Local Warming
By Michael Ford, Aegis Software
A couple of years ago, at a local SMTA chapter meeting in the US, a relatively small EMS provider stated that their business model was constrained and destined to stay small due to their location, often doing only the early prototype runs before the start of volume manufacturing, which they assumed would be performed over a border or two somewhere. Global events and the mood of society, for a variety of reasons, has changed. Even larger companies are now warming to the idea of local production at some kind of volume; but what needs to be in place to make this happen successfully, and without risk?
The so-called “fixed costs” have always been the key measure of a manufacturing business viability. The cost of labor is a large part of that, creating confidence in business performance where the costs of labor are lower than average. Within these areas, there is a lot of competition, as we see continued movement of manufacturing services from one place to another, as competitive rates are exploited. The fact that large manufacturing sites can be created and moved, illustrates the sheer volume of business that is performed, given that margins are supposed to be so small. Consumption of electronics devices and assemblies has never been so high as it is today.
A Real Turning Point
The reality of the situation is that the industry is now at a real turning point. The introduction of technology has started to “level the playing field”. On the hardware side, machines and robotics are now able to perform more and more tasks that were once manual, and do them more flexibly. Designers have been pushed to create ever smaller products, using smaller components in higher densities, such that humans are no longer the target for any assembly work at all. In electronics, the role of the human production operator, while never likely to be completely replaced, will change and evolve, becoming a very flexible mobile or cell-based resource, rather than being part of a production line, aided by technologies such as simple Augmented Reality (AR), which helps guide operators between many varied jobs throughout the factory, providing the opportunity for them to be fully utilized.
On the software side, the situation is similar. The real driver for adoption of software automation on a factory scale to catch up with that of assembly hardware, is the link between hardware and software, which recently has become seamless, in no small way attributable to IPC’s Connected Factory Exchange standard (CFX), which has broken down all of the barriers related to technical issues, cost and risk that hindered the active connection of IIoT MES-based digital twins to their physical counterpart. This bonding of software and hardware across all assembly operations reveals critical information, never before quantified. For many years, the performance of manufacturing has been reported using different “flavors” of data, meaning that different elements were included, not included, or categorized in a way that brought “meaning” to the data from a certain perspective. It is all very understandable, as people responsible for running each production line for example, do not want to be measured on things that are outside of their control, such as lack of orders, or a poorly optimized schedule. Everyone therefore has a claim to a different type of metric.
With the availability of so much data from assembly processes, all feeding into the IIoT-based MES system to gain context and then in turn being made available into a data warehouse for analytics processing, the way Enterprise reporting is now approached is revolutionary as compared to previous reporting options. The question then arises as to whether and how the many different perspectives on data presentation will be retained, or, will the factory performance as a whole, now that it is completely visible, become the dominant driver when business decisions are made. Of course, the answer is the latter. Whether production sites are local or distributed globally, profit and loss are the key drivers. Business intelligence, based on these new data connections is about to reveal what we already know, but seldom have had the confidence to talk about. For an EMS company, looking at the holistic model of a factory, it will include what were previously potentially unreported considerations of, for example, customer line setup times; product line changeovers; run rates that do not match customer delivery schedules; the cost of buffer stock; the time lost due to needless maintenance or breakdowns; time waiting for materials; time to set up machines and processes; time to find people, resources and tools to be ready when needed, as well as the big one, the cost of poor quality.
When factoring in all of these things, as site-based analytics do, it shows that the relative cost of labor between regions, especially as automation grows, becomes ever more inconsequential. Many sites openly admit productivity levels on a holistic basis of only 10% – 20%, based on the high product mix that they have to support. Being smarter about the way we manufacture has become more important than where we manufacture. The balance tips then in favor of local manufacturing becoming at least a significant step forward in terms of viability for volume production, being local to customers, and offering greater flexibility with lower finished goods stock.
There are several cases today where local EMS companies in the US and in Europe, are securing new business, featuring differentiated flexibility and capability, based on the use of the latest machines and software technologies. Investment is being made at least as confidently as those fighting amongst each other in remote locations. The interesting thing is, you don’t need to build a factory and accumulate years of experience in order to know what is viable, and what is not. The use of analytics and Business Intelligence is not only aimed at making cool graphs and charts to impress your boss, it is there to enable you to fully understand the current holistic manufacturing business model, based on complete, detailed, accurate and timely data from live production; see the opportunities to improve, then apply the exact same logic to alternate scenarios; comparing like for like between local and global alternatives; understanding what is needed, to become very confident about finding the next location for that volume production. The cost and risk to do this, as compared to the investment in a new factory that may or may not make money, is trivial. What is needed is just the IIoT-based MES system and an off the shelf Business Intelligence suite. The cost of both of these is fundamentally negative when considering the total cost of ownership, as the benefits experienced from each are most likely to provide ROI within months, not years, from normal operational usage. It takes you then, to a position where the manufacturing business strategy can move to the next level of certainty, something I am sure that we all value in today’s world.
