Using Onshape CAD models for robotic assembly of mass customizable products
Mass customization combines the efficiency of mass production with the ability to customize individual products according to customer specifications. This has been an ambition of manufacturers for many years, delivering the ability to produce bespoke products for the same cost-per-unit as mass production lines.
However, one of the biggest challenges has been incorporating the benefits of robotic assembly to maximize efficiency. In the latest PTC Talk, Peter Harman, Founder and Chief Technology Officer of CAE Tech, spoke about using PTC Onshape models for the robotic assembly of mass-customizable products.
What are the benefits of mass customization?
While assembly lines are still common, today's state of the art for robotics is shifting from large, stationary devices to flexible machines with self-learning capabilities that help optimize processes over time. This move is helping to fuel a much higher degree of customization in production.
Previously limited to software and services, digital manufacturing techniques have made mass customization viable for physical products as well. Before digital manufacturing solutions, mass manufacturing and customization were considered incompatible. Bespoke products were usually produced on a small scale at a far higher cost. Digital manufacturing systems have enabled manufacturers to mimic the approaches seen in custom-made business manufacturing to target mass consumer groups' wants and needs rather than individual clients. Changes in products or services can be as small as a variation in color or as complex as an entirely new product.
More value is delivered to the customer at negligible cost to the producer. Mass customization offers significant opportunities for differentiation, innovation, and disruption. Used creatively, it can be used to substantial competitive advantage, offering ever greater levels of personalization, increasingly distinct from similar products on the market. The infrastructure required for mass customization also makes manufacturers markedly more agile in their operations and better able to respond to externalities, as well as trends in consumption, design, and technology.
Why should manufacturers adopt a mass customization strategy?
When Harman set up CAE Tech, the mission was to bring the simulation he had used in previous sectors, such as Formula One, into the manufacturing industry. Over the last few years, the focus has been on the challenges of mass customization. When any product is being manufactured, the focus is to minimize the use of resources and materials, which feeds directly into the current drive of net zero. Mass customization gives companies the ability to only make the things that are needed in the market, eliminating waste. Another trend within manufacturing is to reduce supply chain vulnerability, and that is lessened the closer you manufacture to the end market, again facilitated by mass customization.
However, we are also in a world where end consumers expect choice; this is something that they have become accustomed to. Because they desire choice and value it, they are often willing to pay more for it. In his book Transition Point, Sean Kewley looks at all the changes we are going through in the latest industrial revolution. One of the critical things he stresses is the idea that future supply chains will be personalized, automated, and local. Products will be made for people as they want, where they want them. The mission of CAE Tech is to develop the technologies that allow us to deliver that.
There is a range of technologies that can facilitate mass customization, one of which is configurators. These configurators allow an end customer, reseller, or someone who is not a design and engineering expert to configure and set up a product ready for manufacturing. A crucial element of this technology is following the data from the actual CAD data to how that product is being made.
Why is robotic assembly not currently used in mass customization?
When he talks to customers about mass customization, the question that Harman often asks is, where are the robots? Many products are being manufactured using robotic assembly, but that is not something that he has ever come across in terms of mass customization. That was the challenge he set for his organization: could they assemble a customizable product using a standard industrial robot? What would that involve?
The challenge here is that if you want to change the program for a robotic cell to assemble a product, you would need to plan the assembly sequences for that product regarding the waypoints, where it opens and closes an end effector, or use additional functions. Then, you need to install that onto the robot controller for execution as part of the manufacturing process; this would require expert engineers. If the robot program must be changed for each product configuration, that will not work.
What CAE Tech set about developing was to define the set of robot actions that are going to take place to build a product but accept the fact that those are parametric. The positions of things might be different, and the number of times certain operations are performed might be different. One of the operations might be to drill a hole multiple times in different places. The solution was to define what that sequence of actions is and then for the waypoints to consider those as being positions within 3D CAD. It was then possible to react to where things have moved to and generate the robot program from those changes in the CAD. If this was connected directly with the robot, the new robot program could be actioned.
How can Onshape help produce programs for robotic assembly?
That was all good, but it needed to be tested in a live environment. CAE Tech took this problem to the UK's Manufacturing Technology Centre (MTC). This MTC is one of the UK technology catapults and is a government-funded, part-industry-funded organization to bring technology to bridge the gap between academia and industry. They came up with the idea of a system of plastic pipes, which might need to be assembled in different configurations. The MTC provided CAE Tech with an ABB robotic cell to prove this case.
CAE Tech used Onshape to make models of every variation of the modular pipes and combined this with a CAD model of the robot. This robot model featured all the joints to allow it to be moved around within the CAD model. All those CAD models were combined using Crusoe Technologies in India, a QA partner of PTC, and crucial to this was the Make Connector functionality that comes with Onshape. Mate connectors were used to define where those parts were going to join and to define where the robot gripper was going to grip them. The table in the robot cell is configurable to use parameters within the CAD to move components around to different places. Then, everything was combined using another mate connector to define what the coordinate system is, as far as the robot is concerned.
One part of the software CAE Tech did not build is how to get that robot program onto the ABB controller because of the inherent risk of that operation. Instead, MTC developed tooling to push new robot programs into the ABB controller. In this case, the CAE Tech software generates ABB Rapid, their robot programming language. This ensured the simulation and running of that program were all done using ABB's software, eliminating any risks of driving robots through tables. Within just two weeks, CAE Tech proved that the robot could assemble the system of pipes correctly, entirely defined by parameters and CAD.
To learn more about how robotic assembly can support mass customization in greater detail, watch the full Talk here.