Why does everyone today, want a Digital Twin?

It was a few years ago, while attending an industry 4.0 conference that it hit me. Amid talks of IOT, BigData, AI & ML, I noticed everyone dismissing the Digital Twin, due to the complexity and potential cost of digitally mirroring the product and manufacturing processes.

It was at that moment when everyone was going one way and trying to take avoid all risk, that I realized I had to go the other – I’d realized the Digital Twin wasn’t some new tech gimmick, but would be sooner or later, the de facto for every organization to survive and thrive over the years ahead.

And the Covid-19 era has only served to accelerate that!

There are so many possibilities, ranging from virtual models of products and processes, complete shop-floors in manufacturing and production, to smart cities, and even the human body.

Creating complex networks and models of digital twins which can gather relevant data regarding almost anything is probably something that the not so distant future holds for industry and humankind everywhere.

The Digital Twin is now

The technology has however been expanding rapidly over the last few years. More and more companies offer solutions for digital twins, including ourselves (www.xdinnovation.com ), within different areas and for various industries.

The concept of the digital twin is made possible by the advances in CAD, Simulation and Internet of Things (IoT) technology. It is now much a easier to visualize components, equipment, processes and people. It’s straightforward to attach sensors and actuators to a physical object– such as a part within an asset or the entire asset itself – to capture contextual and operational data and control the object from its digital twin.

This is often characterized as a bi-directional flow of data from the product to the manufacturing line, from the physical to the digital and back.

However, each digital twin will involve multiple threads of information. This is not simply between the physical object and its twin, but also between the twin and the enterprise systems – such as CAD, PLM, ERP & MOM – that help create the initial model and supply supporting data, in real time or near real time, to build a complete picture of the object or asset.

In addition, data flows between the asset, its digital twin and everyone that needs access to the twin to view, manage or manipulate the data it holds. The more advanced analytics and simulations you perform through the digital twin, the more people will be involved in the process.

But creating these complex systems is far from easy – a good amount of resources are needed ranging from people, money, and time, to smart components, sensors, servers, and processing power which at the same time needs to be compatible in order to communicate with each other.

The process digital twin

The two fundamentals of the Digital Twin creation process are:

1. Data:

The input data is:

• number of machines,
• routes,
• number of employees,
• warehouses,
• internal transport,
• number of robots,
• operation times,
• product data (MBOM, PDM)

This data typically resides in PLM/ERP/MRP systems. But for the needs of the twin, this data must be available in databases for: data analysis, adjustment, supplier data, order dates, etc

With a digital twin, this data also becomes more available for decision makers across the enterprise in order to speed up and simplify the decision making process.

2. Simulation

Secondly, a simulation model is developed both of production processes and for analysis of individual components and models in CAD/MBSE/PLM programs. These documents also are transferred to the database for further analysis and interpretation.

Simulation experiments are performed. The model consists of several machines (processor), operated by workers, a robot, and warehouses and to complete the items on the line, transport vehicles/AGVs are used.

3. Reap the rewards

By creating a digital twin of the manufacturing process, it’s possible to continuously test, analyze, and make improvements and retest, to get the best results. In this way, many improvements can be made without actually testing physically.

This solution saves considerable time and money, as changes on the lines are implemented only when simulations produce the required results. The output data from the simulation for example are:

• process bottlenecks,
• storage problems,
• transport problems between machines,
• idle times,
• productivity,
• lead times,
• machine layout, etc.

Further advantages of the Digital Twin

The advantage of the digital twin is to improve manufacturing agility. The ability for the users to access documentation, models and simulations at anytime, provides the adaptability to neceessary in today’s ever-changing world.

Add to this:

• demands from the B2B and B2C Customers, requiring a more personalized product, flexible production and potential batches of one.
• introduction of certain standards both in terms of product performance, but also in terms of how the company operates, when production is changed.

In making these changes in a digital version, these can be made at anytime.

Barriers for creating a digital twin

The benefits of introducing a digital twin are significant, however, there are also some barriers to implementing these solutions:

The Required data

In creating a digital twin, there exist a few concerns:

The first barrier is the production data processed. This data is often basic data, from which the functionality of not only the production process, but also the entire enterprise is based. In this area there is data about suppliers, lead times, costs, set of machines, employees, failures, etc.

In addition, this area includes also all documentation, i.e. EBOM/MBOM, CAD drawings, standardization cards, quality cards, machine breakdown cards, results from analyses and simulations, etc.

With such a wide range of data, manufacturers are afraid, that by creating a digital twin of the documentation and this data, the data will leak.

Simulation of manufacturing processes

Due to the implementation of the digital twin in this area, the most common problem for manufacturers, especially for small manufacturers is the problem of access to appropriate software.

The second challenge is confidence in the results of the analyses and simulations, along with the proposed improvements.

Reading data from machines and robots

When using your Digital Twin to make almost real-time adjustment, the challenge surrounds the volume of connected data from multiple machines, equipment and robots at one time and all in real-time.

This is a large amount of data that, in order to be used as a digital twin, must be properly stored. Suitable databases and folders are required in order to then analyze this same data. Much of this data gathering is realized from PLC, HMIs, which are integrated with MES/MOM and beyond.

Standardization

Manufacturing requires multiple tasks to be performed at the same time and for most at an ever-increasing level of quality and complexity. For this purpose, standards are introduced for the implementation of individual production tasks, so that each employee shift can make products to specific standards. In creating a digital twin, all standardized instructions and procedures for the implementation of work, should be available in the digital version in the databases of the enterprise, so that every employee can have access to them.

And this assumes each worker is open to changing their approach and accessing this information.

Appropriate IT infrastructure

In terms of IT infrastructure, the biggest problem with the digital twin is building the right databases to store the data, and the servers. Currently, mostly databases are created in the cloud. However, many manufacturers are afraid of data leakage and loss of sensitive data.

Data security

Due to the collection of large data sets, enterprises are choosing to store them in the cloud. This is a flexible solution, in which dedicated people have access to data from any place. However, it is necessary to monitor movements on servers, but also to implement appropriate security solutions.