Trending for years and still widely discussed among simulation enthusiasts and experts – the topic of digital twins.
Contents:
- Two types of digital twins
- Digital twins and simulation
- In supply chain
- For business processes
- In manufacturing
- For ports and rail logistics
- For oil and gas
- Cloud computing
What is a digital twin? You would be hard-pressed to find the same exact definition twice. Eventually though, they all can be narrowed down to two types: observer and virtual.
Two main types of digital twins
In its essence, a digital twin is a set of computer-generated models that map a physical object onto a virtual space with the most up-to-date data.
As Leon McGinnis, Professor Emeritus in the Stewart School of Industrial and Systems Engineering stated at the Winter Simulation Conference 2022, there are 2 types of digital twins:
Observer
- Serves to observe how objects and systems work.
- Used largely for asset management as it always shows the current state of the real twin.
- Created mostly for such things as wind turbines, jet engines, and generators at manufacturing facilities.
The fundamental part of these digital twins is capturing information about the state of resources and the units of flow (jobs, patients, work orders, etc.). Then when the state changes, it is detected by sensors or transactions and sent to a digital twin helping to always track the health of the real twin.
Virtual
- Serves to design new things and test them before a real twin is created.
- Created for such things as aircraft, rotating machinery, automobiles, and integrated circuits.
We will focus on the observer digital twins as these are the ones that can be built upon simulation models.
How is this related to simulation modeling?
The two key elements of a digital twin are a dynamic simulation model and data that reflects the current state of a live system. With the model and the data, it is possible to build a powerful digital twin for experimentation, analysis, and interpretation of results, so that you can ask ‘what-if’ questions, understand system behavior, and verify at multiple levels.
See below how companies around the world combine simulation modeling and digital twin technologies to resolve business challenges.
Reducing supply chain costs and better forecasting
Accenture created a digital twin based on a simulation model of the entire supply chain for a US-based exercise equipment brand. It focused on predicting order to delivery times to help reduce them, and on providing a basis for a smart inventory-allocation solution to improve planning.
The digital twin took the input data from various Amazon services and spreadsheets all connected through Amazon S3. Meanwhile, for the business analytics on the model’s outputs, it was connected to Tableau.
As a result of the digital twin initiative, Accenture saw a 57% increase in accuracy for order to delivery forecasting and a 20% cost reduction for inventory allocation logistics costs.
Business processes optimization
A digital twin that was built for Siemens aero-derivative gas turbine division emulated global maintenance repair and overhaul operations. It used vast quantities of data about the customers, supply chain, production, and maintenance to improve productivity and efficiency in customer operations and asset management.
With the new solution, Siemens could capture and forecast the system’s KPIs, visualize the fleet and maintenance facility operations, identify bottlenecks in the system, and run both quick ‘what-if’ and detailed scenarios to support investment decision-making.
One of the largest turbine manufacturers in the world had a very promising five-year portfolio of gas turbines to produce and was planning an optimistic 30% net margin. However, soon the company realized that some projects in the portfolio were facing significant delays.
The digital twin solution helped the company’s management see the whole picture of the projects. In addition, they identified the best scenario which allowed them to reduce the inevitable delay from two years to nine months while still having a net profit of $104 million.
Case study and video →Improving production lines and testing maintenance policies
CNH Industrial, a global leader in capital goods, wanted to test a digital solution for evaluating and selecting different maintenance policies for their production line. It was crucial to identify the optimal one because the cost of a single minute of downtime could be more than $160k, therefore even a very small improvement could save a lot of money.
As a pilot project, the team decided to simulate a single manufacturing line for Iveco Daily van chassis welding and focus on automatic welding stations.
CNHI used a simulation-based digital twin to monitor and forecast the health of a critical component of the welding station and facilitate significant downtime reduction. The tool provided a wide variety of data and helped to analyze and compare scenarios — enabling a quick understanding of how changes could impact maintenance costs.
Lagor, an Italian manufacturer of magnetic transformer cores, faced issues scaling the manufacturing processes while ramping up production and expanding the business.
A digital twin of the shop floor, working with the real data, helped simulate the production and decision-making processes as well as investigate the production plan.
With the new digital twin tool, Lagor engineers could successfully rearrange production sequences in a risk-free environment using a ‘what-if’ approach. Therefore, they could efficiently avoid bottlenecks in real-world production processes.
Container yard planning and longshore worker training
Terminal San Giorgio in Genoa, Italy created a decision-support system combining simulation, digital twin, and AI technologies and built a reliable evacuation strategy for emergency situations at the terminal. The system could recalculate paths to safe zones on the fly when accidents occurred and communicate the paths to the in-house alert tool.
The engineers also showed that AI capabilities together with simulation would improve the overall terminal throughput by 20%.
Digital twins also come in handy in human resource management. The British Columbia Maritime Employers Association is an organization created by maritime employers operating in the five port regions of British Columbia, Canada. The association represents over seven thousand active longshore workers and is involved in worker training and job dispatch.
To begin, the BCMEA created datasets and dashboards to understand current and historical states for longshore labor dispatch in British Columbia. Following on from the datasets and dashboards, the team developed a predictive analytics solution. It would enable the BCMEA to look into the future and analyze scenarios, such as the opening of a new container terminal or the effect of training more truck drivers.
A digital twin of Vancouver longshore labor dispatch served as the basis for developing predictive analytics and helped determine an optimal three-year return on investment for training.
Oil and gas well construction
At the drilling rig, all technical units are integrated, so any kind of delay in machinery increases the critical time path. As a result, it reduces the efficiency of the overall performance and can lead to financial losses.
To handle possible variations and inefficiencies of the well construction process, Transocean engineers collected and assessed measurements at dozens of rigs, including machine and crew timing. They built a digital twin of the oil well to process this data and analyze the interdependencies among various operations.
The statistics from the digital twin were fed back to operations personnel and rig managers so that they could assess the well crew performance and identify causes of time loss. The initial results indicated that over 20% of time could be saved by implementing the digital twin.
Cloud computing for digital twin solutions
Together with dynamic simulation and current data, high-performance computing capabilities are vital for building a digital twin. AnyLogic Private Cloud is a secure and powerful platform that simplifies the integration of simulation models into operational workflows and eases the creation of digital twins.
In Cloud, you can perform complex multi-run experiments with models faster and more efficiently than on a regular computer. Being scalable by design, it can rapidly respond to meet computing demands and execute experiments on multiple nodes and cores.
Companies around the world from various industries turn to digital twins as a solution to complex business challenges where traditional Excel spreadsheets are no longer efficient. Start by building a simulation model, add current data, run experiments in AnyLogic Cloud, and combine them all into a solid and powerful digital twin that will support your decision-making.
Simulation-based digital twins are also the focus of our white paper, An Introduction to Digital Twin Development. It contains case studies that help demonstrate the development of digital twins and their benefits.
