A Digital Twin of the Milling Process

Case Study Overview

In the manufacturing industry, digital twins are increasingly being used to optimise production processes and reduce downtime. One area where digital twins are particularly useful is in milling, a widely used manufacturing process that involves the removal of material from a workpiece using a rotating cutting tool.

De-risking software systems at sea using a digital twin

In the manufacturing industry, digital twins are increasingly being used to optimise production processes and reduce downtime. One area where digital twins are particularly useful is in milling, a widely used manufacturing process that involves the removal of material from a workpiece using a rotating cutting tool.

By using a digital twin of the milling process, manufacturers can monitor and optimise their milling operations in real-time, providing insights into the machine’s performance and productivity. This technology can help identify potential issues and suggest optimal settings and parameters, ultimately improving efficiency, reducing waste, and increasing productivity.

Productive Machines have created a digital twin of the milling process to identify vibrations and then automatically personalise the tool path parameters to eliminate chatter.

The digital twin of the milling process created by Productive Machines is a virtual model that replicates the behaviour and performance of a physical milling machine and its operations in real-time. This digital twin uses sensors and data from the milling machine to generate a live model that simulates the machine’s behaviour.

Furthermore, using the Productive Machines digital twin technology allows the user to access products more quickly and cheaply due to the reduction in waste and resources. Although additional computation is required on computers, it is minor compared to the resources involved in traditional machining processes. Thus, the Productive Machines digital twin technology is a valuable solution that can help the machining industry achieve its productivity and sustainability goals while reducing costs and lead times.

How does it work?

The digital twin of the milling process helps manufacturers monitor and optimise their milling operations, providing insights into the machine’s performance, productivity, and potential issues. The digital twin also identifies ways to improve efficiency, reduce downtime, and prevent machine failures.

The digital twin-based learning platform comprises three modules: a learning system, a process planning tool, and a process monitoring and control tool. The process planning tool is integrated into existing CAM software like Siemens NX and provides an offline digital twin of the machining process. Process planners can use this tool to proactively identify potential problems and take countermeasures to avoid them while optimising the process.

The process monitoring and control tool includes an online digital twin that detects anomalies such as tool wear and chatter and adapts key parameters to control them, thereby optimising productivity and reducing waste. All three modules are interconnected, and one unique feature of the software is that machine tools can learn from each other to optimise machining processes.
One of the main advantages of the Productive Machines digital twin is the two-way integration between simulation and measurement. The technology enables the identification of the dynamic characteristics of the machine, which can be used to improve the machining process. It is also possible to characterise material data so that the technology can identify the type of material used (aluminium, titanium, or anything else). The measurement data from the machine is fed to the simulation environment, and the simulation provides the reference data for future predictions. This means that it is possible to start monitoring the process in a matter of seconds, rather than having to wait for an hour or more.

By using data from the physical milling machine and applying advanced analytics and machine learning algorithms, the Productive Machines digital twin predicts potential problems and suggests optimal settings and parameters for the milling process. The Productive Machines digital twin is not based on predicting trends from mass observations, but on predicting the physics of a process. Their predicted maintenance service looks at trends based on signals coming from machine tools.

Benefits

In the manufacturing industry, there is often a wastage of resources in the form of time, energy, and materials due to the need to typically produce three parts before achieving the desired productivity and quality of the process. This is particularly problematic for high-volume sectors where materials can be a significant cost item. Furthermore, recent industry trends aiming for net-zero emissions by 2050 highlight the need to address these issues as the emissions generated during the trials contribute to an increased carbon footprint.

The Productive Machines digital twin enables users to run numerous trials in a simulation environment and identify the best process parameters without wasting resources on the machine tool. As a result, the benefits of cost reduction and lead time reduction are significant, and customers can access products quicker and cheaper.

Moreover, the adoption of Productive Machines digital twin technology allows companies to optimise their machine tools by enabling them to reach the optimum settings quickly. In the absence of digital twin technology, it is challenging to determine how close one is to the optimum settings. Therefore, companies tend to adopt a conservative approach to avoid breaking cutting tools, causing wastage, and scrapping parts through failure to meet dimensional tolerance or surface finish.

Challenges and lessons learnt

One challenge faced by Productive Machines is convincing customers of the benefits of their solutions. In order to operate effectively, production machines tools will need to have time allocated for cutting tool calibration. The additional machine productivity far outweighs this downtime, but production is typically measured on operational availability rather than output efficiency. This lack of awareness and understanding can be a barrier to providing these solutions to clients. While Produductive Machines have alternative solutions that do not require measurements from machine tools, they may not be as powerful as those that do. Thus, the team needs to educate customers and address any concerns they may have in order to fully implement their solutions and maximise their impact.

Another challenge comes in the form of the skills gap issue in the manufacturing industry. The Productive Machines digital twin aims to bridge this gap. Many companies face challenges in adopting new solutions due to the lack of qualified engineers to operate them. In some cases, companies have more software than their engineers are capable of running, which creates a skills gap issue.

To address this issue, Productive Machines has been developing an automated version of their product that can be run with very low skill levels. They aim to deliver a solution that does not require a special set of skills. The only additional skill level required is for the user to upload a manufacturing file to their website.

Overall, the aim of their solution is to make the life of the client easier by providing an automated solution that can bridge the skills gap and improve productivity in the manufacturing industry. By addressing the skills gap issue, companies can adopt new solutions more easily and increase their productivity.

Outcomes

In the machining industry, many companies face productivity and quality issues that result in the production of around three parts before achieving the desired results. This process is not only wasteful in terms of resources but also increases emissions, which contradicts the industry’s aim of net-zero emissions by 2050.

Thus, improving productivity and reducing costs are always important goals. The digital twin created by Productive Machines addresses these challenges. Using the digital twin to optimise manufacturing processes can lead to savings before production starts, during production, and even after production.

The easiest thing to measure in manufacturing is cycle time. By measuring how long it takes to produce a part, companies can easily see improvements in productivity. When implementing the Productive Machines digital twin, companies can measure how long it takes to complete the milling operation and compare it to their previous process.

In some cases, companies have seen up to a 20% improvement in productivity compared to their previous process, with an improved range of 7% minimum to 110% maximum. The savings range is large due to the varying levels of optimisation that companies may have already achieved. For companies with skilled engineers who have optimised their processes, the savings may be less, but still considerable.

In addition to waste reduction, employing the Productive Machines digital twin technology also results in cost reduction and lead time reduction. Continuous improvement is a standard practice in the industry, with an aim to reduce costs by 3 to 4% every year. After five years, this results in a cost reduction of roughly 15 to 16%. However, using the Productive Machines digital twin technology, some companies have reported a 20% cost reduction achieved in just one month.

By minimising the need for post-processing inspection, companies can save time and resources, especially for high-volume manufacturing. The digital twin can help identify areas that require improvement, even if it is impossible to eliminate all problems.

Special thank you to Erdem Ozturk, Chief Executive Officer of Productive Machines, and Richard Nevill, Commercial Director at Productive Machines, for their willingness to share their time, expertise, and experiences with us.