The EMS industry has posted several straight months of what some consider excessively high book-to-bill ratios. The April peak of 1.62 has only marginally fallen over the past couple months and, as of this writing, was 1.48 in June, the most recent data available.
As a refresher, the ratio is calculated by dividing the amount (in dollars) of bookings by the amount in shipments. In other words, if over a set time period a company gets $110 worth of orders and ships $100 worth of product, its book-to-bill ratio would be 1.10. A ratio over 1.0 is considered an indicator of future market growth.
So a positive ratio is a good sign, generally speaking, but too much of a good thing makes folks nervous. And ratios in the 1.40 and above range are historically at the high end.
Some are concerned of an overheated market, but conversations with several leading EMS firms suggest instead that OEMs are offering longer forecasts, which are inflating the numerator. For instance, if the typical window was six months, it might be nine or even 12 months now. That pushes more “orders” into the data pile, but it’s a mathematical anomaly, not a sign of double-booking.
I don’t expect the sky to fall, at least this time.
I read with interest Zohair Mehkri’s SMTAI 2020 paper titled“How Quantum Computing (QC) will Revolutionize Electronics Manufacturing.”I will start by saying that he gives a very good Quantum Computing 101 overview. This is no easy feat, as QC is a difficult technology to understand. I will humbly state that I still struggle to understand the basics, and I’m sure I don’t understand QCs as well as he does.
However, I have two main concerns with Zohair’s paper. One is that it may give the impression that QC is becoming a practical technology and will soon be widely available — to the point that we can use it to solve electronics manufacturing problems.
QCs are rare; there are about 30 worldwide, 15 of which are owned by IBM. Although to be fair, Shenzhen SpinQ Technology gave this recent announcement: “On 29 January 2021 Shenzhen SpinQ Technology announced that they will release the first-ever desktop quantum computer. This will be a miniaturized version of their previous quantum computer based on the same technology (nuclear magnetic resonance) and will be 2 qubit device. Applications will mostly be educational for high school and college students. The company claims SpinQ will be released to the public by the fourth quarter of 2021.”
Since the device has only two qubits, it will more than likely be for educational purposes not intended to solve real problems. It will be interesting to see how it emerges later in the year.
Almost all QCs are superconducting, meaning that they require very low temperatures to operate as cold as -460°F, which is colder than liquid helium. They are also extremely delicate; even slight vibrations causes them to fail.
So, we might be able to rent time on a useful QC sometime in the future, but QCs won’t be common any time soon.
The other concern I have is what is the need for QCs? Most of the practical problems that face us can be solved by conventional computers. In addition, only certain types of problems can be solved by QCs. As stated in Wikipedia: “However, the capacity of quantum computers to accelerate classical algorithms has rigid upper bounds, and the overwhelming majority of classical calculations cannot be accelerated by the use of quantum computers.”
QC is an exciting technology and many wonderful discoveries will no doubt come from it. However, I am skeptical that it will solve practical problems anytime soon.
It’s been awhile since I used this space to make any predictions about the coming months, but the end of the year is always the logical (if cliched) time to do so.
So here goes:
The migration of manufacturing to North America will accelerate, and the mainstream media will begin to report that OEMs are also reestablishing internal production lines.
Flextronics will buy at least some of RIM.
Robots as substitutes for human labor will be heavily hyped but lightly used.
Ousted Altium founder Nick Martin will hook on with a budding cloud-based software company and build a PCB CAD tool.
At least two new PCB CAD vendors will emerge.
Electronics manufacturing companies will end 2013 with less cash in the bank but brighter prospects for the future.
