consumption, emissions intensity, and resource efficiency before deployment
• Budget for integration: Allocate 40-60% of total project cost to data integration, not just platform licensing
• Plan continuous validation: Build sampling-based verification comparing twin predictions against actual outcomes into operational processes
• Start with highest-value use cases: Focus initial deployment on areas where emissions reduction aligns with cost reduction—energy optimization, waste reduction, logistics efficiency
• Secure executive sponsorship: Digital twin initiatives require cross-functional authority; ensure sponsor can mandate data access across organizational silos
• Develop internal capability: Train operators and sustainability teams on twin interpretation; technology without adoption delivers zero value

FAQ

Q: What's a realistic timeline and budget for a digital twin pilot targeting sustainability outcomes? A: Plan for 9-15 months from initiation to validated pilot results. Budget $200,000-$500,000 for a single-facility manufacturing pilot including platform licensing, integration, and change management. Enterprise-wide deployment typically costs 8-15x pilot investment when accounting for scale-related complexity. Organizations expecting to replicate pilot costs across multiple facilities consistently underbudget by 40-60%.

Q: How do digital twins help with Scope 3 emissions that occur in supplier networks outside our control? A: Digital twins enable two capabilities for Scope 3: visibility and scenario analysis. By integrating supplier-provided Product Carbon Footprint data (using frameworks like PACT), digital twins model emissions across your value chain. More importantly, simulation capabilities let you test alternative sourcing strategies, identify emissions hotspots, and quantify the impact of supplier engagement initiatives before committing resources. Unilever's approach—engaging 291 suppliers covering 44% of Scope 3—demonstrates the progressive coverage model.

Q: Our organization has tried digital twin pilots that failed to scale. What typically goes wrong? A: Three patterns dominate: First, pilots avoid integration complexity by using clean, curated data feeds that don't exist in production. Second, pilots receive exceptional attention from vendors and internal teams that cannot be replicated at scale. Third, pilots often target friendly user populations whose adoption doesn't predict organization-wide behavior. Successful scaling requires designing pilots that explicitly test production conditions: messy data, minimal support, skeptical users. If it works under adverse conditions, it will work at scale.

Q: What role does synthetic data play in sustainability applications specifically? A: Synthetic data addresses three sustainability-specific gaps. First, supplier emissions data is often confidential—synthetic representations preserve patterns without exposing competitively sensitive information. Second, extreme events critical for resilience planning (facility closures, supply disruptions, extreme weather) rarely appear in historical data; synthetic generation creates training examples. Third, scenario modeling for decarbonization pathways requires testing conditions that don't yet exist; synthetic data enables training AI models on futures rather than just historical patterns.

Q: How should we evaluate digital twin vendors for sustainability use cases? A: Prioritize four criteria: First, integration depth with your existing systems (MES, ERP, SCADA)—ask for reference customers with similar technical environments. Second, sustainability-specific capabilities including emissions modeling, LCA integration, and regulatory reporting features. Third, demonstrated validation approaches—how does the vendor ensure twin accuracy over time? Fourth, total cost transparency including integration services, ongoing maintenance, and the internal resources required for success. Request references from sustainability leaders, not just IT or operations.

Sources

• Fortune Business Insights, "Digital Twin Market Size, Share & Growth Report 2025-2032," January 2025
• Siemens Industrial AI Sustainability Survey, "AI-Powered Sustainability Gains," Q3 2025
• Mordor Intelligence, "Synthetic Data Market Size, Share, Trends & Research Report 2030," January 2025
• Gartner, "Predicts 2025: Synthetic Data and Privacy-Enhancing Technologies," November 2024
• Deloitte, "From Manufacturing to Medicine: How Digital Twins Can Unlock New Industry Advantages," 2024
• MarketsandMarkets, "Digital Twin Market 2025-2030," December 2024
• Unilever Climate Transition Action Plan, "Supplier Climate Programme Progress," 2024
• Siemens-Ecolab Climate Intelligence Joint Announcement, "Reducing Greenhouse Gas Emissions in Process Industries," 2024