Unlocking the potential of digital twins to achieve sustainability in seaports: the state of practice and future outlook

Abstract

This paper examines the role of digital twins (DTs) in promoting sustainability within seaport operations and logistics. DTs have emerged as promising tools for enhancing seaport performance. Despite the recognized potential of DTs in seaports, there is a paucity of research on their practical implementation and impact on seaport sustainability. Through a systematic literature review, this study seeks to elucidate how DTs contribute to the sustainability of seaports and to identify future research and practical applications. We reviewed and categorized 68 conceptual and practical digital applications into ten core areas that effectively support economic, social, and environmental objectives in seaports. Furthermore, this paper proposes five preliminary potential applications for DTs where practical implementations are currently lacking. The primary findings indicate that DTs can enhance seaport sustainability by facilitating real-time monitoring and decision-making, improving safety and security, optimizing resource utilization, enhancing collaboration and communication, and supporting the development of the seaport ecosystem. Additionally, this study addresses the challenges associated with DT implementation, including high costs, conflicting stakeholder priorities, data quality and availability, and model validation. The paper concludes with a discussion of the implications for seaport managers and policymakers.

1 Introduction

Maritime transportation plays a pivotal role in globalization and international trade, facilitating the worldwide export of various goods. Despite being a relatively slow mode of transportation, it remains popular for intercontinental shipments due to its high capacity, cost-effectiveness, and lower emissions (Acciaro et al. 2014; Homayouni and Fontes 2018). Seaports serve as critical nodes connecting sea and land transport, and even marginal improvements in their performance can lead to significant cost savings, emission reductions, and improved social well-being.

Ensuring the sustainability of seaport operations and logistics requires addressing environmental impacts, supporting local communities, and maintaining economic viability. This complex task extends beyond the capabilities of any single organization and necessitates a multidisciplinary approach involving collaboration between government, industry, academia, and civil society (Hossain et al. 2021; Del Giudice et al. 2022; Fobbe and Hilletofth 2021; Gerlitz and Meyer 2021). Seaports play a crucial role in driving sustainable development for themselves and the broader communities they serve. This involves developing innovative technologies, implementing effective policies and regulations, and promoting collaboration and information sharing (see, e.g., Belmoukari et al. 2023; Heikkilä et al. 2022; Jović et al. 2022; Alzahrani et al. 2021). One such innovative technology is Digital Twin (DT). DT technology has emerged as a vital tool for comprehensively and holistically modelling physical assets or systems (Yang et al. 2024; Agostinelli et al. 2022; Attaran and Celik 2023; Wang et al. 2021). A DT can be seen as an advanced decision support system, providing real-time insights and predictions based on data collected from the physical system (Meierhofer and West 2020).

While DT technology has shown promise in various domains, from manufacturing to logistics (Loaiza and Cloutier 2022; Perno et al. 2022; Madusanka et al. 2023), its development is still far from mature. There is no consensus on the design and specifications of a DT for complex systems like seaports (see discussions in recent reviews, e.g., in Wang et al. 2021; Klar et al. 2023; Neugebauer et al. 2024), and its application is still evolving. Recent works emphasize the importance of DTs for seaports, and seaport professionals worldwide have acknowledged their potential (Port Technology International 2022, 2018), investing more in DT development projects. Surveillance and operations management are identified by experts as two urgent applications (Choi and Yu 2020), while comprehensive reviews cover a wide range of areas, such as cargo handling, transportation management, terminal operations management, data communication, risk prediction, resilient operations, and environmental protection (Wang et al. 2021; Neugebauer et al. 2024; Zhou et al. 2021).

While the potential of DT technology in seaports is widely recognized, its specific contributions to seaport sustainability remain unclear which underscores the significance of this study. This study seeks to address this gap by conducting a comprehensive literature review to investigate both established and emerging applications of DTs in seaports. The key research questions guiding this study are:

• How do digital twins contribute to the sustainability of seaport operations and logistics?

• What are the possible future research and practice gaps?

• What are the challenges and potential applications associated with their implementation?

Building on previous review papers (Wang et al. 2021; Klar et al. 2023; Neugebauer et al. 2024; Zhou et al. 2024), this study broadens their scope and proposes several innovative applications for DTs in seaport sustainability. These novel applications aim to provide researchers and practitioners with valuable avenues for future DT implementations in seaports.

The significance of this study lies in the growing significance of seaports in global trade and the increasing need to improve their sustainability in the face of various challenges, such as climate change, disruptive events, and the greater use of renewable energy resources (RESs). Furthermore, seaports help to improve the sustainability of global supply chains due to their significant role in global distribution networks. Therefore, this paper makes three main contributions:

• It thoroughly examines the existing literature on how DTs are used in seaports to promote sustainability

• It investigates the potential applications of DTs in areas often overlooked, such as stakeholder management, energy management, and waste management

• It highlights the challenges and opportunities associated with implementing DTs in seaports, providing valuable guidance for future research and development in this field.

The remainder of the paper is divided into nine main sections. First, the methodology of this work is described in Section 2, then, in Section 3, DT technology and its implementation frameworks in seaports are introduced. A holistic overview of the literature review and a framework for promoting sustainability in seaports through DTs is presented, in Section 4. Section 5 focuses on applications of DT-based systems to promote operational efficiency and economic viability in seaports, Section 6 reviews the applications of DTs to promote social goals in seaports, and Section 7 examines the applications of DTs for environmental management in seaports. Furthermore, Section 8 discusses the challenges and barriers to DT implementation in seaports and Section 9 proposes and discusses five potential applications of DT-based systems and outlines future research lines in these areas. Finally, Section 10 concludes the paper based on the insights and findings presented.

2 Review scope and methodology

To address the research questions in this study, we adopted a Systematic Literature review based on content analysis. As stated (Tranfield et al. 2003), systematic literature reviews are appropriate to develop a knowledge base within a field and provide a reliable basis to formulate new research actions. Therefore, motivated by the works of Tranfield et al. (2003), Seuring and Gold (2012), a three-stage research methodology is proposed here as depicted in Fig. 1. The systematic literature review not only ensures a comprehensive coverage of all relevant literature allowing for replication by other researchers, but also makes it easier to identify key research gaps.

In the first stage, we defined the purpose and scope of the study. Then, we conducted a heuristic search for recent literature reviews and technical papers and a series of brainstorming sessions to understand the boundaries of digital twins and sustainability practices, particularly in a complex system like seaport operations and logistics. The result of this step was a definition of digital twin technology and its main characteristics (further discussed in Section 3) and extracting relevant keywords.

Given the extensive volume of literature, this research narrows its focus to scholarly articles published in English that delineate the deployment of DT technology within seaport operations and logistics. The scope of our literature review was confined to peer-reviewed journal articles and book chapters indexed in bibliographic databases. Moreover, only documents that have been published or formally accepted for publication were included. The literature search used two of the most established bibliographic repositories: Web of Science™ and Scopus^®. Furthermore, the temporal parameters of our search were set to encompass works published from the year 2010 onwards. It is noteworthy that, despite this temporal range, the initial relevant publications on the subject emerged in the year 2019.

In the second stage, we searched for publications that included the following combinations of generic keywords in their title, abstract, or list of keywords:

• (“digital twin” OR “smart port") AND

• (“seaport” OR “container terminal” OR “maritime" OR “marine")

The data gathering stage consists of the following four steps:

• 1. Structured search: Bibliographic databases were searched using specific keywords and considering the date range from 2010 onwards;

• 2. Data refinement: The sets of records resulting from the structured search were then combined, screened, and cleaned as explained below;

• 3. Search expansion: Additional data gathering through backward/forward reference search; and

• 4. Practical applications:Added those applications that were not reported in a research publication.

Fig. 1

The three stages of the research methodology according to a systematic literature review

The literature search was conducted in June 2024. Initially, we gathered 261 and 254 entries from the bibliographic databases Web of Science™ and Scopus®, respectively. The data refinement process involved several steps. We began by merging the two sets of results and removing duplicate entries based on their Digital Object Identifier (DOI), title, or abstract similarity, resulting in 323 unique papers. In the first screening round, we reviewed the titles and excluded 135 papers, such as those related to shipbuilding yards, marine navigation, underwater vehicles, and offshore infrastructure.

In the second screening round, we skimmed the abstracts and excluded another 76 papers. By the end of this round, 112 records remained. During abstract skimming, we looked for indications of digital twin usage and applications related to sustainability or its pillars. In the final screening round, we read the full papers of these 112 works, ultimately excluding another 77 papers that did not align precisely with the study’s scope.

From the remaining 35 works, including 10 review papers, we expanded our search using backward and forward reference searches based on the references cited in recent literature review papers (e.g., Wang et al. 2021; Neugebauer et al. 2024; Zhou et al. 2024). This step added 24 more works to our review. Additionally, we expanded our search to include use cases cited in review papers but not described in research papers, as well as those found through heuristic web searches, increasing the total to 68 applications and use cases.

In the final stage, we conducted a content analysis. We thoroughly analyzed and synthesized all theoretical and practical applications to extract their key features. A preliminary categorization scheme was developed based on these key features, focusing on ten core sustainability applications (in Section 4). This framework was evaluated through peer discussions and brainstorming sessions, which helped identify research and application gaps, propose future research directions, and define digital twin implementation challenges, particularly in seaport operations and logistics.

3 Digital twin technology

The concept of the digital twin was first introduced as a living model during NASA’s Apollo mission in 1960. It was officially presented by Dr. Michael Grieves at the University of Michigan in 2002 (Grieves and Vickers 2017). The process of DT involves creating a virtual replica of a physical object or system. This enables continuous and automatic exchange of data between the physical and virtual systems. It comprises a model that mirrors the physical system, bidirectional data flow, and frequent updates; and has recently gained widespread adoption in various sectors (Loaiza and Cloutier 2022; Perno et al. 2022; Madusanka et al. 2023). DTs undergo classification across five maturity levels (Klar et al. 2023): reality capture and replication, connection, synchronization, interaction, and autonomous empowerment.

Implementing DTs in seaports demands tailored frameworks that accommodate port operations and logistics’ unique characteristics and challenges. While some similarities exist with other applications like manufacturing and logistics, the nature of port operations requires distinct frameworks. A recent definition of a DT for seaports has been proposed as “a grouping of models and algorithmic components that jointly describe the complex interplay of port processes and operations allowing the characterization, estimation, and prediction of the most efficient operations at the process level, but also for the port as a whole” (Klar et al. 2023; Kaklis et al. 2023).

Selecting the appropriate DT characteristics depends on factors such as port size, system complexity, data availability, and stakeholder requirements. Essential requirements for seaport authorities and terminal operators include high concurrency, real-time feedback, and minimal system delay, typically no more than 0.2 seconds (Yang et al. 2024). Effective DT implementation in seaports requires data fusion from heterogeneous sources, integration of physical and virtual data, and seamless connection with external systems and stakeholders like shipping companies, agencies, cargo owners, and customs offices.

Several frameworks have been developed for implementing DTs in seaports. For instance, a five-layer framework for DT deployment in port management is proposed in Wang et al. (2021), comprising physical, data, model, service, and application layers. This framework enables operational management through DT-based models and platforms, supporting functions such as cargo handling and transportation, container operation and storage, data communication and sharing, risk prediction, and environmental protection in smart ports.

At Qingdao Port, a hybrid bi-level DT framework for managing container terminal operations has been developed (Yang et al. 2024). The framework operates across two layers - physical and digital. The central control system in the physical layer receives client orders and allocates resources accordingly, while the digital layer provides terminal operators with tools to manage equipment scheduling, track container movements, optimize yard space allocation, and make informed decisions based on real-time operational data. A similar framework has been extended to a multi-layer hybrid system in Yang et al. (2024). This next-generation system utilizes cyber-physical fusion, enabling holographic visual management and control patterns. The system is composed of five layers - physical and virtual resources, data connection, twin data, and application services. Together, these layers describe the port’s physical elements, infrastructure, DTs, data integration, and customized data visualization. The framework incorporates DT modelling, global ubiquitous perception, data mapping, and model fusion to facilitate intelligent operations, enhance decision-making, and enable autonomous management capabilities. This innovative system has been successfully deployed at Qingdao Port, enabling three-dimensional visual monitoring and optimal dispatching based on real-time perception data.

Another DT implementation framework is developed for Yangshan Port, Shanghai, China (Ding et al. 2023). This framework consists of five layers: a physical layer for collecting real-time data using Internet of Things (IoT) devices and sensors, a data layer for storing and integrating real-time and operational data, a model layer with simulation models to represent system behaviour, a virtual layer for visualization and real-time monitoring of the terminal layout and operations, and a decision support layer for early-warning assessments, real-time feedback, and optimization. These frameworks offer valuable guidance for implementing DTs in seaports, addressing the unique complexities and requirements of seaports operations and logistics.

Fig. 2

Digital twin applications to improve sustainability in seaports operations and logistics

4 DT for seaports’ sustainability

In Section 2, we outlined that the main outcome of the literature review and content analysis, which is a framework to categorize the reviewed works into ten core sustainability applications. In this section, we will provide a brief overview of the classification framework. The DT technology is considered a significant catalyst for digitalization in seaports, as it enables greater transparency, control, and data-driven decision-making. A digital twin revolves around data acquisition from diverse sources, including IoT sensors, cameras, and weather forecasts. Collected data relevant to promoting sustainability at seaports include environmental data (e.g., temperature, humidity, energy consumption, air and water quality, noise level), physical asset and infrastructure data (e.g., berths, cranes, vehicles, storage facilities), logistics data (e.g., movement of goods, people, energy, handling equipment), safety and security data (e.g., incidents, breaches), and performance data (e.g., accuracy, response time, reliability of the DT).

The DT facilitates implementing various tools and methods such as predictions, visualization, real-time monitoring, optimization, simulations, anomaly detection, communication channels, and automation. These tools and methods enable the development of different applications that promote sustainability in seaports. This study identifies ten core applications of the DT that promote sustainability in seaports, three core applications for the economic pillar, four core applications for the social pillar, and three core applications for the environmental pillar (refer to Fig. 2). While sustainability goals are interconnected and actions targeting one goal can affect others, this study concentrates on identifying the primary goals of each initiative and application.

Table 1 lists 68 DT tools implemented or conceptually designed for seaports. These tools are aimed at promoting sustainability and have been categorized into the ten core applications of sustainability. Additionally, this table includes real case studies to provide a practical understanding of their application in real-world scenarios. These applications utilize real-time data, ensuring that the information they provide is up-to-date and reflects the current state of the port. They are designed to be dynamic and responsive, updating their outputs in real time as new data becomes available. Additionally, they feature bi-directional data flows, allowing for seamless communication and integration with other components of a DT. Lastly, these applications serve as fundamental building blocks for a comprehensive DT of a port. They can be implemented gradually and progressively, enhancing the sustainability of ports step by step. These applications will be reviewed in Sections 5 to 7.

Table 1 Applications of DT-based systems in seaports sustainable operations and logistics

5 DT applications for operational efficiency

To maximize the economic impact of seaports, it is essential to optimize their operational efficiency and ensure their long-term viability. Efficient port operations and logistics are characterized by lower turnaround times, faster cargo handling, and streamlined logistics management. Such improvements decrease costs for shipping companies, importers, and exporters, and enhance their global competitiveness. DT-based tools have the potential to significantly improve the operational efficiency of seaports. This section intends to elucidate the diverse applications of these DT-based tools.

5.1 Operations management

DT technology provides an effective solution for improving seaport operations management. The real-time data gathered by the DT is used to develop an operational plan, which includes vessel schedules, resource allocation, and workflow optimization. A simulation model is then employed to assess the plan’s performance, taking into account uncertainties. By continuously integrating real-time data, DT ensures the simulation models remain up to date. The DT considers critical variables like berth availability, energy availability, crane utilization, labour distribution, and traffic flow, assisting in identifying resource allocation plans that optimize efficiency and minimize bottlenecks. Once an optimal plan is generated, it is transmitted to the physical system for execution.

Numerous applications of DTs have demonstrated improvements in seaport operations management. For instance, the Danish Port of Esbjerg, the world’s largest base port for offshore wind turbine transport, deployed DT to triple the port’s wind turbine transport capacity (Memija 2023). With the DT, the port can simulate future projects and operations, enabling proactive planning. By using vast amounts of data to analyze all port processes related to the shipping of offshore wind turbine installations, the DT helps the Port of Esbjerg aim to deliver 4.5 GW of offshore wind turbines annually by 2025, instead of the current 1.5 GW, reported in January 2023.

A digital berth planning optimizer has been developed in South Carolina Port’s terminals by Portchain (Port chain 2021). This digital tool enables planners to swiftly evaluate various scenarios and select the optimal plan while efficiently adapting to new customer information. By connecting customers and internal stakeholders to the platform with real-time information, the digital berth planner fosters transparency and improves collaboration. Adopting machine learning (ML) and artificial intelligence (AI), the platform enhances plan reliability, capacity, and cost-effectiveness. It effectively plans and tracks available quay, cranes, and labour resources, taking into account vessel calls, maintenance schedules, and labour constraints. Another example of berth planning is outlined in Wang et al. (2024), which describes the optimization of autonomous surface ship berthing through the use of a DT-based tool. The autonomous berthing process for these ships involves three key levels. Initially, at the motion planning level, a DT model is developed to represent the physical ship. Subsequently, a model-predictive berthing planner plans a seamless trajectory for automatic berthing based on the DT model, taking into account port and ocean disturbances, as well as ship uncertainties. Finally, at the motion control level, a line-of-sight guidance law and a parallel antidisturbance control law are implemented to enable the ship to accurately follow the planned berthing trajectory and berth automatically. In a related work, a DT was deployed in a Thai port to expedite data generation for port operations, incorporating configurable uncertainty (Wattanakul et al. 2022). The data generated from the DT was used in a simulation model to forecast the estimated time of arrival and the estimated time of departure for incoming vessels. The results demonstrated that the DT accurately replicates the behaviour of the actual system, affirming its reliability and effectiveness.

With the advancements of communication channels, e.g., 5G networks, seaports are developing a network of drones for data collection and security control. A DT model using drone-assisted data collection architecture is proposed in Yigit et al. (2023). A basic data collection protocol using IEEE 802.11p to communicate between ships and drones is also introduced. The authors also developed a recommendation engine to ensure accurate ship navigation within a port during the docking processes. Their experimental results in Port of Leith reveal that the trajectory performance is improved by approaching the desired shortest path and by reducing the financial costs and fuel consumption.

A DT-based AGV scheduling framework that considers the charging of AGVs is proposed in Gao et al. (2024). In this framework, the physical conditions (e.g., congestion, battery usage, charging rate, etc.) are monitored and synchronized with the DT model. A Q-learning optimization algorithm was embedded in the framework to minimize the transportation time and charging time of the AGVs. A similar concept of a DT for vehicle dispatching assistance in port logistics was introduced in Hofmann and Branding (2019). The DT continuously evaluates the current dispatching policy and configuration alternatives, providing a performance forecast based on the current system status. It incorporates a dispatching algorithm based on starvation avoidance to balance the probability of overstock and shortage situations at the system’s bottleneck. The study deployed a simulation using Python SimPy as a cloud service that integrates with IoT-connected sensors, minimizing infrastructure administration and seamlessly integrating into port operations. Another work in Widyotriatmo et al. (2024) describes the conception, evolution, and actualization of a high-precision autonomous container truck docking control mechanism. The DT captures dynamic data and simulates the physical truck behaviour. The docking sequence consists of two distinct stabilization issues: point stabilization to ensure a seamless transition from the initial position to the docking slot, and orientation control to accurately position the container truck at the terminal docking with a permissible deviation of 5 cm in position and 0.0087 rad in orientation.

A data-driven conflict prediction method utilizing DT is proposed in Lou et al. (2023). This method constantly compares the differences between the physical and virtual terminals to identify potential conflicts in advance. Additionally, the study explores a conflict resolution method to address the predicted conflicts. Both the prediction and resolution methods are integrated into an AGV scheduling and routing algorithm, allowing for efficient management of conflicts within the automated container terminal environment.

A decision support system for the Shanghai Yangshan Phase IV automated terminal based on the DTs is described in Ding et al. (2023), which continuously monitors operations, triggers planning processes and simulates future scenarios to adapt operational plans and processes. The system identifies issues such as truck timeouts and vessel delays, providing early warnings to operators and supporting drill-down analysis to locate bottlenecks accurately. It also highlights operational abnormalities, such as stacking plan issues, and warns operators about potential delays. An operations optimization framework in the Yangshan terminal that combines DT with ML techniques was proposed in Li et al. (2021). This framework utilizes twin data to dynamically generate models that adapt to changing environmental conditions. Real-time data from the terminal is used to validate and refine the trained model. The results demonstrate that the DT-based operation optimization significantly enhances operational efficiency by reducing crane and vehicle operation time while achieving higher loading and unloading efficiency compared to conventional terminal operations.

Another work related to Shanghai port was developed in Du et al. (2023). The authors constructed a DT model for the yard system that is mainly composed of physical space, virtual space, data, services, and intelligent agents. The model is used to enhance the efficiency of container delivery and loading processes. After receiving a delivery/ or loading request, a reservation is made and the DT model runs a simulation based on real-time data to estimate the processing time of the request. If the simulation result is satisfactory the execution order is released, otherwise a new reservation is made. After the execution, the real state of the yard system is updated in the DT model. The experimental results showed that the average waiting time for a request decreased from 27 to 19 minutes. The workflow in container terminals was enhanced in Zhang et al. (2023) through training a reinforcement learning (RL) algorithm by a DT-based simulation model which accesses real-time data of the terminal. After being trained, the RL model can be transferred to a real environment using only a small amount of real data. The experimental results show a decrease in average completion time from 200 minutes when the scheduling is done based on a first-in-first-out rule to 127 minutes using the RL algorithms. Also, the average equipment utilization rate increases from 12% to 19.4%.

A similar approach was developed in Gao et al. (2022) focuses on enhancing operations in storage yards. The study develops a framework that optimizes critical resources such as the storage area, automated stacking cranes, and AGVs by utilizing a DT. The framework continuously monitors potential disruption scenarios and visualizes real-time data in the virtual space, allowing for real-time adjustments to the initial operation plan. A case study involving the rescheduling of cranes due to the dynamic arrival of external trucks demonstrates the effectiveness of the framework and highlights the importance of considering uncertainties in port optimization. The application of the framework resulted in a significant reduction in crane waiting time compared to an offline scheduling approach.

5.2 Asset management

Asset management involves the systematic and strategic approach of acquiring, operating, maintaining, and disposing of assets to achieve optimal performance, minimize costs, and maximize value throughout their lifecycle. Effective asset management requires technical expertise, data-driven analysis, financial considerations, and proactive planning. It is particularly crucial in seaports to ensure efficient asset utilization, enhance operational reliability, comply with regulations, and ensure safety and security. DTs play a significant role in asset management by providing real-time monitoring and predictive maintenance capabilities for physical assets in seaports and maritime logistics.

One application of DTs in asset management is real-time monitoring through the creation of virtual replicas. Data analytics and ML techniques analyze asset data to predict failures or maintenance needs, enabling proactive maintenance actions, reducing downtime, and improving overall asset reliability. This allows maintenance personnel to remotely monitor equipment, perform diagnostics, and reduce the need for physical inspections. The Belfast Harbour has implemented a platform that utilizes a DT to remotely visualize and make action decisions on port equipment, improving maintenance efficiencies (Port Technology 2021). A similar port infrastructure monitoring system has been developed by GISGRO (2023) that provides an up-to-date situational view of port equipment and required maintenance actions. This platform enables maintenance teams to access and record information easily with their mobile devices and reduces manual data collection and communication.

Another direction for asset management is to forecast the health situation of infrastructures and equipment. For example, a mooring monitoring system has been implemented in the Port of Hamburg to reduce the risk of ships breaking loose during storms and improve mooring practices (Tesse et al. 2021). The DT’s virtual model captures data from multiple heterogeneous sensors and fuses the data using a Bayesian neural network. The solution forecasts the proximity of moorings to their breaking point. Another DT-based monitoring system for marine infrastructure, developed in Li and Brennan (2024), comprises four components: virtual monitoring, data-driven forecasting, fatigue reliability, and inspection planning. In the virtual monitoring, cyclic wave loads are simulated, and structural degradation such as fatigue cracks is simulated. Then, the remaining fatigue life of the structure is forecasted to build a fatigue reliability score which is used to efficiently plan physical inspection plans. Additionally, a prediction tool for the fatigue crack growth trend of marine structures, that are under the pressure of sea waves, was proposed in Fang et al. (2022). A DT model of the infrastructure is developed which is synchronized with the real-time data. A finite element surrogate model is used to synchronize the DT model of the structure and real-time data captured from the physical structure. It was validated that the prediction method can accurately predict fatigue crack growth under multiple load changes.

Maintenance decision-making models based on DT have been proposed for various assets in seaports. An integrated maintenance decision-making model for quay cranes (QCs) in container terminals considers transportation processes and crane maintenance scheduling to ensure robust performance in stochastic environments (Szpytko and Salgado Duarte 2021). The model estimates crane operation risks using a sequential Markov chain Monte Carlo simulation model and optimizes scheduling using particle swarm optimization to minimize inefficiency. The effectiveness of the proposed model has been demonstrated using data from actual container terminals. Another DT-based operation state monitoring system for port cranes has been proposed in Zhou et al. (2022). This system utilizes the DT of a rail-mounted gantry crane to create an online data simulation test environment and enables real-time data visualization during the control process of port cranes. It combines multi-sensor data acquisition and plug-in programming methods to fuse virtual and real data, facilitating control algorithm testing, movement process mapping, and remote control. The system demonstrates high real-time performance and effective visualization for crane operation monitoring.

5.3 Visibility & transparency

Enhancing visibility and transparency in seaport operations and logistics would enhance the efficiency and effectiveness. With visibility, stakeholders can track the movement of cargo, monitor the status of shipments, and identify any bottlenecks or delays. This allows for proactive decision-making and timely interventions to ensure smooth operations. Transparency, on the other hand, fosters trust and collaboration among different parties involved in the logistics process.

The Port of Singapore has implemented a groundbreaking digital bunkering transaction system, revolutionizing the efficiency and transparency of bunker trades (DBS 2021; Port technology 2021). This system utilizes a digital bunker delivery note to minimize fraud risks by enabling direct verification of trade data. By utilizing the mass flow meter system for electronic bunkering, transparency, visibility, and certainty are improved within the bunker trade supply chain. The digital bunker delivery note simplifies financing procedures and streamlines the bunkering process, ultimately optimizing fuel supply to ships and enhancing energy consumption.

Initiatives like Container 42 (Container 42, 2019) enable intelligent tracking and monitoring of individual containers. Equipped with advanced sensor technology and solar panels, Container 42, a collaboration between technology companies and the Port of Rotterdam provides real-time location tracking and monitoring, eliminating the need for physical inspections. It gathers valuable data to drive the port’s digital transformation, continuously analyzing environmental conditions both inside and outside the container during its global voyage. Another real-time container movement intelligent monitoring and analyzing system is developed in the Port of Hamburg (Kapkaeva et al. 2021). An intelligent transport system incorporates traffic data from all stakeholders in the transport chain, including schedule and cargo data for inland waterway craft, terminal/lock/bridge availability, waterways water levels, and other current traffic data from the Harbourmaster’s Office and the Nautical Centre enhancing traffic safety and reducing costs for inland waterway vessels. This system enables flexible responses to non-scheduled delays or changes, improving efficiency for inland waterway skippers and terminals. Integrating this system with the smartPORT (Port of Hamburg 2023) – see Section 7.1, the project enhances transparency and efficiency throughout the transport chain, including inland waterway shipping.

The Port of Rotterdam provides a real-time digital platform offering a comprehensive overview of crucial environmental factors within the port area (Port of Rotterdam 2023). This platform includes information on wind speed and direction, water depth and current, tides, and visibility. Logistics and transport planners can proactively consider predicted phenomena based on real-time and forecasted environmental conditions. The platform acts as a reliable source of information, enabling stakeholders to optimize their operations, particularly in green logistics, based on environmental conditions.

6 DT applications for social goals

The interconnected relationship between seaports and their surrounding social and economic environment underscores the importance of prioritizing social goals such as safety, emergency preparedness, and security in their operations and logistics (Klar et al. 2023; Batalha et al. 2023; Caliskan 2022). By demonstrating a commitment to the well-being of the community, seaports can enhance their reputation and gain a social license to operate, which fosters trust and positive relationships with stakeholders.

DTs hold tremendous potential in advancing social goals within seaports. They facilitate knowledge transfer and support worker training initiatives, enhancing the capabilities and expertise of the workforce. Furthermore, DTs promote effective communication, collaboration, and coordination among stakeholders, leading to improved transparency and trust, and they contribute to safety and security management by monitoring data from various sources and identifying potential hazards and security threats (Agnusdei et al. 2021a, b).

6.1 Situational awareness

Maintaining situational awareness is crucial in achieving social goals in seaports due to their complex and dynamic nature. Surveillance and situational awareness help in identifying and managing risks, such as security breaches, equipment malfunctions, environmental hazards, or adverse weather conditions, allowing for early detection and proactive response to mitigate their impact. Furthermore, situational awareness supports regulatory compliance by monitoring and assessing port activities and conditions, ensuring adherence to safety regulations, environmental guidelines, and international standards.

One notable example is the development of an intelligent shipping container network that leverages DTs to capture, fuse, and use data from multiple sensors and contextual information from diverse security subsystems. This network enables effective decision-making for control and safety actions, enhancing operational planning and security (Jakovlev et al. 2021). Another example framework was developed in Troupiotis-Kapeliaris et al. (2023) for real-time monitoring of the vessel traffic. A DT-based platform visualizes maritime traffic in real-time, based on data from the Automatic Identification System (AIS), and provides forecasts of future movement based on machine learning and deep learning techniques. Another vessel monitoring system was developed in Liu et al. (2021) where a DT model based on relay cooperation IoT is constructed. The proposed model provides a higher secrecy rate and can keep a data transmission delay between the DT model and the physical entities below 700ms which is essential in real-time data usage in the DT model. A vessel traffic monitoring was described in Wu et al. (2021), where a DT model is developed for an inland waterway safety monitoring system in Changhu Shen Line. The model involves combining drone tilt photography and Building information modelling (BIM) technology to construct a 3D scene model of inland waterways. The DT model enhances the safety and security of water transportation by providing real-time monitoring, surveillance, and management of maritime activities. It also helps detect and respond to potential risks, collect evidence of violations, and facilitate emergency response in case of accidents or incidents.

Autonomous drone networks are being widely used for situational awareness (see Section 5.1 for their usage in vessel navigation). In the Port of Antwerp, six autonomous drones are deployed to enhance overall security (Port of Antwerp Bruges 2023). The flights are remotely controlled from a command and control centre in the heart of the port. They provide aerial surveillance, helping to coordinate operations, manage infrastructure, detect oil spills and floating waste, and support security partners during incidents. The drones offer a unique perspective and enable effective management of a large area, enhancing safety and efficiency in the port. Another example in the Port of Antwerp involves the use of DTs to detect anomalous ship movements by utilizing ships’ automatic identification system data and fusing it with tracking and monitoring data from cameras and sensors implemented throughout the waterways (xyzt.ai 2021; Farahnakian et al. 2023). ML techniques and big data analysis are employed to identify and explain such movements, enhancing security measures.

In the Port of Barcelona, DTs combined with customs data enable accurate traffic forecasts, enhancing container terminal management and facilitating real-time decision-making for business safety. DTs provide visibility of cargo, waterways, assets, and personnel, improving situational awareness in the port (Pier Next 2020). Moreover, Belfast Harbour has installed private wireless 5G networks over 35 acres of operational port to accelerate the ambition to become a regional smart port (Belfast Harbour 2023). This private 5G network facilitates data sharing, which is necessary for situational awareness and also for automating processes across transportation.

6.2 Early warning

DTs utilize real-time data from sensors and IoT devices to provide early warnings and predictions for adverse weather conditions, equipment malfunctions, and other safety-critical events. This enables proactive measures to mitigate the risks of accidents or injuries in seaports. For example, a DT can alert operators to potential collisions and safety risks by monitoring the movement of containers and heavy equipment. Additionally, DTs can notify the appropriate authorities if a vessel enters a restricted area or an unauthorized person attempts to access the port.

DTs facilitate regular risk assessments by providing a comprehensive and dynamic representation of a seaport’s physical assets, processes, and operational conditions. This allows stakeholders to identify potential hazards, evaluate risks, and develop appropriate mitigation strategies. For instance, a safety management approach based on DT is proposed in Wang et al. (2023). Another example is the integration of quantitative risk analysis within a DT, enabling stakeholders to easily identify risk drivers and isolate accident scenarios (Wattis et al. 2021). Also, an early collision warning system in marine piling construction projects based on a DT framework is developed in Li et al. (2024). The main twin objects are piles. Based on the physical data, a pile positioning algorithm, and a collision prediction algorithm early warning system alerts the hazardous situation.

Furthermore, DTs aid seaports in hazardous cargo management by providing visualization and tracking of container parks and hazardous freight, offering simulation and training for emergency preparedness, and providing decision support through comprehensive and accurate representations of the port’s environment. An example application is a DT for 3D and real-time georeferenced visualization of container parks and the location of hazardous containerized freight in Terminal de Contentores de Leixões and Terminal de Contentores de Santa Apolónia (Oliveira et al. 2022). This tool combines different modules that allow for the visualization of container information, movement, and the surrounding area, including a realistic and dynamic 3D representation of the port’s vicinity. Another example study is the management of hazardous chemicals based on DT in Li et al. (2022). The approach involves designing a monitoring system that utilizes DT and multi-source data acquisition to enable operation visualization, information fusion, cargo tracking, and hazard source monitoring. The application of DTs is illustrated using the example of liquid natural gas road transportation, showcasing their use in the four stages of the process. This system helps visualize the progression of accidents in different stages and assists in evaluating the effectiveness of comprehensive disposal plans.

The historical sea level rise was studied in Rouja et al. (2022) to build up a digital twin which can be used to create a spatio-temporal reference model of the Royal Naval Dockyard seawall. The digital twin incorporates multiple sets of image assets, including a high-resolution handheld photo survey, a drone-based aerial image set, and historical photographs. These images were used to create a dense point cloud model and align the historical photographs with the reference model. The DT model was used to study and understand the effects of sea-level rise and other environmental factors on the seawall. Additionally, it was used to test the viability of black-zone lines as a proxy for sea-level-rise measurements in Bermuda.

6.3 Training management

DTs facilitate training and knowledge transfer through realistic and immersive simulations. Virtual training sessions in DTs replicate real-world scenarios, enabling workers to gain valuable experience and enhance their safety knowledge and skills. This proactive approach helps reduce accidents and human errors by improving situational awareness, decision-making abilities, and overall safety competency.

The Port of Livorno utilizes advanced technologies such as smart sensors, 3D LIDARs, and wide dynamic range cameras to capture comprehensive data and create a DT model of the port area and cargo (Puleri et al. 2020). This integration empowers port workers to make well-informed decisions, enhance operational efficiency, and access immersive virtual reality (VR) and augmented reality (AR) applications. The VR application allows operators to navigate the DT, accessing container information without physically being present in the storage area. This improves safety by reducing the need for on-site visits and enables remote review of container details, minimizing accidents and injuries. The AR application provides real-time visual guidance overlaid on the vision of forklift drivers, enhancing accuracy and efficiency in cargo handling and reducing errors. Another training simulator was developed in Liu et al. (2024) for a ship loader crane based on a DT 3D model in a risk-free environment. The simulator is integrated with different physical domain models, including the ship, environmental forces, hydraulic power systems, and controllers which can train the operator under various scenarios.

Integrating chatbots into DTs in seaports presents an innovative solution for enhancing worker training and knowledge transfer for both inexperienced and expert workers. By integrating chatbots with VR, AR, and real-time prediction models, knowledge transfer is facilitated, and workers receive alerts (Gärtler and Schmidt 2021). The use of voice and image processing technologies in chatbots can offer helpful advice and assistance, resulting in enhanced safety and operational efficiency in seaport settings. Chatbots offer customized empowerment, mentoring, hands-free and eyes-free user experiences, and flexible support. Their availability and accessibility make them ideal for on-the-job training and equipping workers with the necessary knowledge to remain competitive.

In the maritime sector, chatbots have been applied to vessel traffic control and safety, as well as enhancing health and safety for field workers in container terminals (Choe et al. 2017; Colabianchi et al. 2022). The Popeye chatbot (Colabianchi et al. 2022) includes a voice service, a spoken language understanding component, and an image processing app. Popeye is primarily used for training new employees involved in container safety-critical quality inspection and control operations in dock areas.

6.4 Communication management

Seaports encompass a wide array of stakeholders, including port authorities responsible for overseeing operations and infrastructure, terminal operators involved in cargo handling, shipping companies engaged in vessel transportation, freight forwarders coordinating logistics, local communities affected by port activities, government authorities ensuring regulatory compliance, and infrastructure providers offering essential services. Effective stakeholder management is paramount for the success and sustainability of seaport logistics (Fobbe and Hilletofth 2021). It entails addressing stakeholders’ interests, fostering collaboration, and achieving overall operational excellence. This collaborative approach helps mitigate conflicts, streamline decision-making processes, and promote sustainable development within seaport logistics.

DTs improve communication and stakeholder management within seaports by providing a platform for real-time data sharing. DTs enhance information exchange between various parties involved in seaport logistics. This clear and timely communication facilitates effective coordination, issue resolution, and informed decision-making. Real-time data sharing empowers stakeholders to work together more efficiently, streamline processes, and collectively address challenges. This collaborative approach not only improves operational efficiency but also enhances stakeholder satisfaction. By granting stakeholders access to relevant data and analytics, DTs empower them to contribute their expertise and insights, fostering a sense of ownership and involvement.

As an exemplar use case, the Blue Visby solution demonstrates the benefits of stakeholder engagement and decision utilization facilitated by a DT model (Blue Visby Solutions 2021). This initiative introduces a sharing mechanism that enables stakeholders, including shipowners, charterers, and cargo interests, to collectively share the costs and benefits associated with the implementation of a vessel arrival planner. The advantages encompass fuel savings, costs related to extended ocean passages, and the financial value of emissions reductions, where applicable. Participants in the initiative also become members of an association that regulates their relationships. Through the establishment of a contractual basis via the Blue Visby Solution, stakeholders effectively harness the benefits of the platform while promoting transparency, fairness, and cooperation among all stakeholders involved.

Another communication management system for a network of IoTs in a DT is developed in Qian et al. (2023) where the network communications are subject to eavesdropping attack. The digital twin model is constructed based on the Federated learning approach where the local IoT models are trained individually and transmitted to a high-altitude platform to be aggregated and to build a global model. The objective of the communication management system is to minimize total energy consumption for constructing the digital twin of marine IoTs by jointly optimizing the global accuracy, local accuracy, transmission power, and transmission duration.

7 DT applications for environmental goals

Ensuring the environmental sustainability of seaports has become a pressing concern, driven by growing awareness of environmental degradation and increased regulations. Seaports and maritime logistics are major contributors to global carbon emissions, accounting for roughly 3% of total emissions (Alzahrani et al. 2021). Seaports face significant challenges, including pollution, rising sea levels, and extreme weather events that can disrupt their operations and threaten their existence (Homayouni and Fontes 2018; Hossain et al. 2021; Fontes and Homayouni 2023). The activities within seaports, such as fuel-powered cargo handling equipment, ships, trucks, trains, and power plants supporting port operations, contribute to air and water pollution, impacting the health and well-being of nearby communities (Acciaro et al. 2014; Hossain et al. 2021). To address these challenges, it is imperative to tackle the environmental impact of port activities alongside its economic viability.

Seaports are embracing initiatives like electrification, energy-saving practices, and smart energy management systems, and exploring opportunities for local energy generation using renewable energy resources, like solar panels, wind turbines, tidal power, and energy storage solutions. Digitalization has become a powerful tool in this context (Gerlitz and Meyer 2021; Yu et al. 2022; Bortolini et al. 2022; Teng et al. 2021; Ghenai et al. 2022). Specifically, DTs are capable of integrating data from various sources to provide a comprehensive understanding of ecological footprints, air and water quality, energy consumption, and waste management in port environments.

7.1 Green logistics

In logistics and transportation, the IoT and sensors play a pivotal role in cargo and vehicle tracking, as well as monitoring environmental conditions (Sahal et al. 2021; Wu et al. 2022). When integrated with DTs, these data sources provide a comprehensive real-time view of equipment and systems’ performance. By exploiting the combined power of IoT, sensors, and DTs, stakeholders can proactively address potential challenges and implement measures to enhance sustainability throughout the logistics and transportation processes.

In recent years, several digital tools have emerged to promote green external logistics practices in seaports. One prominent example is Routescanner (Route Scanner 2023), an application developed in close coordination with the Port of Rotterdam. Routescanner provides cargo owners worldwide with comprehensive insights into door-to-door connections for containers across seaports, inland ports, rail, and road networks. With integration across more than 4500 terminals, 1200 ports and inland hubs, and collaborations with over 200 operators, Routescanner transforms ports into transparent components of digital transport chains. This enables efficient intermodal route planning from origin to destination, effectively reducing pollutant emissions, operational costs, and arrival times, contributing to enhanced sustainability in logistics operations. Another noteworthy application is the real-time ship routing DT, addressing decarbonization regulations (Wei et al. 2023). This application utilizes a data model and a physical model to predict carbon emissions, enabling real-time prognosis and evaluation of compliance with decarbonization regulations. Incorporating meteorological and AIS data into the physical model allows for real-time prediction and updating of resistance, power, and carbon emissions, while uncertainty analysis quantifies probabilistic regulatory compliance performance.

A collaborative scheduling method between ports and vessels based on a DT model was established in Eom et al. (2023). The DT model enhances vessel arrival time prediction, optimizes the scheduling and coordination between the port and vessels, and through reducing vessel waiting time near ports reduces carbon emissions. While predicting the vessel arrival and service time, the DT collects real-time data from several sources such as weather conditions, port congestion, and operational status, and provides more accurate estimates. A case study conducted in Pusan Newport International Terminal showed that the DT-based scheduling method saved over 75% of carbon emissions compared to the actual operation case.

Blue Visby has also made strides in enhancing collaboration between ports and vessels with the development of a vessel arrival optimizer (Blue Visby Solutions 2021). This application optimizes the arrival of a group of vessels sailing to the same destination, considering vessel specifications and performance, port congestion, and environmental conditions. By employing voyage speed optimization techniques, the platform aims to achieve a significant reduction in emissions, estimated at around 15% (Blue Visby Solutions 2021).

The SmartPort project (Port of Hamburg 2023) in the Port of Hamburg integrates real-time data to improve traffic flow, offering personalized navigation, traffic situation updates, parking and infrastructure information, and other crucial operations information for efficient traffic management. Anyone driving around the port benefits from personalised navigation. In a relevant study, the optimization of truck appointment systems is investigated to mitigate waiting times and congestion for both external trucks and internal trucks within the container terminal (Zhang et al. 2019). The research adopts a vacation queuing model to portray the coordinated service process of yard cranes and employs non-stationary queuing theory to achieve more accurate estimations of truck waiting times. Through the integration of vacation queuing theory, the study effectively describes the complex coordinated service process of yard cranes, while non-stationary queue methods are employed to estimate waiting times with greater precision.

For green internal logistics in seaports, DT and ML techniques are utilized to optimize the path planning of AGVs within a container terminal (Gao et al. 2022). Real-time data from the physical terminal, including container block information, cranes, and dynamic changes during operation, is leveraged to formulate a mathematical model for minimizing AGV transportation time. ML techniques, specifically Q-learning, are employed to optimize transportation paths, considering task destinations, conflicts, and congestion. The solution is validated and optimized using a virtual terminal, ensuring efficient container transportation under current and future tasks. In a related work, the optimization of energy consumption for automatic stacking cranes in container terminals is explored (Gao et al. 2023). The solution approach involves determining the sequence of tasks, which are then simulated in a virtual container yard integrated into the terminal’s DT. The synchronization between the virtual representation and the physical container yard enables observation and validation. Notably, the simulation model incorporates considerations of the stochastic availability and arrival time of AGVs, which further enhances the accuracy and realism of the optimization process. This integrated approach of utilizing the virtual container yard within the DT contributes to a more accurate and practical solution for optimizing energy consumption in container terminal operations.

7.2 Energy management

Effective energy management systems are vital for protecting the environment in seaports. As handling tasks in ports and transportation routes are energy-intensive and new advances in local power generation emerge, energy management systems have become increasingly important. Digital twins can improve these management systems by creating opportunities for energy consumption and production pattern modelling and prediction.

DTs use historical and real-time data to develop predictive models that forecast energy demand and consumption patterns. By continuously collecting and analyzing data from sensors and other sources, DTs enable proactive energy management, identifying opportunities for energy conservation, load balancing, and demand response. The Port of Livorno has implemented a DT initiative as part of its decarbonization strategy. The DT utilizes real-time data for precise tracking, positioning, and inventory management (Cardone 2020). It also integrates with the port authority control platform and employs advanced vision techniques for accurate positioning estimations. The pilot project at the Port of Livorno is expected to yield significant benefits These include a reduction in registration time from three minutes to two minutes, shorter average movement times from eight minutes to seven minutes, a 10% improvement in storage utilization, and a decrease in the average unloading/loading time of ships from 18 hours to 16 hours. These optimizations result in a notable reduction in fuel consumption and associated carbon emissions by 8.2%.

Abu Dhabi Terminals achieves complete traceability of container transporters from the quay to the yard. Implementing Microsoft® Azure services, valuable data is collected to digitally map a crane’s box moves per hour, enabling the reduction of unnecessary movements and minimizing energy consumption associated with handling equipment. By utilizing Microsoft® Azure Stack Edge (Microsoft 2021) and computer vision capabilities, efficiency and productivity are enhanced. This process involves running computer vision models for container placement tracking and employing ML techniques for model training, maintenance, and establishing a historical data repository for future analytics.

Seaports can manage their energy demand by implementing smart grid technology, which allows for real-time monitoring and management of energy usage. This responsiveness allows operators to quickly adapt to fluctuations in energy demand and optimize energy consumption accordingly. Efficient energy management in ports can be achieved through decentralized control systems with localized computations. These systems collect data to predict energy requirements, supply, and local production within a limited time horizon, effectively managing demand and supply. An example is the multi-agent control system proposed in Kanellos (2017), which focuses on real-time control of flexible loads in ports. The system addresses the challenge of rapid power fluctuations from offshore wind utilization, and simulations conducted in a large-scale port demonstrate its accuracy in tracking power demand set-points and minimizing errors. Furthermore, the energy consumption of an office building located in the Port of Middlesbrough is modelled in Segovia et al. (2022). The study incorporated data from various sensors measuring electricity and fuel oil consumption, temperature and humidity distribution, and subjective thermal experiences reported by occupants. The energy consumption model developed in this study facilitates the exploration of different scenarios related to demand management, demand response, and emissions reduction. The findings highlight promising outcomes, including a potential 33% reduction in annual CO2 emissions by transitioning from fuel oil heating to heat pumps, and a 14% decrease in peak electricity demand through the implementation of demand response controls.

An important development in the energy landscape of seaport logistics is their transformation into prosumers, where they actively consume and produce energy concurrently. This shift has captured the attention of local communities, leading to increased interest in energy markets that promote accessibility and sharing of services within the community (Acciaro et al. 2014; Pieńkowski 2021). DTs play a crucial role in supporting the decarbonization process in this context. They offer valuable capabilities, such as smoothing cleaner energy supply and facilitating the seamless integration of RESs into seaport energy systems.

In the context of hydrogen production, DTs collect real-time data on critical parameters like temperature, pressure, flow rates, and energy consumption (Gerard et al. 2022). They control the production process, such as electrolysis or steam methane reforming, and optimize safety measures within hydrogen production facilities. Furthermore, DTs facilitate the management of the hydrogen supply chain in seaports, overseeing storage, transportation, and monitoring considerations like pressure, temperature, and leak detection. By integrating data from sensors and meters, DTs ensure the reliability and integrity of the hydrogen supply chain. This information empowers operators to optimize hydrogen utilization, minimize wastage, and ensure safe operations.

To enhance the integration of Renewable Energy Sources into a seaport’s energy system, DTs can be employed. For example, a DT can use real-time data from sensors and the IoT to monitor the performance of RESs. By integrating this data with real-time meteorological information and energy consumption forecasts, the operations of RESs can be fine-tuned. This can involve adjusting the angle of solar panels to maximize sunlight exposure and altering the pitch of wind turbine blades to capture more wind energy. A DT for integrated port energy systems is proposed in Pang et al. (2023). The architecture encompasses a digital space that captures the integration of multiple energy flows, electrification initiatives, and the coupling of physical and informational aspects in integrated port energy systems. The authors also highlight the key challenges and technological advancements necessary for the successful implementation of a DT for integrated port energy systems.

The Anzio Port in the zero energy district in Italy initiated a project to develop energy-saving procedures, integrate RESs, and promote sustainable mobility (Agostinelli et al. 2022). The implementation of DTs conducts an energy analysis, demonstrating the potential for energy self-sufficiency within the infrastructure. The digital and ecological transformation can be extended to the surrounding territory, including sea, land, road, and rail transportation. Additionally, it aims to replace carbon-based electricity production with renewable energy sources, thereby reducing the economic and environmental costs of public lighting and electricity supply for moored boats.

7.3 Environmental protection

The transportation of chilled and frozen goods in refrigerated containers (reefers) across intermodal transport chains involves both seaports and land terminals. During their stay at terminals, these containers are connected to the electricity grid to maintain their temperature. In addition to regular storage costs, seaports charge fees for providing refrigerated container services, including the power supply during storage. Studies indicate that reefer storage can account for up to 40% of a container terminal’s total energy consumption (Filina-Dawidowicz and Filin 2019; Wilmsmeier et al. 2014).

The concept of intelligent containers has been trialled in the banana supply chain (Jedermann and Lang 2021). These containers are equipped with environmental sensors to monitor various parameters that affect the quality of fresh products. The primary objective is to track the quality conditions of each product and predict its remaining shelf life. Using these predictions, items with a limited remaining shelf life can be prioritized for delivery to nearby retailers, while those with a longer shelf life can be scheduled for later deliveries or more distant customers. Further advancements in this area have been explored in Jedermann et al. (2022), where simulation models for predicting the remaining shelf life were integrated with real-time data shared within a DT. The DT includes adapters to handle different standard protocols, and additional product-specific sensors can be incorporated for specialized measurement tasks. Another application in the cold supply chain of fresh horticultural products has been investigated in Defraeye et al. (2019, 2021). DTs prove particularly beneficial for perishable species and low airflow rates. They offer actionable data to exporters, retailers, and consumers, such as the remaining shelf life for each shipment, which can inform logistics decisions and marketing strategies.

The DT models are also widely used to create dashboards and monitoring systems to control emissions within the seaport area. For example, in the port of Oulu, Finland, the implementation of DT has significantly improved the sharing of real-time environmental monitoring data. A customized DT-driven platform called “AURA" has been developed to integrate dynamic data from IoT equipment, sensors, and digital representations with environmental data (Port of Oulu 2020). This platform collects data from various sources within the port, allowing for the creation of DT-based models. These models enable the simulation of the port’s regional, river basin, and spatial environment, facilitating the prediction of pollutant transmission, dilution, and diffusion in different environmental media. With the use of DT-driven management, all stakeholders, including vessels, trains, trucks, construction projects, area maintenance, waste management, and customs have simultaneous and direct access to up-to-date comprehensive information presented in an intuitive spatial visual context.

Another method to evaluate and predict carbon intensity in actual operational and environmental conditions was proposed in Vasilikis et al. (2023). The DT model predicts the fuel consumption and carbon intensity of mechanical, electrical, and hybrid propulsion systems under the aggregate effect of operational and environmental uncertainties. The method combines data from the DT model and ML techniques which allows to prediction of instantaneous fuel consumption with an accuracy of less than 5% error and carbon intensity over voyage intervals within 2.5% at a confidence interval of 95% for the case study on an Ocean-going Patrol Vessel. A similar approach was applied in Yang et al. (2022) to collect operation and carbon emission data from automated rail-mounted container gantry cranes using IoT sensors. The data was used to create a DT model which was employed for simulation studies used in cranes’ hydrogen energy transition. This model can fully show the potential benefits and losses of all possible investment schemes, estimate the present value of low-carbon transition investment schemes for different equipment, and provide references for port decision-making. The DT model also is used as a remote dashboard to visualize the cranes’ operation and carbon emission behaviour to operators.

8 DT implementation challenges

The implementation of DTs in seaports is accompanied by various challenges and barriers, which have been partially addressed in the existing literature. Technical barriers, data availability, and interoperability have been recognized as significant challenges (Wang et al. 2021; Klar et al. 2023). However, it is crucial to further explore these barriers, particularly those relevant to seaport sustainability. One can adapt and extend an understanding of the challenges for implementing DTs in seaports by drawing insights from digital initiatives in ports, logistics, and industry such as those discussed in the literature (Senna et al. 2022; Wankhede and Vinodh 2021; Attaran and Celik 2023; Deepu and Ravi 2021; Sarkar and Shankar 2021).

One notable challenge is the high investment and ongoing costs of implementing DTs in seaports. DTs require regular maintenance and updates to ensure the accuracy and currency of the information they provide (Jia and Cui 2021). Additionally, seaports need a skilled workforce capable of operating, maintaining, and developing the DT effectively.

Another critical challenge lies in the involvement of multiple stakeholders with potentially conflicting or competing interests (Klar et al. 2023). This challenge becomes particularly relevant when seaports aim to achieve sustainability goals through DT implementation. The objectives of different stakeholders may diverge. For instance, the seaport authorities are seeking to maximize total throughput while minimizing operational costs, while a renewable energy supplier may impose technical constraints that pose significant challenges or lead to increased pollutant emissions. Managing these conflicting interests and goals is essential, especially considering that the port authority usually plays a central role in regulating relationships among stakeholders. Throughout the implementation of DTs, it is essential to understand and address the challenges and barriers summarized in Fig. 3. All stakeholders involved in the seaport ecosystem should actively engage in addressing these challenges and work collaboratively to overcome them.

Fig. 3

Main challenges and barriers for DT implementation in seaports

The lack of stakeholder collaboration presents significant barriers to the implementation of DT projects in seaports, particularly in promoting sustainability (Attaran and Celik 2023). These barriers arise from a lack of understanding and shared vision among stakeholders regarding the positive impact of DTs on seaport operations, environmental impact, and long-term sustainability. Several key challenges contribute to these barriers. First, there is a lack of awareness among stakeholders about the benefits and potential of DTs, which can lead to resistance or hesitation towards their implementation. Concerns related to data security, privacy, system reliability, and vulnerabilities further contribute to the perception of risk associated with DT implementation. Secondly, some stakeholders resist adopting DT solutions due to concerns about disruption to existing processes, fear of job displacement, or a lack of familiarity with new technologies. Thirdly, differing priorities, goals, and interests among stakeholders make it challenging to achieve consensus on the implementation of sustainable DT practices. Conflicting agendas can impede progress and decision-making. Fourthly, limited collaboration and cooperation among stakeholders hinder the collective effort towards implementing sustainable DT solutions, as it requires alignment and coordination among multiple entities. Finally, financial considerations pose a barrier, as stakeholders may be deterred by the costs associated with implementing DTs and the uncertainty of immediate or quantifiable returns on investment, especially in terms of sustainability benefits.

Legal barriers are significant factors that can greatly impact the implementation of DT projects in seaports, especially in the context of promoting sustainability (Glover 2020; Nezhad et al. 2024). These barriers encompass a range of challenges that need to be addressed. First and foremost, ensuring regulatory compliance with existing maritime, environmental, and safety regulations poses a complex task as it involves integrating DT systems into established regulatory frameworks. However, a potential obstacle arises when these regulations do not adequately support the implementation of sustainable DT solutions, which can hinder adoption efforts. Intellectual property rights also present challenges as seaports must navigate complex legal frameworks to ensure compliance with applicable laws governing the collection, storage, and sharing of data within the DT framework. Resolving intellectual property rights issues, including negotiating licensing agreements for proprietary technologies and software, can be a time-consuming and intricate process. Moreover, clarifying data ownership, and access rights, and establishing liability frameworks for errors or damages arising from the use of DT-generated insights are essential. Given that DT implementation often involves multiple entities and stakeholders, establishing clear contractual agreements, data-sharing agreements, and governance frameworks becomes crucial to mitigate legal risks and protect the rights and interests of all involved parties. Additionally, legal disputes with external entities such as neighbouring businesses or environmental organizations can pose obstacles to DT implementation, necessitating dedicated legal resources for their resolution.

The successful implementation of DTs in seaports is contingent upon overcoming various technical challenges. Adequate infrastructure, encompassing both hardware and software components, plays a pivotal role in supporting DT implementation in seaports. This infrastructure includes robust computing systems capable of efficiently processing and analyzing large amounts of data, as well as reliable communication networks that enable real-time data exchange, facilitated by sensors and IoT devices. Furthermore, appropriate data storage and management systems are essential to handle the vast and diverse datasets generated by seaport operations. Constructing a comprehensive DT model for seaports’ physical assets and processes presents further challenges, particularly in the integration of geometric and process information models (Wang et al. 2021). Existing geometric modelling tools are not specifically designed to accommodate the requirements of DTs, necessitating adaptations and customized approaches. Moreover, the design of process information models, which rely on the geometric model as a carrier of data, has yet to fully embrace the advancements offered by DT-based modelling. Consequently, limitations may arise in capturing and representing the dynamic nature of seaport processes within the DT framework. Scalability and flexibility are paramount considerations in DT implementation within seaports (Nezhad et al. 2024). Seaport operations are inherently dynamic and subject to continual change. Therefore, the DT must exhibit the ability to scale and adapt to evolving operational requirements, accommodate the integration of new technologies, and align with shifting business needs. This entails the capacity to handle increasing data volumes generated by a multitude of sources and the ability to seamlessly incorporate future advancements in seaport operations.

The successful creation of an accurate DT hinges on the availability and connectivity of a diverse range of data, encompassing crucial aspects such as the seaport’s physical infrastructure, weather conditions, and shipping traffic. However, gathering this data can be challenging. To achieve reliable and effective DT-based models, it is vital to ensure consistency in data collection across dimensions, including physical, geometric, and temporal data. This consistency is necessary to avoid fusion failures and ensure accurate representations within the DT framework. The adoption of 5G connectivity technology is considered the optimal and cost-efficient approach to meet the connectivity requirements of advanced automation applications in port (Du et al. 2022). Ensuring data quality, consistency, and reliability is crucial for accurate modelling and simulation. Additionally, seaports generate substantial volumes of data from various sources such as sensors, equipment, and systems. Integrating this diverse data into a unified DT model poses its own set of challenges. The lack of unified communication protocols and data standards further complicates the integration of DT-based models from different dimensions. This challenge hampers seamless interoperability and standardization across different systems within the seaport environment (Lind et al. 2020). To address this issue, efforts are underway to develop common standards and protocols that facilitate effective data exchange and integration, promoting a cohesive DT ecosystem (Fonseca and Gaspar 2021). Integrating a digital twin with existing systems and processes can also be demanding, particularly when the seaport’s current systems are outdated or incompatible with DTs. Overcoming compatibility issues and addressing the need for system upgrades or replacements is essential to enable seamless integration and maximize the benefits derived from DT implementation in seaports.

Ensuring data security and privacy emerges as a notable challenge in the implementation of DT in seaports. The extensive operations in seaports require stable data transmission to mitigate the risks of data loss during real-time processes, which could otherwise disrupt operations. To mitigate these risks, reliable communication networks must be established to ensure uninterrupted data flow and minimize the impact of data transmission instabilities. Additionally, legal and ethical issues surrounding data ownership, privacy, and security arise in the context of DT implementation. Compliance with data protection regulations and addressing privacy concerns are paramount. Robust security measures, including data encryption and access controls, must be implemented to safeguard data and ensure compliance with relevant regulations. Building trust among stakeholders requires demonstrating a commitment to data security and privacy. Furthermore, DTs are susceptible to cyber attacks that can compromise the integrity of the system and the data it contains (Glover 2020; Fonseca and Gaspar 2021). Implementing robust cybersecurity measures becomes crucial to protect against potential threats and ensure the resilience of the DT. Complying with cybersecurity regulations and addressing liability and responsibility for data breaches are also critical considerations.

9 Discussion on potential applications

The implementation of digital twins within seaports is still in its infancy, with stakeholders holding differing opinions on how to define the borders and boundaries of a DT within such a complex environment. Nevertheless, there exist many untapped opportunities where DTs could make a significant contribution towards enhancing seaport performance, especially in advancing their sustainability. Future research should concentrate on developing new DT applications in seaports, as the potential benefits are significant, but further research is needed to fully realize them. In this section, five potential applications of DT in seaports located in underrepresented areas are discussed.

9.1 Stakeholder management

Enhancing stakeholder management through the use of DTs are relatively uncommon practice in the seaports. To illustrate its potential, a specific potential application is presented in Table 2. In this scenario, the DT serves as a virtual representation of the entire port, encompassing all relevant assets and infrastructure. Real-time updates ensure that stakeholders have access to the latest information at all times. The DT incorporates various features that facilitate communication and collaboration among stakeholders, including chatbots and data-sharing capabilities. The effectiveness of this application can be evaluated by monitoring and analyzing the number of accidents and delays at the port before and after implementing the DT-based stakeholder management system.

Table 2 A DT potential application for enhancing coordination and collaboration in seaports

9.2 Green internal logistics

DTs in internal logistics might aid in identifying bottlenecks and implementing measures to prevent congestion and collisions within the terminal environment, specifically during peak hours. An illustrative potential application, as shown in Table 3, exemplifies the integration of a speed adjustment strategy for internal AGVs (Fontes and Homayouni 2023; Homayouni and Fontes 2021) with a DT, facilitating real-time decision-making and effective AGV operations management. The DT, continuously updated in real-time, aggregates data from various sources and facilitates the joint scheduling of Quay Cranes operations and AGVs handling tasks. Through this integration, several benefits are expected: optimized AGV routing, reduced container transportation time, coordinated QCs and AGVs activities to prevent bottlenecks and delays, and adjustable AGV speed based on traffic conditions, further reducing ship unloading time. The application’s evaluation metrics encompass ship unloading time, terminal throughput, energy consumption, and the cost of handling equipment operations, generating valuable insights for improving logistics efficiency and sustainability in seaports.

Table 3 A DT potential application for energy-efficient QCs and AGVs scheduling

9.3 Alternative energy supply

Diversifying energy vectors and adopting cleaner fuels are essential initiatives for decarbonization in seaports. These include the electrification of port equipment, such as transitioning from diesel-powered machinery to electric alternatives, the implementation of onshore power supply (OPS) infrastructure to enable ships to connect to the electrical grid while berthed, and the production of sustainable fuels like biofuels, hydrogen, and ammonia. DTs offer valuable support for these initiatives in seaports. They aid in the planning and design phase by allowing the simulation and optimization of shore power connections. Once deployed, DTs enable real-time monitoring and control, providing insights into power demand, energy consumption, and system performance. This facilitates power usage monitoring, issue identification, and supply optimization for docked vessels.

Table 4 A DT potential application for integration of renewable energy source via a virtual power plant in seaports

Furthermore, DTs can play a crucial role in the context of virtual power plants (VPPs) (Acciaro et al. 2014). A VPP comprises a network of decentralized power generation units, including RESs, which are coordinated and controlled to provide reliable and flexible power to the grid. By deploying a DT, real-time data from sensors can be utilized to monitor the performance of individual power generation units within the VPP. This data can then be used to optimize the VPP’s operation, such as adjusting the output of specific power generation units to maintain a balance between supply and demand.

Table 5 A DT potential application for peak power shaving through optimizing the scheduling of energy-intensive tasks

An illustrative DT potential application for the integration of RESs through a VPP is described in Table 4. A seaport aims to integrate RESs into its energy system, comprising solar panels and wind turbines with intermittent and unpredictable outputs. Additionally, the installation of a virtual power plant is under consideration to optimize energy demand and RESs integration. By employing the DTs, the seaport can simulate various scenarios, effectively determining the optimal approach to integrate RESs and the VPP, thus enhancing energy decision-making. The application’s evaluation metrics encompass tracking electricity expenditure, revenue generated from surplus electricity sales to the grid, the percentage of electricity consumption met by RESs, and CO2 emissions per operation.

9.4 Peak power management

Increasing the share of RESs in a seaport’s energy mix can result in less reliable power availability due to the intermittent nature of power generation from these resources. On the other hand, the diverse range of activities taking place in seaports results in fluctuating instant power requirements, posing challenges in meeting these demands. Exceeding the available power supply in the seaport can lead to blackouts or brownouts, resulting in equipment damage and production disruptions. Therefore, there is a clear need for peak power shaving measures. Furthermore, in seaports, the electricity bill consists of a fixed cost and a variable cost based on consumption levels (Geerlings et al. 2018; Iris and Lam 2019). The fixed charge is determined by the highest observed peak demand, which sets the cost for the subsequent 12 months. When a new higher peak is reached, the costs are updated accordingly. Peak demand in seaports accounts for approximately 25-30% of the monthly energy bills. Reducing peak demand can result in significant annual savings for container terminals.

An illustrative DT potential application that focuses on optimizing the scheduling of energy-intensive activities like QCs and yard cranes while imposing a peak power limit is presented in Table 5. The DT model integrates real-time data from sensors, control systems, and operational databases, simulating the seaport’s physical assets and operations. The application can be evaluated by tracking the percentage of time power usage remains below the peak limit, per-move energy consumption, and the sufficiency of the seaport’s power supply.

Table 6 A DT potential application for waste management in seaports

9.5 Waste management

DTs offer valuable capabilities for waste monitoring and tracking, providing real-time information on waste generation and disposal within seaports (Attaran and Celik 2023). This technology plays a pivotal role in facilitating collaboration among stakeholders involved in waste management, including port authorities, waste management companies, recycling facilities, and regulatory agencies. Through DT implementation, stakeholders can effectively collaborate on waste management strategies, share and analyze real-time data, predict waste generation patterns, optimize waste collection routes, and allocate resources efficiently (Borchard et al. 2022; European Environment Agency 2021). In hazardous waste management, DTs ensure compliance, enhance safety measures, and simulate scenarios for optimizing storage, handling, and disposal procedures.

A promising potential application of DTs in seaport waste management is the development of a model for the port’s waste collection system. This model would simulate the entire waste flow within the port, encompassing waste generation points, collection routes, waste processing facilities, and disposal sites. An illustrative DT application is described in Table 6. By utilizing historical data and real-time information, the DT could provide accurate predictions of waste generation and enable the optimization of waste collection routes and schedules.

10 Conclusion

This study explored the potential of digital twin technology in improving sustainability in seaport operations and logistics. It presents a framework to help managers grasp the role of digital twins, promote their adoption, and tackle the challenges of implementing sustainable seaport development. In addition, it outlines the path for future research and practical application, positioning managers as innovators in seaport sustainability. It also defines clear boundaries for implementing DT technology in this complex system.

Through a systematic literature review, we identified and categorized 68 digital applications into ten core areas that support economic, social, and environmental objectives in seaports. Our findings emphasize the significant potential of DTs to improve seaport sustainability through real-time monitoring and decision-making, enhanced safety and security, optimized resource utilization, improved collaboration and communication, and the development of the seaport ecosystem. In addition, we investigated potential applications of DTs in areas often overlooked, including stakeholder management, green internal logistics, alternative energy supply, peak power management, and waste management.

Integrating DTs into seaports presents a unique set of obstacles, including high implementation costs, a shortage of technical expertise, and conflicting stakeholder priorities. Data availability and willingness to share data on a common platform further complicate DT integration. Additionally, the lack of model validation poses a significant risk, as many stakeholders may view the DT as a “black box" with hidden complexities.

Despite the comprehensive nature of this review, several limitations exist. The primary limitation is the reliance on existing literature, which may not fully capture the latest advancements and practical implementations of DTs in seaports. Additionally, the study’s scope was confined to publications available in specific databases, potentially excluding relevant works from other sources. Finally, the dynamic and evolving nature of DT technology means that some findings may quickly become outdated as new developments emerge.

Digital twin technology is still in its early stages of implementation within seaport operations and logistics. There are numerous untapped opportunities to utilize DTs to improve seaport efficiency and sustainability. Future research should concentrate on creating innovative DT applications tailored to the unique dynamics of seaports, along with case studies and pilot projects demonstrating their real-world impact. It is important to investigate the cost-benefit aspects of DT implementation in seaports to gain a clearer understanding of the financial implications and potential returns on investment. Staying updated on technological advancements in DTs and their applications in seaports is vital for continuously updating and refining best practices. Additionally, developing standardized metrics and frameworks to measure the sustainability impact of DTs in seaport operations is essential.