A radical shift in manufacturing
In 2012, the Economist brought a special report, hailing 3D printing as the steam-engine of the Third Industrial Revolution. While 3D printing is over 40 years old, only in the past decade has the real potential of 3D printing become clear [ECO, WEF]. With Adidas offering personalized 3D printed shoes, Airbus printing lightweight structures to reduce fuel consumption, Volkswagen printing tooling in-house reducing cost and time of production and Mercedes Benz providing spare-parts on-demand, reducing the need for shipping and invested capital in stock.
The main reason is that the ecosystem is maturing to a point of valuable market applications. It is maturing with a transition from laboratory to fabrication (Lab to Fab). As costs are decreasing and performance increases, 3D printing applications are shifting, from 1) High-margin, low-mix and low-volume sectors, such as jewellery and Formula One racing, to 2) Mid-margin, high-mix and high-volume applications.
The maturity and shift are supported by the M&A activity; # of transactions, € per transaction, switch from 3D printing companies investing to non-traditional players investing; e.g. GE, Autodesk, Siemens and BMW aggressively entering the market (Figure 1). As the performance of 3D printing technologies increases and cost-prices lower, new applications are becoming possible. This supports the shift from;
- Mass production in a traditional supply-driven chain (push) towards,
- Mass customization in a demand-driven chain (pull)
Mass production to mass customization
A traditional supply-chain is driven by the available resources (material, design, factory). Each chain-element (material, design etc) operates in silo’s. Connecting the silo’s and adding intelligence would enable data-driven optimization. As the demand-chain is driven by the use-need and application. This is identified as the main need in the 3D printing industry [TAND]. The result: A radically more efficient and effective way we use and produce products.
3D printing enables mass customization. This is a true paradigm shift. Only now, the maturity of the ecosystem is starting to enable mass customization (see “3D printing”). This maturation enables a new way we use and produce products.
The economic and ecologic benefits of customization
- B2B: Exclusive markets have long used 3D printing (nuclear, aerospace, defense). Due to the shift, commodity industries such as logistics (UPS/SAP), tier-supplier (Jabil) and technical wholesalers (ERIKS) are starting to enter the market. A good example is the recent configurator developed by Trinckle 3D enables the customization of copper inductors by Protiq [PROTIQ].
- B2B2C: Lowe's is introducing in-store and online 3D printing and scanning services that will allow customers to reproduce rare replacement parts or design unique products instead of purchasing off-the-shelf units [LOWE]. Eyewear is increasingly becoming 3D printed and may well follow hearing aids of which > 99% are 3D printed [YUN].
- B2C: It is estimated that between 2040 and 2060, 50% of all products will be 3D printed [ING]. With the cost/performance improving such that the long-tail is starting to see the first 3D printing applications. A great example is the reshoring by Adidas of their speed-factory and the recent launch of the 3D printed Adidas; Futurecraft 4D. The time-to-market was reduced from 18 to 3 months and no longer is there a need for warehouses and transportation. By 2021, Gartner predicts that 20% of the world’s top 100 consumer goods companies will use 3D printing to create custom product. Looking at 2016’s numbers, this would result in a market value in excess of €5 billion dollars, for just 20 companies. With $200 billion in spending power, the millennial market segment is the Holy Grail for brands [RCE]. Millennials expect not only immersive and interactive, customer journeys but also fun, one-of-a-kind experiences. In 2025, Cognizant foresees retailers using AI to replicate personalized assistance in a self-service model. Next-generation tools will enable smartphone-wielding customers to interact with virtual experts [COG].
So, what is holding this back?
Mass customization is held back by two things, identifying the right opportunity and subsequently capturing that opportunity. Seems simple, right? However, the challenge lies with involving the end-users. They are the ones that are able to spot the opportunity and suddenly need to be involved with product-design, a field with which they have no experience.
If you take a look at the 3D CAD software that is currently available, the solutions fall into three categories based on the user of the software:
- Professional - Generic, desktop solutions for the 3D experts, product-designers and engineers
- Prosumer - Less specialized software for the demanding, non-technical hobbyist
- End-user - Application-specific software, to enable the co-creation with end-users
The software further differentiates across the degree of freedom, which the user requires. Engineers in demanding industries - such as the automotive and aerospace industry - are at the end of both scales, being highly competent (hopefully) and demanding high precision. Craftspeople, 'makers' and artists on the other end of the axis.
Looking at the various UI designs for the various packages, the more successful solutions are at either end of the spectrum. CATIA and MAYA for aerospace and special effects, and Tinkercad and gravity sketch for the artists and hobbyists. There seems to be little in between that sticks.
CAD software for experts is dramatically difficult to grasp for anyone not already trained in using 3D software and with good reason. The concepts that are used in the domain of these workfields are intrinsically difficult because they try to capture reality into a computational domain. This is something so fundamentally non-intuitive that it stands to reason that the user experience is highly challenging. Objects like NURBS (Non-Uniform Rational B-Splines) or Voxel Spaces and Subdivision surfaces are commonplace in the high-end high-precision workfield and require extensive training to use and predict. Just like operating any advanced and sophisticated machinery requires training and even certification of the operator.
For the non-technical user however things are not so open to exploration. Even software that is meant to be intuitive can still be quite challenging. Concepts like extrusion and revolution of a curve are commonplace, but require the user to think geometrically. Autodesk's TinkerCAD comes closest to providing truly intuitive concepts like 'making holes' instead of 'subtracting shapes' but still uses the Cartesian plane with its degrees of rotation and scale factors. That is not something you think of when you are building a house of cards or playing with clay. Input devices such as the mouse or touchscreen are interaction funnels that all but filter out our proprioception and physical intuition. Many people are very good at imagining new 3D products and shapes, but use muscle memory and physical intuition rather than fundamentally mathematical concepts. In other words, in most software, -you- are not there creating shapes, your mind is. And although the mind can be trained to mediate creative ideas, it is secondary to how we imagine making things.
There are two developments that may eventually bring us closer to leveraging the power of computational design in a physical, truly intuitive way. One is Virtual and Augmented, the other is Artificial Intelligence.
Borges for Tooling, Jigs & Fixtures
Tooling, jigs & fixtures are an excellent application of 3D printing. Volkswagen Autoeurope has recognized this opportunity and started collaboration with Ultimaker in 2014, 3D printing their tooling in-house. This has led to the following benefits;
- 93% manufactured tools, switched to in-house production
- 91% of cost were saved
- 95% reduction in development time the projected
- 2017 savings of €250,000 had already been surpassed (Sept 2017) and are now expected to amount to €325,000
The project between Ultimaker and Volkswagen Autoeurope changes the classic process towards an Additive process. The classic process for the creation of toolings is shown in Figure X. Once an opportunity is identified, the tooling- and process planner define the requirements, then together with internal or external tooling facilities the process is selected, finally the design is created with the design department. With Additive Manufacturing the process is similar, with three differences;
- Part integration: Due to the design freedom of 3D printing, parts can be integrated in an assembly (Additive), whereas for conventional manufacturing, this often requires multiple production- and assembly steps,
- Internal tooling: 3D printing can be easily used in-house for the rapid, cost-effective production in-house (see introduction)
- AM Expert: An expert on Additive Manufacturing needs to be involved, to define the requirements, select the right process and support with the design
The Borges approach enables end-users to directly create their design. This removes the need for an AM-expert in the process.
Figure X shows the two-step approach towards the implementation of this solution. In a first step a CAD Configurator could be implemented based on a parametrization of existing jigs and fixtures.
Based on existing application the jig and fixture library is used for the creating of easy to manipulate models.
In a following step, workers in the production plants are enabled to realize their own tooling needs via a web interface. Including the information from step 1, this web interface offers various simple design functionalities to manipulate the jigs and fixtures to fit the worker’s requirements. Through the use of intuitive user interfaces and the geometric engine that has been developed, end-users could directly generate their own designs.
Figure X shows the potential cost reduction for the implementation the Borges approach. The initial cost reduction of 90% was achieved and validated by Ultimaker and Volkswagen Autoeurope. The implementation of Borges could further reduce the cost by 70%. Finally, by enabling end-users to co-create their designs, the identification of opportunities increases significantly leading to further optimization of processes and a reduction of cost.