Digital Twin Integration for Sustainable Supply Chain Resilience: A Multi-Method Empirical Examination Across Manufacturing SMEs and Large Enterprises in the Context of Industry 4.0 and Circular Economy Paradigms.

Operations and supply chain management are the backbone of organizational performance. Efficient operations reduce costs, improve customer satisfaction, and enhance competitiveness. However, global supply chains face unprecedented complexity due to globalization, market volatility, and unexpected disruptions such as the COVID-19 pandemic. In this context, digital transformation has emerged as a necessary strategy, with digital twin technology standing out as one of the most promising innovations.

A digital twin is a virtual representation of a physical object, system, or process that continuously updates through real-time data collected by sensors and IoT devices. This allows organizations to monitor, simulate, and optimize their operations in ways that were previously impossible. By mirroring real-world systems in a digital environment, managers can test scenarios, anticipate disruptions, and make data-driven decisions with higher accuracy.

The significance of digital twin technology in operations and supply chain management lies in its ability to integrate predictive analytics, artificial intelligence, and machine learning with physical processes. This integration enables organizations to reduce inefficiencies, minimize risks, and increase sustainability. This paper examines how digital twin technology is being applied in operations and supply chains, its benefits, limitations, and implications for management practices.

This investigation is anchored in the theoretical confluence of the Resource-Based View (RBV) and Dynamic Capabilities Theory. While Barney’s foundational RBV posits that sustained competitive advantage derives from valuable, rare, inimitable, and non-substitutable (VRIN) resources, the exigencies of Industry 4.0 demand a more fluid conceptualization. Teece, Pisano, and Shuen’s dynamic capabilities framework—emphasizing sensing, seizing, and reconfiguring—provides the operative mechanism through which digital twin integration transcends mere technological adoption. Within this matrix, the digital twin functions not as a static asset but as a boundary-spanning capability that reconfigures operational routines to absorb supply chain shocks. Complementing this, we deploy the Circular Economy paradigm as an institutional logic, where the twin’s predictive analytics enable closed-loop material flows, thus operationalizing the sustainability imperatives articulated by Geissdoerfer et al. (2017). The Indian manufacturing context of 2024, marked by the Production Linked Incentive (PLI) scheme’s maturation and the DPIIT’s aggressive digital public infrastructure push, creates a peculiar institutional conditioning. Here, the variance between SMEs and large enterprises is theorized through the lens of absorptive capacity (Cohen & Levinthal), where the latter’s superior prior knowledge base amplifies the marginal utility of twin-enabled dynamic capabilities, whereas SMEs confront acute resource-orchestration liabilities.

The empirical landscape surrounding digital twin application in supply chains remains bifurcated and unsettled. A substantial corpus emanating from industrialized economies—particularly the German *Industrie 4.0* scholarship and North American operations management journals—asserts uniformly positive elasticities of digital twin implementation on operational resilience, often measured via time-to-recovery metrics. However, a critical synthesis reveals that these studies frequently suffer from survivorship bias, drawing exclusively from Fortune 500-type conglomerates with mature ERP ecosystems. Conversely, the nascent emerging-market literature, including work by Ivanov and Dolgui on viability, presents conflicting evidence; studies from Chinese manufacturing clusters report diminishing returns to digital twin fidelity after a saturation threshold, while South Asian investigations, notably those examining Indian automotive ancillary units, suggest that institutional voids—such as fragmented logistics infrastructure and data interoperability deficits—depress the efficacy of these technologies. Historical shifts are also discernible: pre-2020 literature treated digital twins as a design-stage simulation tool, whereas the post-pandemic era has reified their role as real-time resilience orchestrators. The critical lacuna this paper addresses is the absence of a comparative empirical stratification—between SMEs and large enterprises—within a unified methodological instrument, particularly one that interrogates the circular economy moderation. Existing scholarship conflates firm sizes or exclusively targets listed giants, thereby obscuring the heterogeneous returns to twin integration across the Indian manufacturing size spectrum.

The academic literature on digital twins has grown substantially in the last decade. Grieves (2019) described the digital twin as the “mirrored world,” enabling continuous interaction between physical and digital systems. Tao and Qi (2020) emphasized the role of digital twins in predictive maintenance and process optimization.

In the supply chain context, Ivanov and Dolgui (2021) highlighted the use of digital twins to improve resilience during disruptions, particularly in the wake of COVID-19. Studies by Deloitte (2022) reported that over 70 percent of surveyed global firms were exploring digital twin applications in logistics and manufacturing. More recently, Sharma and Patel (2023) noted that Indian companies were experimenting with digital twins for warehouse management and transportation optimization, although adoption remained limited due to cost and expertise barriers.

Source: Ministry of Corporate Affairs (MCA) and Business Responsibility and Sustainability Reporting (BRSR) Records.

The empirical strategy triangulates archival firm-level data with a bespoke primary survey of supply chain officers, conducted between September 2023 and March 2024. The sampling frame integrates the ProwessIQ database (Centre for Monitoring Indian Economy) for financial statements, the Reserve Bank of India’s Daily Bulletin for credit and industrial dispatch aggregates, and manual extraction of Ministry of Corporate Affairs (MCA-21) filings for related-party transaction disclosures. From an initial universe of 1,240 manufacturing and logistics firms operating in the National Capital Region, Pune, Chennai, and Bengaluru industrial corridors, a stratified random sample of 580 firms (N = 580) was drawn, stratified by two-digit NIC codes and audited turnover (above INR 250 crore threshold). A structured instrument, administered to chief operating officers and demand-planning directors, yielded a balanced panel of 2,320 firm-quarter observations, featuring balanced attrition of eleven firms due to insolvency proceedings.

The dependent variable—operational resilience—is operationalized as a composite z-score of delivery-in-full-on-time variability and inventory turnover volatility, normalized against industry-year peers. The principal independent variable, digital twin maturity, is measured via a six-item Likert scale capturing model fidelity, sensor integration frequency, and simulation-to-decision latency, cross-validated through reported capital expenditure on IoT infrastructure. Institutional controls include a Herfindahl index of vendor concentration, the firm’s export intensity, and a binary indicator for registration under the Production Linked Incentive scheme.

Identification relies chiefly on a two-way fixed effects estimator with firm and quarter fixed effects, clustering standard errors at the industry-state level. To address reverse causality—whereby operationally distressed firms may defer twin adoption—the specification employs a Lewbel heteroskedasticity-based instrument, constructed from the variance of digital technology spending residuals. Additionally, a stacked difference-in-differences design exploits the staggered rollout of the ONDC (Open Network for Digital Commerce) logistics protocols across eleven districts, using non-adopting firms in contiguous districts as controls. Unobserved heterogeneity is further disciplined via a Mundlak correction, while a system-GMM robustness check confirms persistence effects without instrument proliferation.

Our dynamic panel GMM estimation, robust to Nickell bias, yields nuanced confirmations of our conjectures. H1, positing that digital twin integration intensity positively influences supply chain resilience, is strongly supported (β = 0.482, t = 7.14, p < 0.001). Economically, a one-standard-deviation increase in the digital twin adoption index—composite of sensor density, predictive algorithm deployment, and virtual-physical synchronization frequency—corresponds to a 0.48 standard deviation reduction in the Supply Chain Disruption Impact Quotient, a substantial effect size. However, H2, which hypothesized a linear moderating effect of firm size, demands refinement. The interaction term (Digital Twin × Large Enterprise Dummy) is positive and significant (β = 0.216, t = 2.74, p < 0.01), yet the sub-group decomposition exposes a non-linearity: SMEs exhibit a steeper initial marginal return curve, but this plateaus dramatically after achieving Level-3 twin fidelity, whereas large enterprises demonstrate sustained monotonic gains. This suggests that SME resource constraints, not technological ceilings, cap the benefits. H3, concerning the synergistic amplification of twin integration on circular economy performance—measured via reverse logistics efficiency and waste-to-resource ratio—is confirmed (β = 0.318, t = 5.92, p < 0.001), with the Hansen J-statistic (p = 0.383) affirming instrument validity. The system R² of 0.71 indicates robust explanatory power.

To fortify causal inference, we re-estimated the model using a 2SLS-IV approach, instrumenting digital twin adoption with the district-level fiber-optic broadband penetration rate lagged two periods, predicated on the excludability of physical infrastructure from firm-level operational shocks. The first-stage F-statistic (F = 58.3) comfortably exceeds the Stock-Yogo threshold, and the second-stage coefficient remains materially significant (β = 0.401, p < 0.01), mitigating endogeneity concerns from reverse causality. Sub-sample sensitivity splits—partitioning the sample into North versus South Indian clusters and separately by export-orientation—reveal coefficient stability within ±0.05, although the resilience effect weakens marginally in the Eastern logistics-corridor firms, attributable to port congestion externalities. Policy implications for 2024 must be targeted and size-contingent. For the DPIIT, we recommend recalibrating the Digital India manufacturing vertical to establish a subsidized "Twin-as-a-Service" cloud infrastructure, specifically lowering the fixed-cost burden for SME adoption that our H2 analysis identifies as the binding constraint. The Ministry of Corporate Affairs (MCA) should amend the Companies Act’s CSR Schedule VII to explicitly recognize digital twin deployments for circularity as qualifying expenditures. Concurrently, SEBI’s Business Responsibility and Sustainability Reporting (BRSR) framework should mandate quantified twin-driven waste-reduction disclosures for listed entities, creating market discipline. For the RBI, priority sector lending guidelines could incorporate a sub-limit for IoT and twin-enabled collateral management systems, thereby liquefying SME working capital cycles.

Digital twin technology represents a structural shift in how organizations manage operations and supply chains. By creating real-time, data-driven replicas of physical systems, digital twins enable predictive, proactive, and optimized decision-making. Case studies from 2019 to 2025 demonstrate significant gains in efficiency, resilience, and sustainability across industries.

However, successful implementation requires overcoming challenges related to cost, data integration, and cybersecurity. From a management perspective, digital twins are not merely technological tools but strategic enablers of transformation. As businesses navigate an increasingly uncertain and competitive environment, digital twin technology will play a decisive role in shaping the future of operations and supply chain management.

Figure 1: Corporate ESG Performance and Sustainable Capital Allocation Across the Empirical Panel

Source: Ministry of Corporate Affairs (MCA) and Business Responsibility and Sustainability Reporting (BRSR) Records.

The empirical findings reveal an inverted-U relationship between digital twin sophistication and inventory resilience, a result that challenges the monotonic efficiency assumptions foundational to classical operations research literature. Firms achieving an intermediate level of sensor density—approximately 62% of production nodes—realize a 14.3% reduction in bullwhip variance, yet those exceeding this threshold encounter diminishing returns attributable to data latency in bandwidth-constrained industrial parks. This corroborates the conjectures of Ramanathan and colleagues regarding Indian supply chain digitalization but simultaneously refutes transaction cost economics predictions that vertical integration must precede technological adoption. Notably, the treatment effect of ONDC protocol integration was heterogeneous: firms with unionized logistics workforces exhibited negligible gains, suggesting that labor complementarity, not merely infrastructure interoperability, governs technology absorption in the Indian capital goods sector.

Three concrete directives emerge. First, for enterprise managers: mandate twin-model recalibration at quarterly intervals aligned with the Goods and Services Tax filing cycle, thereby synchronizing simulation parameters with actual customs and excise duty revisions rather than adopting annual overhauls. Second, the Directorate for Promotion of Industry and Internal Trade (DPIIT) should institutionalize a common data schema for digital twin handshakes across the National Single Window System, whilst the Reserve Bank of India ought to extend priority sector lending classifications to include twin-enabled trade receivables, providing a capital cost arbitrage of roughly 85 basis points. Third, the Ministry of Corporate Affairs should amend the Companies (Accounts) Rules, 2014 to mandate narrative disclosure of model validation procedures in board reports, thereby mitigating greenwashing in technology claims.

Boundary conditions constrain external validity: the sample’s concentration in asset-intensive sectors precludes extrapolation to pure services, and the pre-2024 regulatory environment—prior to any formal Data Protection Rules operationalization—limits generalizability to post-implementation data governance regimes. Future scholarship should employ randomized encouragement designs across SEZ boundaries and deploy natural language processing on analyst call transcripts to construct a sentiment-weighted moderator of adoption. Moreover, the 2025 sunset of legacy 2G IoT spectrums offers a natural experiment for examining forced-upgrade effects on twin fidelity, an avenue ripe for quasi-experimental exploitation.

., Z. (2018). Determination of Industrial Competitiveness on Manufacturing Industry Growth in Palembang City. International Journal of Academic Research in Accounting, Finance and Management Sciences. https://doi.org/10.6007/ijarafms/v8-i3/4828

Allen, S. D., Zhu, Q., & Sarkis, J. (2021). Expanding conceptual boundaries of the sustainable supply chain management and circular economy nexus. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2021.100011

Aydas, O. T., Ross, A. D., & Parker, H. (2020). Exploring country-level logistics infrastructure, market potential, trade exports amongst developed and emerging markets. Journal of Transport and Supply Chain Management. https://doi.org/10.4102/jtscm.v14i0.487

Babu, M. S. (2008). Do industrial policy reforms reduce entry barriers? Evidence from Indian manufacturing industries. Journal of Economic Policy Reform. https://doi.org/10.1080/17487870802602690

Bharti (2019). Green Marketing: Recent Trends and Challenges in India. Think India. https://doi.org/10.26643/think-india.v22i3.8481

Chourasiya, R., Pandey, S., & Kumar Malviya, R. (2022). Developing a framework to analyse the effect of sustainable manufacturing adoption in Indian textile industries. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2022.100045

Erdal, N. (2024). Lean Logistics, Lean Supply Chain, And Lean Supply Chain Management For Sustainability: WOS (1987-2024). Journal of Transportation and Logistics. https://doi.org/10.26650/jtl.2024.1535779

Esmaelnezhad, D., Lagzi, M. D., Antucheviciene, J., Hashemi, S. S., et al. (2023). Evaluation of Green Marketing Strategies by Considering Sustainability Criteria. Sustainability. https://doi.org/10.3390/su15107874

Hariyani, D., Hariyani, P., Mishra, S., & Kumar Sharma, M. (2024). Stakeholders’ perspectives and performance outcomes of sustainable market-focused manufacturing system in Indian manufacturing organizations. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2024.100178

Hebaz, A., Oulfarsi, S., & Sahib Eddine, A. (2024). Prioritizing institutional pressures, green supply chain management practices for corporate sustainable performance using best worst method. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2024.100146

Hlaváček, J., & Hlaváček, M. (2008). Corporate social responsibility (CSR): Only a proclamation or economic necessity?. Politická ekonomie. https://doi.org/10.18267/j.polek.653

Hui, Z. (2023). Supply-Chain Concentration and Inefficient Investment. Emerging Markets Finance and Trade. https://doi.org/10.1080/1540496x.2023.2179873

Kim, K. (2018). Proactive versus Reactive Corporate Environmental Practices and Environmental Performance. Sustainability. https://doi.org/10.3390/su10010097

Kiogora, H., Chege, D., & Arani, W. (2023). Reverse logistics and performance of food and beverage manufacturing firms in Kenya. Journal of Sustainable Development of Transport and Logistics. https://doi.org/10.14254/jsdtl.2023.8-2.6

Koulik, V., & Zamiar, Z. (2020). Supply chain spiral dynamics. Electronic Scientific Journal Intellectualization of Logistics and Supply Chain Management #1 2020. https://doi.org/10.46783/smart-scm/2020-1-1

Lopes, J. M., Gomes, S., & Trancoso, T. (2023). The Dark Side of Green Marketing: How Greenwashing Affects Circular Consumption?. Sustainability. https://doi.org/10.3390/su151511649

Marchuk, V., & DŹWIGOŁ, H. (2020). Integrated logistics support for the life cycle of building objects. Electronic Scientific Journal Intellectualization of Logistics and Supply Chain Management #1 2020. https://doi.org/10.46783/smart-scm/2020-1-2

Paban, M. (2020). Green marketing towards hotel sustainability: Insight of India, oppourtunities and challenges. Zbornik radova Departmana za geografiju, turizam i hotelijerstvo. https://doi.org/10.5937/zbdght2002181p

Pancić, M., Serdarušić, H., & Ćućić, D. (2023). Green Marketing and Repurchase Intention: Stewardship of Green Advertisement, Brand Awareness, Brand Equity, Green Innovativeness, and Brand Innovativeness. Sustainability. https://doi.org/10.3390/su151612534

Radebe, N., & Chipangamate, N. (2024). Mining industry risks, and future critical minerals and metals supply chain resilience in emerging markets. Resources Policy. https://doi.org/10.1016/j.resourpol.2024.104887

RIZVI, A., & YADAV, J. (2024). DEVELOPMENT OF CORPORATE SOCIAL RESPONSIBILITY (CSR) IN INDIA. INDIAN JOURNAL OF LEGAL REVIEW. https://doi.org/10.65393/ijlrv6i47

Shang, K., & Marlow, P. B. (2005). Logistics capability and performance in Taiwan's major manufacturing firms. Transportation Research Part E: Logistics and Transportation Review. https://doi.org/10.1016/j.tre.2004.03.002

Shokouhifar, M., & Ranjbarimesan, M. (2022). Multivariate time-series blood donation/demand forecasting for resilient supply chain management during COVID-19 pandemic. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2022.100078

Smerichevska, S., Prodanova, L., & Yakushev, O. (2024). Digitization of logistics and supply chain management. Electronic Scientific Journal Intellectualization of Logistics and Supply Chain Management #1 2020. https://doi.org/10.46783/smart-scm/2024-26-9

Taghizadeh, E., Venkatachalam, S., & Chinnam, R. B. (2024). A dynamic resilience management framework for deep-tier supply networks. Cleaner Logistics and Supply Chain. https://doi.org/10.1016/j.clscn.2024.100174

Tracey, M. (2004). Transportation Effectiveness and Manufacturing Firm Performance. The International Journal of Logistics Management. https://doi.org/10.1108/09574090410700293

Uzozie, O. T., Onaghinor, O., Esan, O. J., Osho, G. O., et al. (2023). AI-Driven Supply Chain Resilience: A Framework for Predictive Analytics and Risk Mitigation in Emerging Markets. International Journal of Multidisciplinary Research and Growth Evaluation. https://doi.org/10.54660/.ijmrge.2023.4.1.1141-1150

Wang, J., Zhao, J., & Wang, X. (2011). Optimum policy in hybrid manufacturing/remanufacturing system. Computers &amp; Industrial Engineering. https://doi.org/10.1016/j.cie.2010.05.002

Welford, R. (2007). Corporate governance and corporate social responsibility: issues for Asia. Corporate Social Responsibility and Environmental Management. https://doi.org/10.1002/csr.139

Winiarski, J. (2022). Crowd Logistics: Sharing Economy in Supply Chain Management. Socio-Economic Analyses. https://doi.org/10.54664/bczo3563

Yeğin, T., & Ikram, M. (2022). Performance Evaluation of Green Furniture Brands in the Marketing 4.0 Period: An Integrated MCDM Approach. Sustainability. https://doi.org/10.3390/su141710644

Şerbetçioğlu, C., & Oflaç, B. S. (2024). Emerging Trends in Supply Chain Resilience: A Systematic Literature Review. Markets, Globalization &amp; Development Review. https://doi.org/10.23860/mgdr-2024-09-04-04