Envision manufacturing as a left-to-right process with product design on the left and production on the right. Today, the balance of power — and the seat of success — is on the right.
Low production costs win manufacturing jobs. Almost all of the hundreds of thousands of American manufacturing jobs that have disappeared over the last 20 years have moved to countries with cheap labor. From clothing to consumer goods and now even military hardware, low-cost overseas producers are undercutting American manufacturers and forcing them out of business at an alarming rate.
American manufacturers can’t expect to compete on labor costs, but labor isn’t the only factor in the equation. Domestic manufacturing can level the playing field with low-cost producers by moving to the left. Not on the political spectrum, but on the product design, development, and manufacturing processes.
Moving to the left means completely integrating the product design and production processes utilizing well-established technologies such as computer aided design (CAD), modeling and simulation, and product life cycle management into a new collaboration framework. This new product development environment will take enough cost out of product development and enough waste out of manufacturing to even the playing field with low-cost producers. It will also accelerate the pace of innovation so sharply that low-cost producers can’t catch up.
Four other factors have emerged to enable this transformation of product design. They are “big” data collection, intelligent machines, high-capacity computing infrastructures, and widely accessible data analysis tools. They all converge in the federal government’s recently established Digital Manufacturing and Design Innovation Institute (DMDII), which was established by the Obama administration in early 2014 as the second institute of the National Network for Manufacturing Innovation.
The leftward drift
The term digital thread describes the data interchange that will connect design engineering, production engineering, marketing, sales and customer service. It will create a single environment encompassing every function of product development, from idea conception to the customer.
The three “fibers” that make up the digital thread are big data, analysis tools, and intelligent machines. The data encompasses all information relevant to product development: material properties, CAD models, production data, etc. It comes from engineers and material manufacturers, and also from the machines in the manufacturing process, from the design studio to the production floor to the shipping dock. The smart machines will collect data and store it in standard formats that make it usable to everyone in the product development process.
The analysis tools make the big data sets manageable and useful. Modeling and simulation applications, for example, enable engineers to combine material data and CAD models into realistic simulation models that can accurately predict product performance. The final design models from this process are used to optimize product design in addition to optimizing the process by which the product is produced. This process, often referred to as Design for Manufacturability, is forever changing the traditional manufacturing paradigm.
To illustrate how much cost-reduction potential there is at the front end of product development, consider the process which commercial aircraft manufacturers and their suppliers must endure in order to get new components made from lightweight (and fuel) saving composite materials qualified for use on the fuselage and wings.
FAA certification of these components is mandatory before it’s incorporated into an aircraft design. To earn FAA certification, OEMs and their suppliers must subject the material being used and the component itself to hundreds, if not thousands of tests.
To complete those tests in the current product development cycle requires building dozens of physical prototypes called coupons, then destroying them under various conditions to determine how they perform. This process can take many months, and more often than not, many years and can cost millions of dollars.
In the digital thread environment, more than 70 percent of that testing can occur in simulation applications in a fraction of the time and at a fraction of the cost of physical testing. The digital thread will connect the design engineers’ CAE systems with material producers’ data over high-capacity network links.
Combining a part’s geometry and material properties into the same CAD model gives a simulation application all of the data it needs to predict the part’s behavior as a physical object. Manufacturers can also validate their part’s performance by plugging their digital models into the OEM’s models and simulating their behavior in assemblies and systems. Some physical testing will be necessary to verify simulation results, but the overwhelming majority of it has been removed from the process at enormous savings.
The simulation environment extends from product design into production design. Using the digital thread to connect them enables production engineers to plan tooling, production lines and processes while the design is evolving. It also enables them to warn design engineers if a product is going to be particularly difficult and expensive to manufacture.
While the product is being designed and manufacturing planned, marketing and sales can advise engineering on customer needs and market realities that might influence designs. Engineers can modify products during every phase of design at low cost and low risk of failure because they’re working on digital models instead of relying on physical prototypes. Every change can be automatically promulgated through the manufacturing process — changing the settings on a smart machine, for example. This reduces trial-and-error waste.
The same simulation technology that moves a product through design and manufacturing at exponentially lower costs also enables engineers to innovate freely without fear of pushing up costs or producing failures. Changing digital models then simulating how the product will behave after it is manufactured carries very low costs and very low risks. Engineers can continuously improve their products during production runs at a pace that low-cost manufacturers can’t match.
Access for all
A critical piece to making the digital thread into a reality for small and medium-sized manufacturers is affordable access. Companies lower down on the supply chain don’t have the technology budgets of larger suppliers and OEMs, and also don’t have the same needs.
A small company might only need simulation capabilities when testing a new design or material. Forcing them to invest in expensive software and training would wipe out the savings they can realize at the beginning of product development. That’s why a major thrust of the government’s DMDII initiative is making digital thread resources accessible to small and medium-sized manufacturers.
Accessibility is easier now than it has ever been because data networks have evolved enough to run demanding, data-intensive applications over WAN and Internet connections. The links themselves are more robust, and the computing power in the cloud frees small companies from the cost of high-powered servers that they probably only need once or twice a month. Putting digital thread applications in the cloud would enable small companies to access them on a pay-per-use basis, or some other scheme that offers ability and cost-effectiveness.
Accessibility has another dimension in the digital thread discussion. The military supply chain was cited as the major impetus for the DMDI initiative, but the digital thread is as much about the civilian economy as the military. The same product development and manufacturing dynamics that affect the military affect civilian industries, so the digital thread has the same potential to transform its cost structure.
As long as low labor costs are the foundation of low-cost manufacturing, countries that provide high wages and benefits will always be at a disadvantage competing against those that don’t. Winning the game means a move to the left on the manufacturing spectrum — toward agile, efficient design bound by a digital thread.