Digital Twin Used in Real-Time Monitoring of Operations Performed on CNC Technological Equipment

Featured Application

Digital Twin used in real-time monitoring of operations performed on CNC technological equipment that will retrieve the relevant parameters of machine working.

Abstract

This article presents the development and implementation of a real-time monitoring solution designed for CNC machines, specifically applied to 150 industrial printing machines, leveraging Digital Twin (DT) technology. The system integrates an SQL database with Android and .NET interfaces, ensuring seamless data synchronization across all machines and optimizing production processes. The real-time monitoring enables immediate reflection of operational changes, enhancing predictive maintenance and reducing machine downtime. A notable feature of the system is its 1 s average data synchronization rate per machine, managing 150 resources distributed over a 10,000 mp area. This fast synchronization improves workflow coordination, reducing production time by approximately 10%, and minimizing operator delays caused by material issues, machine malfunctions, or product defects. The integration of advanced analytics further supports real-time decision-making, predictive maintenance, and performance optimization, aligning the solution with the objectives of Industry 4.0 and Industry 5.0 initiatives. This version reflects the specific results of the research, including the 1 s synchronization rate, the 10% reduction in production time, and the scalability of the system for 150 resources.

1. Introduction

In today’s industrial landscape, characterized by rapid technological advancements and rising market demands, manufacturing organizations face increasing pressure to boost productivity, optimize processes, and stay competitive. However, a significant challenge remains: leaders struggle to effectively monitor production and machine parameters without real-time monitoring systems that incorporate Digital Twin technology [1].

1.1. Background

Real-time monitoring through Digital Twin systems is crucial for tackling operational inefficiencies, enabling predictive maintenance, and dynamically optimizing manufacturing processes [2]. This reinforces the necessity of advanced monitoring solutions in modern production environments [3].

Our study explores how Digital Twin (DT) technology can be applied to CNC machines, with specific emphasis on printing industrial machines, to create a virtual reproduction of physical production processes. This enables real-time monitoring, data synchronization, and informed decision-making. These applications significantly enhance operational efficiency [4], reduce downtime, and improve overall equipment performance by providing continuous insights into the status and performance of machines and processes. By establishing a virtual representation of equipment and production workflows, the system allows real-time tracking of key parameters, analysis of data, and the identification of potential issues or inefficiencies [2] before they affect production.

The primary objective of developing production monitoring applications using Digital Twin concepts is to improve the efficiency, productivity, and responsiveness of manufacturing environments. Integrating SQL databases and .NET interfaces, these systems allow for seamless synchronization of data, averaging 1 s for each of 150 resources running on a Microsoft Windows Server 2022. This ensures that production managers can make data-driven decisions and reduce inefficiencies by optimizing machine utilization and maintenance processes. Furthermore, this approach supports preventive maintenance strategies that minimize equipment downtime [5], ensuring smoother operations and significant cost reductions [6].

1.2. Research Topic, Goals, and Objectives

This research introduces an innovative approach for monitoring the production parameters of CNC machines by utilizing real-time monitoring software version 1.0 powered by Digital Twin technology. This software solution is designed to facilitate better production control and oversight. It employs Structured Query Language (SQL) [7] to handle information transmission, ensuring that data move efficiently between the systems. Additionally, the interface is developed using .NET [8], providing a robust and user-friendly platform for operators and managers to interact with.

One of the primary benefits of this system is its capability to continuously synchronize data [9] across different systems, facilitating the real-time monitoring of critical performance indicators.

Ultimately, the system’s capability to provide real-time insights, predictive maintenance, and continuous synchronization enables manufacturing companies to optimize their operations, maintain high levels of productivity, and fully embrace the transformative potential of Industry 4.0 technologies [10]. The implementation of such a solution positions organizations to operate at the cutting edge of industrial innovation.

1.3. Novel Contribution

As a novel contribution, this research focuses on the integration of Digital Twin technology into CNC machines, particularly when applied to printing industrial machines, to develop real-time monitoring and performance optimization solutions. Unlike existing approaches, this study emphasizes the synchronization of data in real-time, averaging 1 s for each 1 of 150 resources across a large production environment, which enhances operational efficiency and reduces downtime [11].

This research stands out by not only exploring the theoretical foundations of Digital Twin technology but also by demonstrating its practical applications in real-world manufacturing. By leveraging real-time data synchronization and performance tracking, the study provides manufacturers with actionable insights to optimize machine maintenance, reduce cycle times [12], and minimize operational costs.

Additionally, this research addresses a gap [13] in the literature by evaluating the impact of Digital Twin technology in specific industrial settings, such as CNC and printing machines. By analyzing its influence on operational performance [12] and competitiveness, this study contributes valuable knowledge to the field and establishes best practices for adopting Digital Twin technology in manufacturing processes.

Furthermore, this research tackles the unique challenges of integrating Digital Twin technology into established manufacturing systems, offering solutions to improve adaptability and responsiveness in today’s rapidly evolving industrial landscape [14].

2. Related Work

2.1. Digital Twin Market

The global CNC machine market has seen rapid growth, with a projected CAGR above 6%, driven by increasing demand for precision machining across industries like automotive, aerospace, electronics, and medical [15]. Similarly, the Digital Twin technology market in manufacturing is expected to grow at an annual rate of about 25%, fueled by demand for solutions to monitor and optimize production [16].

Traditional monitoring systems often lack real-time insights into key metrics like machine utilization, tool wear, and quality, leading to inefficiencies and downtime. To address this, the proposed project introduces real-time monitoring software version 1.0 for CNC machines using Digital Twin technology. By creating virtual replicas and synchronizing data between the digital and physical systems [17], manufacturers can identify bottlenecks, predict maintenance needs, and optimize production [13].

Real-time monitoring of KPIs and predictive analytics using machine learning can drive continuous improvement and proactive decision-making. North America, led by the USA as can be seen in Figure 1, dominates the Digital Twin market with around 36% market share, supported by major companies like Rockwell Automation, Microsoft, IBM, and Siemens [18].

2.2. Early Development of Digital Twin Technology

Early applications of Digital Twin technology were seen in the aerospace and automotive sectors, with NASA using a Digital Twin-like system to monitor spacecraft performance during the Apollo missions [19]. Despite limitations in computational power and data storage at the time, these applications showcased the potential of DT technology to improve operational efficiency and decision-making [20].

2.3. Expansion of Digital Twin Technology in Manufacturing

In the automotive sector, companies like General Electric (GE) and Tesla have adopted Digital Twin solutions to simulate and monitor product lifecycles—from design to post-sale performance [21]. By integrating sensors and machine learning algorithms, these systems offer real-time insights into machine performance, enhancing predictive maintenance and production quality [2].

According to Gartner, over 50% of large industrial companies are expected to use Digital Twin technology by 2025, potentially achieving efficiency gains of up to 10% [22]. The market for Digital Twin technology is projected to reach USD 39 billion by 2026 and USD 110 billion by 2028 as can be seen in Figure 2, driven by increasing adoption across various sectors, including healthcare, smart cities, and energy distribution.

2.4. Digital Twin in CNC Machine Monitoring

Digital Twin (DT) technology is transforming CNC machine monitoring and optimization. Companies like FANUC [23] and Siemens offer DT solutions that create virtual replicas of machines, enabling real-time tracking of metrics such as tool wear, efficiency, and product quality [24].

Siemens’ Sinumerik ONE integrates DT technology with CNC systems, providing real-time feedback for optimizing operations and reducing downtime [25].

2.5. Digital Twin Technology in Healthcare and Smart Cities

Digital Twin (DT) technology is making significant strides in healthcare and smart cities. Companies like Philips and GE Healthcare use DT technology to create virtual models of medical equipment, enabling predictive maintenance and improving reliability in critical care settings [26]. Researchers are also developing patient-specific Digital Twins to simulate disease progression and treatment outcomes, advancing personalized medicine [27].

In smart cities, DT technology helps optimize urban environments [28] by simulating traffic flows and managing resources like water and energy. Singapore’s Virtual Singapore project is a leading example, offering real-time insights for efficient urban planning [29].

2.6. Future Prospects of Digital Twin Technology

The future of DT technology looks promising, driven by advancements in AI, machine learning (ML), and the Internet of Things (IoT). A McKinsey report notes that AI integration will enhance DT systems’ simulation and predictive capabilities, optimizing complex processes [30]. In manufacturing, DT technology will further integrate with autonomous systems, enabling machines to make real-time adjustments, boosting efficiency and allowing mass customization [31]. Additionally, combining DT technology with 5G connectivity will open up new possibilities in remote surgery, autonomous vehicles, and real-time smart city management [32]. As the DT market grows, its applications will expand across various sectors of the global economy.

2.7. The Architecture of Digital Twin

Digital Twins create a virtual counterpart of a physical system, offering a representation that mirrors its operational state. This virtual environment enables a 3D visual monitoring system for comprehensive oversight of the physical system. Digital Twin (DT) concepts vary by domain due to the specific nature of information in each field [33].

The architecture of Digital Twins consists of the physical world, the virtual world, and their connectivity [34] as is designed in Figure 3. In the physical world, sensors gather data, edge computing enables initial analysis, and security measures protect information [35]. The virtual world includes a digital replica, advanced data processing using ML and AI, and databases. Connectivity ensures secure data transfer through interfaces like the internet and Bluetooth.

2.8. The Maturity Levels of Digital Twin Technology

Most industrial companies are expected to adopt Digital Twins for increased efficiency, but currently, less than 5% have made tangible progress. A key aspect is the classification of Digital Twin maturity levels present in Figure 4, independent of industry or technology.

In the literature, most DT concepts fall within maturity levels 0–3 [35], with few integrating real-time data due to challenges like data collection, filtering, and device malfunctions, which can result in anomalies or missing data points. So far, 3D simulation modeling with a temporal component has been preferred, as it enables running multiple “what-if” scenarios using real data.

2.9. Digital Twins in the Smart Manufacturing Process

Smart manufacturing systems, as depicted in Figure 5, as defined by the National Institute of Standards and Technology [36], must be capable of responding in real time to changes in conditions and requirements. Digital Twins, which are digital representations of physical elements, utilize IoT data and sensors to monitor operations and support decision-making processes. They are used at various stages of the product lifecycle or the manufacturing system.

3. Research and Development Methodology

3.1. Models—WPF Applications with the Model-View-ViewModel (MVVM) Design Pattern for CNC Machine Interface

Creating a CNC machine operator interface with Windows Presentation Foundation (WPF) [38] utilizing the Model-View-ViewModel (MVVM) design pattern offers a structured, scalable, and maintainable solution for desktop applications. WPF, as a UI framework designed for developing visually appealing desktop applications, when paired with the MVVM pattern, facilitates a clear separation between the user interface (View), business logic (Model), and data binding logic (ViewModel). This makes it a favored choice for industrial applications, including CNC machine monitoring. In the context of a CNC machine operator interface, MVVM and WPF provide a flexible and powerful platform for real-time monitoring, control, and diagnostics.

• Real-Time Monitoring: The application can monitor critical CNC parameters such as spindle speed, feed rate, tool wear, and temperature. Using data binding, the ViewModel updates the UI in real-time with the latest sensor data from the CNC machine.

Example: a graph displaying spindle speed over time can be dynamically updated by binding the graph’s data source to a collection in the ViewModel, which receives live updates from the machine’s sensors.

• Machine Control: The interface allows operators to control the CNC machine remotely. Buttons in the UI can be bound to commands in the ViewModel, enabling operations such as starting, pausing, or stopping the machine. The ViewModel would then translate these commands into machine-specific API calls or instructions.

Example: a “Start” button would be bound to a StartMachineCommand in the ViewModel, which communicates with the machine’s control system to initiate machining.

• Data Visualization: Using WPF’s data visualization libraries (like OxyPlot or LiveCharts), operators can view graphs, tool paths, and other CNC performance data. These can be customized to show historical trends or real-time machine data, providing insights into the CNC process.

Example: a real-time plotting graph that displays tool wear progression during machining, enabling the operator to predict when maintenance is required.

• Fault Detection and Alerts: The MVVM pattern allows for the seamless display of alerts and notifications when the machine encounters errors or operational anomalies. The Model layer would handle the logic for detecting faults, while the ViewModel would notify the View to display alerts.

Example: when the CNC machine detects a tool failure, the ViewModel can trigger an alert in the UI, highlighting the issue and providing options for operator action.

• Historical Data and Reporting: Operators can access historical machine data, such as previous jobs, tool paths, and machine performance, using the WPF interface. This allows operators to review past performance and adjust parameters for optimal efficiency.

Example: an operator could pull up a report of machine performance over the last 24 h, presented as a set of graphs showing key metrics like spindle speed and tool wear.

• CNC Machine Connectivity: The ViewModel acts as the mediator between the UI and the CNC machine’s control system (e.g., via an API or communication protocol like OPC UA). This enables the application to send instructions to the machine and receive updates on its operational status.

Example: using the CNC machine’s API, the ViewModel can send toolpath data to the machine and receive feedback on the operation’s progress.

Key Components of a CNC Interface:

• Data Binding: WPF’s powerful data binding feature allows for real-time updates in the UI whenever the underlying data (Model) change. For a CNC machine, this could include real-time updates of the machine’s status (e.g., temperature, spindle speed, tool wear).
• Commands: In MVVM, user interactions (e.g., button clicks) are handled through commands in the ViewModel, which executes business logic in the Model. For example, a “Start” or “Stop” button in the CNC machine control panel would trigger commands that interact with the machine via the ViewModel.
• Dependency Injection: using frameworks like Prism or Unity, MVVM allows for dependency injection, where machine services (e.g., machine control APIs) are injected into the ViewModel, making the code more modular and easier to test.
• Notifications and Alerts: The application can be designed to notify operators of any anomalies, errors, or performance deviations in the CNC machine, allowing for timely interventions. WPF’s binding to ObservableCollection or INotifyPropertyChanged interfaces ensures real-time updates of any changes in the machine’s operational status.

3.2. Real-Time Notifications and Actions Using Microsoft SignalR Hub

SignalR, developed by Microsoft, patch 2.4, as depicted in Figure 6 for current research, facilitates real-time communication between clients and servers in web applications. It is used for interactive applications, providing live updates for chats, notifications, and online games.

Core Technologies SignalR utilizes WebSocket [39] for efficient communication. In the absence of WebSocket, it adapts other technologies such as Server-Sent Events (SSEs) and Long Polling, ensuring extensive compatibility.

Key Features:

• Simplified API: sending and receiving messages is easy to implement.
• Multi-Platform Support: compatible with various platforms and languages, including .NET and JavaScript.
• Connection Management: automates reconnections and redirects for reliability.
• Integrated Security: supports authentication and authorization, protecting against unauthorized access.

Common Uses:

• Chat and Messaging Applications: provides instant updates and bidirectional communication.
• Live Notifications: sends real-time notifications for news and social media platforms.
• Real-Time Monitoring: displays updated data in monitoring applications.

3.3. Dapper ORM

Dapper is a high-performance Object-Relational Mapper (ORM) designed for .NET, as depicted in Figure 7, optimized for scenarios where both speed and simplicity are essential. Unlike traditional ORMs, Dapper minimizes overhead by executing SQL queries [40] directly and mapping the results to .NET objects with minimal performance loss. It is developed by the Stack Overflow team and is widely adopted in performance-critical applications.

In the context of Digital Twin technology for monitoring manufacturing processes, Dapper can be a crucial tool for efficiently managing large datasets generated by IoT sensors, machines, and real-time monitoring systems. Here is how it can be used:

• Real-Time Data Querying: Dapper can fetch real-time sensor data from SQL-based databases efficiently. This is particularly important in Digital Twin environments where up-to-the-second data accuracy is necessary for simulating the physical manufacturing environment.
• High-Performance Data Operations: the performance advantages of Dapper make it ideal for high-frequency data operations in Digital Twin systems, where large volumes of machine and sensor data must be processed quickly.
• Batch Updates for Monitoring States: Manufacturing systems may need to frequently update machine states, sensor readings, or production statuses. Dapper can handle batch updates, ensuring that the digital twin stays in sync with real-world conditions without lag.
• Compatibility with Various Databases: whether your manufacturing system stores data in SQL Server, PostgreSQL, or other relational databases, Dapper can easily integrate, allowing for flexible data management across different platforms.
• Efficient Resource Management: with minimal overhead, Dapper can reduce the computational resource consumption of your monitoring system, which is critical when scaling a Digital Twin to monitor multiple production lines or machines.
• By using Dapper in your Digital Twin system for manufacturing, you ensure that data flow efficiently between the physical environment and the virtual model, enabling real-time decision-making and optimization of production processes.

3.4. Android Interfaces

Developing an operator interface for Digital Twin monitoring in manufacturing using Android Studio and Java [41] enables the creation of a highly customizable, portable, and efficient mobile application. With Android, operators can monitor real-time data from the manufacturing floor on tablets or smartphones, improving accessibility and response times.

Developing a Digital Twin operator interface for Android offers several key benefits to manufacturing environments, especially when real-time monitoring and quick decision-making are critical.

• Real-Time Data Access: Operators can monitor live data from the factory floor via an intuitive mobile app. The interface can display metrics such as machine status, energy consumption, temperature, pressure, or any other IoT data generated by the Digital Twin system.
• Remote Access and Mobility: Operators can access the manufacturing system’s Digital Twin from anywhere on the plant floor or remotely, increasing responsiveness and decision-making speed. This is especially valuable for large-scale facilities where operators are not always near desktop workstations.
• Interactive Dashboards: By utilizing libraries like MPAndroidChart or Android Plot, developers can integrate interactive graphs, data trends, and KPIs (Key Performance Indicators) into the operator interface. This allows for more dynamic monitoring, enabling operators to zoom into specific data ranges or investigate anomalies.
• Alert and Notification System: with Android’s notification APIs, the app can push critical alerts to the operators’ devices when real-time data signal equipment failure or process inefficiency, enhancing proactive maintenance and reducing downtime.
• Seamless Integration with Existing Systems: the app can integrate with the backend infrastructure of the Digital Twin, pulling data from SCADA systems, PLCs, and ERP systems, providing operators with a unified interface to monitor the entire production environment in real-time.
• Offline Mode: for environments with unreliable connectivity, the app can be designed with an offline mode, allowing operators to view and update data locally, syncing with the backend when connectivity is restored.

By developing an Android operator interface using Android Studio and Java, manufacturing operations can benefit from enhanced mobility, real-time insights, and faster response times, optimizing the performance of Digital Twin systems and overall production efficiency.

4. Detailed Design and Implementation

4.1. The Architecture of the Proposed Solution

The architecture of the proposed solution, presented in the Figure 8, brings in discussion as main technologies: .NET WPF, Android Java, Swagger API, SignalR, and MSSQL.

4.2. The Architecture of the Database

The proposed application uses a database created in MySQL using the SQL programming language. After its organization, it contains 10 tables.

During the design process of the internal MySQL database, normalization was aimed at avoiding redundant fields, synthetic keys, and circular references, while applying the following principles:

• A clear structure of the tables was established, using descriptive column and table names.
• Each table contains information strictly related to a single defined object.
• Each table has a distinct primary key.
• The data types of the fields have been optimized by applying appropriate functions.
• In Figure 9, we will illustrate the database diagram before deriving the tables.

4.3. The Monitoring Windows Application

The Windows software module is built using .NET 8 WPF. It contains the following components present in Figure 10:

• Controls: Contains reusable interfaces for other windows. For example, ResourceControl contains the design for each resource on the real-time map and is reused for each resource without the need to define these settings for every individual resource. This allows for configuring a responsive interface for each resource in the live map without having to rewrite the same code.
• Helpers: contains classes necessary for executing commands, value converters, and implementing the dialog system.
• Resources: contains all project resources such as images, styles, and global information that is available across all interfaces for reuse.
• Models: defines the tables and fields existing in the database for an entity.
• ViewModels: provides the functionality for each interface to allow the separation of processed data.
• Windows: the main interface that contains the other interfaces used in the application.

4.3.1. Configuring Real-Time Events

To receive real-time refresh signals, we created a connection to the Notification Hub server and configured the reception of the automatic refresh signal, as in Figure 11 and Figure 12.

In order to calculate the Connection Establishment Time, Retry Interval, Message Handling, and Notification Update we have used the formulas below:

Connection Establishment Time:

$$T_{connect}=T_{init}+T_{auth}+T_{SignalR}$$

(1)

where $T_{connect}$ = total time to establish the connection.

$T_{init}$ = initialization time of the SignalR and Notification Hub setup.

$T_{auth}$ = time to authenticate with the Notification Hub.

$T_{SignalR}$ = time to open the SignalR connection.

Retry Interval:

$$T_{retry}=T_{init}\times 2^n$$

(2)

where $T_{retry}$ = time between retries.

$T_{init}$ = initial retry time (e.g., 5 s).

$T_{auth}$ = number of failed attempts (exponential backoff).

Message Handling:

$$M_{received}=\left\{ms{g}_1, ms{g}_2, \dots ., ms{g}_n\right\}$$

(3)

where $M_{received}$ = a set of messages received from the server or Notification Hub.

Each message $ms{g}_i$ is processed based on predefined event handlers.

Notification Update:

$$N_{update}=M_{push}+M_{realtime}$$

(4)

where $N_{update}$ = notification updates shown to the user.

$M_{push}$ = messages from the Notification Hub (push notifications).

$M_{realtime}$ = real-time messages received from the SignalR hub.

4.3.2. Data Model Configuration

To map the fields from the database with those used in the code project, we created three data models: user, production-required information, and CNC machine operating parameters as depicted in Figure 13.

4.3.3. Logging Method in the Application

To perform SQL queries, an asynchronous method was implemented using the Dapper package for SQL queries. The password is encrypted in the database, and to verify if the password is correct, an AES decryption method was implemented, as in Figure 14.

The decrypted password is calculated using the formula below:

$$D_{pass}=E_{pass}^{-1}\left(key\right)$$

(5)

where $D_{pass}$ = decrypted password.

$E_{pass}$ = encrypted password.

$key$ = secret decryption key.

4.3.4. Retrieving the Necessary Information for Each CNC Machine to Process the Orders

To retrieve the information necessary for processing orders on each CNC machine, a new SQL query was created, as depicted in Figure 15.

4.4. Development of the Android Module for Recording the Operations Performed by the Operator

To record all operations and actions performed by the operator, a module was created for the CNC machine operator, an application developed in Android Java.

4.4.1. Global Declaration of Reusable Information at the Code Project Level

To work more efficiently and ensure high application performance, all reusable information at the code level was defined in a global file, as in Figure 16.

4.4.2. Defining the Data Model for the Information Required by the CNC Machine Operator

A model was defined that contains the fields necessary for the operator as an information flow in the CNC machine setup or production process, presented in Figure 17.

4.4.3. Method for Retrieving the Data Required by the Operator (Figure 18)

A method was created that generates a query using the JDBC driver for MySQL.

4.5. Create the 3D Virtual Model of the Machine

Digital Twin technology enhances manufacturing by providing a virtual representation of physical assets for real-time monitoring and analysis. Key steps in creating a 3D model of a machine using AutoCAD include the following:

• Data Collection: gather real-time operational data from CNC machines to inform the Digital Twin.
• Model Development: create an accurate 3D model in AutoCAD, reflecting the machine’s design and specifications.
• Integration: embed monitoring features into the model using data from the Digital Twin to visualize performance and detect anomalies.
• Simulation: use the Digital Twin for scenario simulations to predict machine behavior and optimize operations.
• Continuous Improvement: update the Digital Twin with new data for ongoing refinement of the 3D model.

The integration process begins with data collection from CNC and printing machines, using the Notification Hub Server to capture real-time operational data (e.g., spindle speed, temperature, print pressure). These data are aggregated and synchronized across machines at an average performance of 1 s for each 1 of 150 resources. Data synchronization relies on SQL databases, and real-time data are linked to a 3D model using platform NET for visualization.

The simulation process involves running scenario-based simulations to monitor machine performance under different conditions.

The continuous improvement process is driven by ongoing data collection, with real-time data feedback used to update the Digital Twin. Models are refined as machines wear down or as production conditions change.

Below, the 3D CNC Model design can be found in Figure 19:

5. Results

5.1. Logging Interface for Windows and Android

Logging into the application, whether it is the Windows app or the Android app, is done using a username and password created by the application administrator. The password is encrypted in the database.

5.2. Windows Module for Monitoring the Production Process

This module is designed primarily to visualize the necessary details for the production process at the level of each CNC machine, as well as to monitor the operational parameters of the CNC machines.

When an operator performs a specific action on a CNC machine, the information in this module automatically updates according to the action taken by the operator. This real-time monitoring allows for immediate insights into machine performance, facilitating quick decision-making and enhancing overall efficiency in the production environment. Additionally, the module can generate reports and analytics based on the operational data collected, further supporting process optimization and maintenance planning. An example using the Windows application is shown in Figure 20.

5.3. Android Module for Monitoring Operator Activities on CNC Equipment

The Android module for the operator is designed to record the actions taken by each operator at each specific CNC equipment and to provide visibility into the details of each order. This module requires each CNC equipment to be equipped with a tablet on which this software module version 1.0 is installed.

Operators can view a list of all scheduled orders for the respective CNC equipment and can start or complete setup or production operations. At that moment, the information is updated in the Windows module for monitoring the production process, ensuring real-time synchronization between the two modules an average of 1 s per action.

This integration allows for enhanced tracking of operational activities and improves communication between the operators and the monitoring system. Moreover, the module can facilitate documentation and the reporting of actions taken, contributing to more efficient production management and accountability. An example using the Android Module is shown in Figure 21.

5.4. Results of the Software Modules Applied to 150 Printing Resources

The Digital Twin-based monitoring system was implemented across 150 printing machines distributed over 10,000 mp on a Microsoft Windows Server 2022 infrastructure.

Server specifications:

• Processor: Intel(R) Xeon(R) Gold 6346 CPU @ 3.10 GHz 3.09 GHz (4 processors)
• Installed RAM: 16 GB

The system performed real-time data synchronization at an average rate of 1 s per each 1 of the 150 resources. With 150 machines being monitored in parallel, data updates were efficient and occurred every 1 s for each one resource from the group of machines. The production time on each resource was reduced by 10% due to the synchronization of the information between operators that work on the machines, maintenance, and coordinators of each group of resources.

• Total resources monitored: 150 machines.
• Data synchronization rate: 1 s average/resource.
• Production time reduced by approx. 10%.

We have tracked the performance of the resources for 1 week and, as the result of real-time actions performed by the resource, we have obtained 714,644 calls for resource actions, as showed in Figure 22.

These actions are reported to approx. an average of 500 orders performed per day.

In Figure 23, we have the number of resources connected to the Notification Hub server in 1 week of resource productivity and performance tracking:

The performance of synchronized data resulted in an average of approx. 1 s per resource as can be analyzed in Figure 24:

The average of production time had a decrease of 10% for the analyzed week compared with the previous week. The results are presented in Figure 25, where we are looking at the Act.Hours (Actual Production Time Duration):

In order to bring in evidence of the unicity of the current research solution, we have conducted a comparison of current similar existing software solutions in Figure 26 below.

Real-time operational data from the CNC machines were collected, synchronized, and analyzed through the proposed monitoring software, integrated with a .NET interface and an SQL database for data storage and retrieval.

6. Conclusions

The implementation of the Digital Twin-based monitoring system for 150 printing resources demonstrated significant improvements in production efficiency and real-time machine monitoring capabilities.

• Real-time Synchronization:

The system achieved a 1 s average synchronization rate per resource across 150 machines. This rapid data update allows for immediate reflection of operational changes, ensuring smooth communication between operators, maintenance staff, and coordinators. The 714,644 actions logged within a week reflect the high-frequency monitoring and the system’s capacity to manage multiple resources efficiently.

• Reduction in Production Time:

The production time per resource decreased by 10% compared to the previous week, highlighting the effectiveness of real-time data synchronization and improved coordination between various departments. This time reduction is crucial for increasing throughput and reducing machine downtime, contributing to better overall productivity.

• Windows and Android Modules:

The Windows module effectively monitors CNC machine parameters and updates operational information in real-time as operators perform actions. The Android module complements this by enabling operators to log activities and track orders at their respective CNC machines, with real-time updates across both platforms. This dual-module system enhances operator accountability, operational transparency, and the coordination of production tasks.

• Server Performance:

The infrastructure utilized, including a Microsoft Windows Server 2022 with Intel Xeon Gold 6346 processors and 16 GB RAM, performed efficiently in managing the large volume of data updates and synchronization tasks for 150 machines in parallel. This scalability and performance indicate that the system could handle larger or more complex operations in the future.

• Comparison with Other Solutions:

When compared to existing software solutions, the proposed system demonstrated several advantages, particularly in real-time data synchronization, customized functionalities, and integration with IoT devices. The unique combination of .NET interface and SQL database for storage and retrieval offers flexibility and scalability for future expansions or integrations with other systems.

• Order Tracking:

The system tracked an average of 500 orders per day, showcasing its capability to handle a high volume of operational data, further affirming its potential to enhance workflow efficiency and ensure timely completion of production tasks.

• Performance Tracking and Notifications:

The integration with the Notification Hub to track resource connections demonstrated a reliable communication mechanism between machines and operators, ensuring that any discrepancies or required maintenance could be addressed promptly.

Overall, the Digital Twin system for CNC machine monitoring has proven to be a highly efficient and scalable solution, delivering 10% faster production times, real-time updates, and effective operator–machine synchronization. This makes it a promising tool for enhancing productivity in the manufacturing industry.

Development Directions

This research marks the initial step in exploring and implementing the Digital Twin concept. Currently, the focus is on real-time monitoring of production processes and the operating parameters of CNC (Computer Numerical Control) machines. Future development will involve advancing towards a system that automatically collects operating parameters directly from the CNC machines.

Key aspects of this evolution include the following:

• Automated data collection system: The development of software or hardware capable of automatically retrieving CNC operating parameters without manual operator input. This will ensure more accurate and efficient data collection.
• Performance analysis and reporting software: A dedicated software solution will be created to analyze the gathered data and generate detailed reports on CNC performance. These reports will provide valuable insights for optimizing production processes and identifying potential issues early on.
• Predictive productivity and maintenance system: leveraging historical data and advanced analytical algorithms, the system will predict future CNC performance and identify the need for preventive maintenance, helping to avoid downtime and improve overall efficiency.
• Monitoring the activity of one specific equipment using cameras and sensors.

In summary, this research and development in the Digital Twin domain aims to enhance the efficiency and reliability of production processes by integrating cutting-edge monitoring and data analysis technologies.

In conclusion, the project demonstrated that the use of modern technologies and innovative concepts can significantly contribute to the improvement of industrial processes, leading to increased efficiency, productivity, and competitiveness in the industry.