Digital Twins in Sustainable Supply Chains: A Comprehensive Review of Current Applications and Enablers for Successful Adoption ^†

Abstract

Digital Twins (DT) are an emerging trend in diversified industrial sectors and their value chains. This study aims to explore current DT applications within SCs, focusing on sustainable inbound and outbound logistics and identifying key enablers that will facilitate the successful adoption of DTs in SCs. Using the PEO model and employing the PRISMA framework, this study screened articles from the Scopus database to explore the existing knowledge related to this topic. A steep increase in articles related to DTs over the past 10 years indicates that there is growing attention in exploring and leveraging this technology for various applications. The successful adoption of DTs is driven by several key factors, including advanced technological infrastructure; standardized processes; continuous improvement; knowledge workers; technologies like IoT, IIoT, AR/VR; and managerial support.

1. Introduction

A supply chain (SC) is a complex, global network of diverse, independent, and interconnected organizations collaborating to create customer value. An SC encompasses multiple elements belonging to various stakeholders and includes every step involved in delivering a finished product or service to the customer. There are six primary processes in an SC: plan, source, make, deliver, return, and enable [1]. These processes are cross-functional across different business stakeholders and coordinate the flow of materials, information, and other resources within the SC. The structural hierarchy of the SC is inherently complex, consisting of several operational levels with multiple stakeholder involvements and high interoperability [2].

Due to this complex and interoperable nature, organizations rely heavily on digital transformation to optimize performance in supply chains (SCs). Digital transformation is defined as “a strategic initiative that incorporates digital technology across all areas of an organization” [3]. Within this framework, organizations are exploring emerging technologies in the context of Supply Chain Management (SCM) to remain competitive in the global market. Consequently, supply chains are now adopting the fourth wave of the industrial revolution, known as Industry 4.0. Industry 4.0 involves a combination of emerging technologies, such as cyber-physical systems (CPS), the Internet of Things (IoT), radio-frequency identification (RFID), big data, artificial intelligence, smart sensors, cloud computing, 3D printing, and additive manufacturing [4].

Digital Twins (DTs) play a crucial role in Industry 4.0. A DT is a digital representation of virtually anything. The National Aeronautics and Space Administration (NASA) initially created the concept of DTs to model real-world spacecraft situations [5]. In the industrial context, the terminology has different variants, including Digital Twins, Digital Models, and Digital Shadows, which differ based on the level of data integration between the physical entity and its virtual representation. A DT typically involves three components: (1) a physical entity; (2) a virtual entity, usually the digital representation of the physical entity; and (3) a connecting layer that enables bi-directional communication between the physical and virtual entities. Digital Twins are crucial for sustainable mobility in sustainable supply chains because they create real-time, virtual models of physical assets, processes, and systems. This innovation allows for enhanced monitoring, analysis, and optimization of the supply chain. With Digital Twins, businesses can simulate and predict the outcomes of various scenarios, improving the efficiency of goods movement and reducing waste. They aid in planning optimal routes, reducing energy consumption, and minimizing emissions, thereby supporting more sustainable and efficient mobility within the supply chain [6]. This study solely focuses on Digital Twins (DTs), where the information flow is bi-directional between virtual and physical entities.

The use of Digital Twins (DTs) in industries beyond space has been a trending topic over the past decade, with manufacturing taking a prominent role. Manufacturing firms now have a high maturity rate in DT adoption, which, in turn, influences supply chains (SCs). Utilizing DTs in SCs allows for the creation of a virtual simulation model of entire supply chain processes, providing end-to-end visibility [7]. However, expanding the use of DTs from manufacturing to SCs has been challenging due to the complexity of SC elements. Before implementing DTs in SCs, it is crucial to identify the key enablers and success factors for successful adoption. This undertaking helps organizations understand the potential benefits and pinpoint the areas that require focused attention.

Therefore, this research aims to explore the current applications of Digital Twins (DTs) in Supply Chain Management (SCM) and identify the key enablers and success factors for implementing and successfully adopting Digital Twin concepts in supply chains.

The research objectives are as follows:

• To identify how Digital Twins enhance the capabilities of Supply Chain Management;
• To identify key enablers and success factors for successful Digital Twin adoption in supply chains.

The research questions are as follows:

RQ1.

What are the current applications of Digital Twins in Supply Chain Management that facilitate the delivery of value-added products to the market?

RQ2.

What are the key enablers and factors affecting the successful adoption of Digital Twins in a supply chain context?

2. Materials and Methods

This study aims to address the current literature gap between the fields of Digital Twin (DT) applications and Supply Chain Management (SCM) by consolidating DT trends in SCM and identifying critical enablers for successful DT adoption in supply chains (SCs). To achieve this, the researchers searched the existing literature and critically evaluated it using the systematic literature review (SLR) methodology, followed by a mapping study and a narrative summary. This study employed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) framework for a methodological and transparent SLR. Following this framework, SCOPUS was selected as the primary database for the literature search. Several keywords were identified through an initial overview search and then organized based on the Population, Exposure, and Outcome (PEO) framework. Although this model is frequently used in health sciences [8], it provided a qualitative framework to organize and conduct the keyword search for this study. While our primary focus is Supply Chain Management, manufacturing was also considered due to the interoperability between elements in these two fields.

Based on predefined inclusion and exclusion criteria, the literature identification was narrowed down to 292 articles for screening. Analyzing the publication pattern of these articles showed a clear increase in research on Digital Twins (DTs) in supply chains (SCs). Compared to 2017, when only one article was published, there was a nearly 30% increase in publications by 2023. This increase indicates growing interest in this field and suggests that SCs are becoming more digitalized. These 292 articles were initially screened for relevance based on their titles, which filtered down to 206 for abstract screening. During this phase, 23 duplicate articles were found and removed. Subsequently, 122 articles were selected for a full-text critical review. Of these, 24 articles were excluded because they focused on the software architecture of Digital Twins and associated systems. Additionally, 14 articles were identified as extended works by the same authors, and 42 articles were excluded for not being domain specific. This process resulted in 42 articles being included in the literature review.

3. Results

3.1. Current Applications of DT in SC

The prominent objectives of using Digital Twins (DTs) in supply chains (SCs) include visibility and monitoring, optimization, prediction, and simulation [9]. For example, a heuristic optimization algorithm running in a DT of a warehouse optimized material handling [9]. Additionally, architectural and functional structures for using DTs to support operational control in order-picking systems have been presented [10]. In the material handling process, a DT framework has been used to analyze worker fatigue during material handling motions [11]. While several simulation approaches address the complexity of automated storage and retrieval systems (AS/RSs), many models do not rely on real-time data, limiting the scope of digitization [12]. A DT-based approach to creating a discrete event simulation model, integrating both the physical system and the information technology architecture, was developed, allowing for the validation of simulation models using a DT [13]. Human–robot interfaces operating on DTs of a warehouse can guide and navigate automated robots through secluded workstations [14]. DT-based AS/RS systems can collect data from industrial processes, mimic them in the warehouse’s DT, and feed the information back to the AS/RS system after processing the simulation-based digital model. This method has increased the scalability and reliability of AS/RSs [15].

By developing a location management system built upon the DT of the warehouse, a simulation-based approach to optimize internal transport within storage locations was introduced [14]. Automated guided vehicles (AGVs) are used in warehouses to improve transport efficiency. An AGV scheduling system based on a DT of the warehouse can collect real-time data and enhance the efficiency of the logistic system inside the warehouse [16]. A dynamic DT-based AGV scheduling system addresses the charging problem by reducing energy consumption by 1.32% [17]. Last-mile logistics (LMLs) involve high organizational costs and time inefficiencies, leading to customer dissatisfaction and organizational losses, and account for 25% of total GHG emissions. DT-based solutions can optimize last-mile routing, assess these transport modes’ environmental and monetary values, and make LMLs greener [18]. Based on past data, DTs can evaluate the performance of suppliers and contractors, optimize transport routes to improve distribution efficiency, and enable the real-time scheduling and monitoring of transport vehicles [19]. Intelligent logistic distribution systems are a growing trend in the era of AI. A simulation assignment introduces visual sensor imaging technology based on DTs, which captures documents and receipts related to order receipt, distribution, and return management, effectively tracking and monitoring information along the logistic distribution process [20]. E-commerce has allowed many companies to excel in their businesses; however, the distribution networks used in this category, multi-echelon supply chain networks, often face challenges in network planning. A study introduced a DT-based 4PL planning and scheduling model, enhancing transportation flow and resource utilization while shortening waiting times within this multi-echelon supply network [21].

3.2. Key Enablers and Success Factors

Despite the growing application of Digital Twins (DTs) in Supply Chain Management (SCM), it is necessary to identify the key enabling factors for successful DT adoption in supply chains—various categories of DT enablers impact process industries [22]. AI, IoT, and VR/AR have been identified as Industry 4.0-related DT enablers. AI and machine learning (ML) are major driving factors that enhance the functionalities of DTs, enabling them to operate at full capacity [23,24]. In processing documents and images related to supply chain functions, AI tools like Microsoft AI Builder are widely used in industries as they require minimal developer knowledge. The Internet of Things (IoT) and the Industrial Internet of Things (IIoT) are other key enablers in supply chain DT applications [25]. A core requirement of a DT is communication between physical entities, which IoT/IIoT facilitates by collecting and transmitting information [26]. Operators in a supply chain environment can use virtual reality (VR) and augmented reality (AR) for various purposes, such as conducting simulations in a virtual environment, virtual commissioning, remote assistance, and gaining a deeper understanding of the production process [27,28]. Also, DT enablers are identified as proper hardware, communication technologies, and development technologies like blockchain, open-source software, and centralized databases [24,29].

Critical enablers for successful DT adoption in supply chains include network, collaboration, trust, sophisticated planning, ICT/ITS technologies, physical infrastructure, legal and political framework, awareness, mental shift, and pricing/cost/services [30]. According to research, having proper traceability software, using appropriate data analytics practices, and maintaining optimization and simulation integration platforms are critical success factors for adopting and implementing DTs in supply chains [31]. The key success factors/enablers of DT adoption are categorized into four themes: economic factors, technological factors, ethical factors, and operational factors. Economic enablers include the availability of funds and financial investments and having a profitable sharing model with supply chain partners. Technological enablers encompass access to state-of-the-art technologies, high data security and management protocols, standard information-sharing methods like ERP systems, and access to real-time data analytics. Ethical enablers include maintaining proper relationships with supply chain partners to encourage change, improving the commitment among SC partners for DT adoption, and making necessary cultural alignments between SC entities to embrace new digital initiatives. Operational factors considered key for successful DT adoption in SCM include top management commitment and support, extended enterprise, long-, medium-, and short-term planning, and proper risk mitigation strategies. Organizational competencies such as knowledge building through reskilling and upskilling the workforce, knowledge-based intelligent skills, and asset modeling are also identified as key enablers for successful DT adoption in process industries [32]. Furthermore, organizational practices like standardized processes, continuous improvements, process designs based on simulations, and optimization techniques also act as enablers in the DT adoption journey [33].

4. Conclusions

Digital Twins (DTs) are being integrated into Supply Chain Management (SCM), acting as major catalysts for digitalization with a notable emphasis on sustainable logistics and transport. DTs provide comprehensive capabilities to solve critical SCM problems such as visibility, monitoring, optimization, prediction, and simulation. The significant use of DTs in the SCM sector covers various areas, including warehouse management, material handling, order-picking systems, and automated storage and retrieval systems. While DTs offer promising solutions, integration, complexity, and scale, challenges require a nuanced understanding and strategic approach for implementation. The multifaceted nature of implementing DTs within the SCM context is highlighted by the identification of key enabling factors for their successful adoption. These include advances in Industry 4.0 technologies, such as Artificial Intelligence (AI), the Internet of Things (IoT), and Virtual Reality (VR), as well as organizational competence, ethical considerations, and operational factors. To realize the full benefits of DTs, one must adopt a holistic approach that considers technological, managerial, and ethical dimensions while addressing challenges related to their integration and deployment within supply chains.