Following a directive from the US military in the early 1990s, the defense industry made a shift from using custom embedded electronic components made to military specifications to commercial-off-the-shelf (COTS) components. Since the overall share of the DoD as a consumer was expected to shrink over time, this move to reduce costs took a practical […]
Saying that something is “good enough for government work” is often meant as a joke and the reference implies “mediocre work.” The irony is that “government work” is often highly sophisticated; systems are designed and engineered to operate in the most extreme environmental conditions for a very long period of time.
I recently had the pleasure of having lunch with a talented component engineer who has spent much of his career working in the defense industry. During the course of our discussion I learned that some aviation systems need ICs to operate in temperature extremes ranging from -55°C to 125°C; ground units often travel in harsh environmental conditions (e.g. fighting extreme heat and sand storms in deserts) while being exposed to hostile attacks; satellites traveling through orbit are exposed to protons and heavy ions from solar flares, yet must operate reliably in space.
In general, defense sustainment and counterfeit avoidance has been left to DMSMS teams and logistics or engineering tactics. However, so far the solution has primarily been to develop standards, authentication and anti-counterfeit technologies. These responses have been critical, but have largely remained reactive and have not produced the dynamic collaboration crucial to maintaining a healthy, proactive supply chain. Instead, each player is left facing inward — focusing on solutions from their own particular positions in the supply chain — but without the resources to truly be proactive.
In the past we’ve talked about the challenges of Last-time Buy and overstock. In Dr. Sandborn’s CALCE Obsolescence Management training, this question illustrates the challenges and risks in regards to what customers can face, at the time of EOL. The answer might be easy if you were looking at a “bridge buy”, where you only need enough to get you to the point of a planned upgrade. If I had to only buy shoes to get me through five years it would be challenging but I could probably come up with a pretty good estimate based on the last five years of my life.
How End-of-Life sometimes feels…
Ask anyone who drives an older car. As the system ages, it develops its own quirks. You have to jiggle the shifter in park to get the keys out of the ignition. You have to pump the gas twice before it starts up on a cold day. The AC has to be turned off when going up a steep hill on days over 96 degrees. A particular brand of brakes work better when driving in California, as opposed to Montana. You know that you can get away with just 2/3rds of the thread on the bolts, but only for 6 weeks.
In short, you know that system inside and out: all the bugs, the features, and quirks that impact operation and repair.
Unless your mechanic is into vintage cars, though, he’s not going to relish working on an older vehicle. A younger mechanic may tell you the car is “old” and you’re better off just getting a new one – just when the old one was about to become a “classic.”
They both get harder to maintain as they get older, and if you don’t plan for obsolescence, they can both fail.
It’s common sense. As things get older, they become more expensive to maintain. For example, an antique car was state-of-the art when it first came out. It performed beautifully, and the parts were easy to find. If it had any real problems, it could be taken into the dealer for repairs. However, now that the car is a classic, it requires a lot more upkeep. In the past, it only needed to be taken in for oil changes and tune ups. Now it needs a new transmission, replacement brakes, a new timing belt and a new radiator… and as time passes, the mechanic can’t even get the parts he needs to fix it.
As the components become harder to find, the odds that your car can even feasibly be repaired get more remote. At first, you might scour junkyards and advertise online, looking for those crucial pieces of equipment, but eventually you will probably end up having to find someone who can reverse engineer or custom build the needed parts for you. And now a part that may have been $300 new is going to cost you hundreds more — if not thousands.
When I first began my work with GDCA one of the questions I had was “Why is dealing with obsolete components not just about making more parts?”
As I have come to learn, unfortunately, obsolescence management is not just as simple as “making more parts.”
Imagine you manufacture various components. In the 1960s, the computers you were making parts for were relatively simple, without many customers who could even afford computers; quantities were low, the manufacturing was relatively easy, and products generally lasted longer.
Let’s jump forward to today. Over time, and as technology has evolved (Moore’s Law), your fabrication company’s production has also evolved. Now with each product line, you are cranking out hundreds of thousands of parts each day. Customers who need 50 parts are not happy to hear of a 5000 part minimum order quantity (MOQ). And besides, to some the manufacturers even a 5000 MOQ on an older part can be a distraction.
On one of the blogs I read, someone commented: “If you’re concerned about counterfeits in obsolete components… don’t worry about to-be-discontinued components — just design them in, and buy what you need to support the product anyway. Then you won’t have to worry about counterfeits.” On the surface and if you are only worried about […]
Is there a downside to new technology innovation? We all love and encourage innovation, but what is the hidden cost?
Critical embedded applications in the Defense and Medical industry are a great example of where this question comes into play. Both these applications have people’s lives relying on them, and both require extended life cycles due to critical verification and certification requirements.
If an OEM experiences sharp drop in demand for a particular embedded board, it doesn’t make any business sense to continue building more, and the board will likely become obsolete. Everyone understands that an OEM can’t remain competitive if they have to support every product they’ve ever developed… forever. But if that board is still being used in the defense or medical industry, suddenly the systems engineer is faced with diminishing manufacturing sources and material shortages (DMSMS) and higher risk of exposure to counterfeits if obsolete components must now be sourced.