A Better Way: Finding Efficiencies in the Product Design and Manufacturing Process
Product realization: arguably it’s the most vital part of the design and manufacturing process. It’s when a real product, the physical embodiment of your IP is turned from an idea into reality. What’s more, without a well-crafted product design and production process – all coordinated to work together to ensure that the assembly flows seamlessly across the manufacturing floor – your idea will remain nothing more than a beautiful drawing or something less your original desire. For years, the way products are designed and their manufacture planned has remained more or less similar – complete with some inherent fault lines that can increase costs and delay development. In the spirit of innovation that’s essential for manufacturers to survive, Siemens has looked at the process to see if it can be streamlined. In this article, Aaron Frankel, Senior Marketing Director Manufacturing Engineering Software at Siemens PLM Software, and Jan Larsson, Senior Marketing Director EMEA at Siemens PLM Software, first discuss the inefficiencies that need to be ironed out before introducing the concept of the Digital Twin and what this means for the way goods are created and produced.
A remarkable symphony
To visit a modern manufacturing plant is to witness a jaw-dropping symphony of people, parts, materials, robots and machines – all working down to the minute or second to hit schedules. It looks incredible.
But behind the scenes the way goods are designed and planned for production remains based on aged processes. This is not to criticise anyone. A great piece of design is a major accomplishment. And it can be a hugely complex task that, in some instances, can involve millions of parts and thousands of people and partners – often coordinated across countries. Furthermore, in key markets such as electronics (faster processors, miniaturisation), automotive (the green agenda and emissions) and aviation (the green agenda and the drive for composite-based aircraft), there’s a relentless drive for improvement that means new designs must be achieved more quickly. Given the complexity, there’s an understandable reluctance to move beyond tried and tested development processes. This said, our customers report common problems across the development and production chain with some areas susceptible to causing costly delays.
One of the most important issues we see is that the design team uses separate systems to their manufacturing colleagues. In practice this can mean that designers pass over their creations to the manufacturing guys who have to try to create the subsequent process plan using the software they’re used to. In this scenario – which is quite common – information can get out of sync so it’s hard for everyone to see what’s happening. This increases the scope for failure.
Moving through into the creation of the manufacturing floor layout, we regularly encounter problems here too. These tend to be based around the fact that layouts are created using 2D floor plans and paper blueprints that take time and effort to create. While they are an essential part of the process, they are quite inflexible and we often find that any changes to the floor’s layout do not get reflected in the plans. This can become especially problematic in fast-moving markets such as consumer electronics where production systems must be continuously extended and refreshed. Why? Because 2D plans lack the intelligence and connectedness for manufacturers to know exactly what’s in production to make smart decisions and act quickly.
Following on from the layout, the manufacturing workflow typically progresses through into process validation. Here we also find a potentially major barrier to efficiency. It’s that manufacturers typically wait for actual equipment to be in place to see how it performs. If it doesn’t do as well as expected, it’s late in the day to look for an alternative solution and, in our experience, any breakdown in this process can cause serious delays.
Finally there are two other areas at the end of the production chain where customers report challenges: throughput performance and manufacturing execution.
Because of the complexity of the modern floor and, more often than not, a lack of coordination between different software and planning systems, it can be hard to isolate areas or cells in production that are delaying the line. And, when it comes to the last piece of the jigsaw – manufacturing execution – customers report that it’s often difficult to measure performance and see if how the process was planned to perform is actually how it is performing. Again the issue here is complexity, with challenges in feeding back information from the shop floor into product design, engineering and manufacturing teams.
So, what’s to be done? Below I take a look at our central idea around which we think all the key steps in the production process can gravitate and be improved: the Digital Twin.
The Digital Twin
A Digital Twin is a virtual copy of something that’s modelled to behave realistically. Without wishing to get too much into our products here, we’ve aligned our Project Lifecycle Management (PLM) tools to provide a complete digital framework around which Digital Twins can be modelled to realistically replicate the product design and assembly processes – from beginning to end. So what does this mean? Using the same stages we discussed above, we’ve highlighted some of the key capabilities we think are most valuable in this approach.
Design: A digital twin of the Siemens Nanobox industrial PC is analyzed for manufacturability by using Variation Analysis in Teamcenter to virtually build thousands of digital Nanoboxes. During the analysis, we discover that 9% of the builds are out of spec due to an insufficient tolerance specification. The issue is easily corrected in the digital design model and warranty claims avoided.
- Design: Using NX software (and other CAD systems), we can create a model of our product – and open them in Teamcenter as a 3D JT model. The software can virtually build literally thousands of variations of the product, just as it would be physically built, in seconds. It uses big data techniques, descriptions of the Product and Manufacturing and Information (PMI) (that specify the product’s tolerances and components), and a basic description of the manufacturing process to determine if we have any clashes. We trialled this approach on the design of one of our own electronics products. We were able to instantly see that the connector screws and corresponding video output connector holes had become misaligned. Unchecked, this could have resulted in warranty claims as the connector would separate from the PCB due to the manufacturing fault. Identifying design problems this early in the piece can save serious time and money – into and beyond the manufacturing process.
Process planning: Design and manufacturing work together to make more successful products by defining manufacturing process plans that link what needs to be made, how to make it, the resources required and where it’s made. This image shows definition of the steps for installing the outer case of the Nanobox and the estimated time for the shop floor operator to complete each action.
- Process planning: The Digital Twin can improve collaboration between the design and manufacturing teams to better plan what needs to be made, how it should be made, the resources needed and where it can be made. Let’s take an example of an updated assembly. Working with our tools, your planning team can use the new Bill of Materials (BOM) to input the new steps into a 3D working model of the current process. You can model any production system anywhere, so a team in Paris can be planning production for a site in Rio. With time estimates available for the new processes, the team can see if the workflow will still ensure that the average unit product targets are met. If not, the revised or new cells can be moved up or downstream and a simulation run again until the sequence ensures that production targets are met. The revised plan can be easily accessed by all stakeholders and signed off. If there are any issues revealed during the process, the design and planning team can work together to rectify these.
Layout: Using the manufacturing process plan information in Teamcenter, a digital twin of the Nanobox production system layout is designed in NX. The production system layout is also managed in Teamcenter ensuring the product design, manufacturing process plan and production system stay synchronized so that production starts without delay.
- Layout: With the floor layout, we recommend creating the Digital Twin – with all the mechanical, automation and resource details – and inseparably linking it to the product design and manufacturing eco-system. Using a combination of PLM tools, you can simply drag and drop cells, equipment and people into place on your line and simulate the operation. It’s a very simple but hugely effective way to design your floor and make changes. So, if a product is changed and this requires the use of a new robot, simulation engineers can see if the robot’s size will, for example, interfere with one of the conveyors. The layout engineer can then make the adjustment and issue a change request notifying purchasing that a new piece of equipment is needed. Furthermore, impact analysis can be run when changes are needed to avoid mistakes and inform any suppliers that might be affected.
Process validation: A digital twin of the production system is simulated to validate that the planned process details are correct and will work properly. Simulation models are very intelligent and can run quantitative analyses to validate a wide range of human safety factors for the global worker population.
- Process validation: With process validation, the Digital Twin can be used to digitally validate the assembly process. Intelligent modelling using quantitative analysis can assess all the human factors associated with the build to advise on issues such as working posture to help prevent employee fatigue and injury. The report can be used for training with videos and process guidelines produced for staff.
Throughput optimization: A digital twin of the planned Nanobox production system is used to statistically simulate and evaluate it for performance. In this image we see that the operators and robot can work together effectively, that all of the resources in the workcell are fully utilized, and that the target throughput can be achieved.
- Throughput optimisation: The Digital Twin can also be used to statistically simulate and assess your planned production system. It can evaluate whether to use people, robots or a combination of the two. It’s possible to simulate all workflows – even down to the how much energy is utilized by the production equipment – to look to streamline the process as much as possible. The analysis can show you how many parts will be produced by what process so you can be sure to hit schedules prior to creating the physical line.
Manufacturing execution: The digital twin is connected directly to the manufacturing operations management system to ensure it is executed properly on the shop floor. In this image electronic work instructions, linked to the manufacturing process plan, are displayed on a SIMATIC IT shop floor console for production operators.
- Manufacturing execution: You can improve manufacturing execution by using the Digital Twin to close the loop between the physical and virtual worlds. Manufacturing instructions are released direct to the design floor where operators can view them along with associated videos. Operators can feed back data from the production floor (e.g. the gap between two panel screws) while other automated systems can also collect performance data. This can be used to assess if there’s any difference between the build designs and results to isolate and rectify any issues.
Close the loop: The digital twin closes the loop between the virtual and physical worlds of manufacturing so you can compare the as-design model to its as-build result. In this image, Big Data techniques are used to collect live quality data from the shop floor, which is overlaid on the Digital Twin in Teamcenter for analysis.
A new way of doing things
Using a Digital Twin, which truly replicates a physical product, can help you to spot problems more quickly to accelerate production and reduce costs across the production chain. What’s more, it ensures that you know the design can be made; the plan is always up to date and synchronised; the strategies will work; and production will perform as you anticipated. It also helps you to see how new technologies can be integrated into your lines without the risk of buying and installing them to see how they perform. For one of the world’s most advanced industries, manufacturing has long relied on proven but dated approaches to planning its products and lines. Now’s the time to bring the spirit of innovation that drives success to the way we approach design, process planning and execution; now’s the time to try something new.
Guest Post submitted by: Siemens PLM