![port-of-baltimore-iStock-1299521189.jpg Cranes and shipping containers along the water at the Port of Baltimore.](https://www.supplychainbrain.com/ext/resources/2024/04/08/iStock-1299521189.jpg?height=100&t=1723060077&width=150)
Visit Our Sponsors |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
There’s a severe shortage of semiconductors at the moment, caused by a combination of factors, including the coronavirus pandemic and unexpectedly high demand for autos and high-tech consumer products. In this conversation with SupplyChainBrain Editor-in-Chief Bob Bowman, Richard Barnett, Chief Marketing Officer with Supplyframe, delves into the impact of the shortfall on electronics supply chains, and what kind of solutions might alleviate the problem.
SCB: What is the current state of semiconductor availability?
Barnett: Highly constrained, and will be for quite a while. At least over the next nine months, and it may extend into next year.
SCB: What are the factors that led up to it?
Barnett: It’s almost like a perfect storm, with a few factors landing at the same time. One has been the overall growth in consumption of semiconductors, integrated circuits and sensors, and the new hardware applications that require them. The growth is being driven by multiple downstream industries, not just high tech but also automotive, medical devices, industrial equipment and smart building applications. It's pervasive because there’s an increasing trend toward digitizing products and services, including sensors on things that might traditionally have been built with metal fabrication. Packaging for single use is becoming a digital service, with sensors.
The second key factor is at the foundry level, upstream from the semiconductor companies. You're seeing increasing concentration, with reliance on just a handful of key foundries — TSMC, UMC, and GlobalFoundries in particular. It's a highly cyclical pattern, where they go through major outlays of investment. We're currently in a growth trajectory, with new capital expenditures rapidly increasing. And those expenditures are helping to develop new fab capacity. That was tracking when COVID-19 hit, causing disruptions to many supply chains, and specifically high tech. In addition, you had a change in demand mix, due to increasing consumption of consumer electronics and remote learning systems by people working from home. Digital services like Zoom and Microsoft Teams suddenly saw a huge spike in usage that was completely unplanned for.
SCB: How did that affect the automotive industry?
Barnett: The major automotive OEMs [original equipment manufacturers] immediately saw an impact on passenger and light truck sales in almost every major market globally. There was an interesting domino effect, where the automotive OEMs started decommitting to their forecasts for vehicle production, then rolling that decommit into their tier-one suppliers, which had been holding the liability and long-term agreements with key semiconductor suppliers. So the foundries and semiconductor companies quickly began reallocating available supply to consumer electronics sellers of games, infotainment systems, mobility services and the like. And automotive manufacturers, which had created a just-in-time delivery system for parts to the plant, saw a huge increase in lead time. Then demand for autos came back, to 80% of 2019, and that caused a major supply constraint for ICs and semiconductor components across the board. Foundries couldn’t quickly absorb the new demand requirements, and that started impacting production schedule commitments for most of the automotive OEMs. We’ve seen cyclicality and supply constraints here and there over the last 20 years, but what we're seeing right now eclipses any of those past events.
SCB: Then the snowstorms in Texas made things even worse.
Barnett: Yes. Obviously, weather is often an unplannable event, and major semiconductor suppliers in Texas were given extremely limited lead time to respond — they lost power and had two- to three-hours’ notice, which didn't allow them to do a soft ramp down. They had to do a hot stop, basically. And when you do that in a fab process, it has a really severe impact. These fabs are meant to run at very high utilization, 95% plus, because they're multi-billion-dollar investments. The whole goal is to maximize asset utilization from a profitability perspective. But many of the end-process fab and wafers starts had to be scrapped or produce a very low yield. And that impacted what was in process for the next two or three weeks. The issue was just compounded by what was already a super-challenging supply-market dynamic that we were already in the middle of.
SCB: What solutions is the industry considering that might prevent this situation from happening again?
Barnett: There's a short-term set of moves that’s very reactive, to look at alternative components that might be available but weren't in the original design for some of these subsystems. They’re also thinking about using different components on platforms — to swap them out or change the mix of the vehicles that are being produced to allow for improved production. For medium- to longer-term solutions, you're seeing a lot of interesting discussions with the Biden Administration regarding its executive order for a 100-day review of critical supply chains in the U.S. — for example, reviewing economic incentives for increased stock production. But none of those proposed policy changes will have any impact on the 18-month horizon of what we're dealing with right now, with regard to available capacity of semiconductors.
RELATED CONTENT
RELATED VIDEOS
Timely, incisive articles delivered directly to your inbox.