Option Value and Test Strategy
Larry Desjardin- March 26, 2012
Do you know what “option value” is? It comes in many forms, but the essentials are that there is value in having a choice at a future date. Take stock options as an example. Here, a company purchases its own stock at today’s price, and offers it to an employee at that price for the next several years. Question: Without the stock option, the employee could have purchased the stock themselves on that same day at that same price, so why is there value to the stock option? Answer: Because we live in an uncertain world. The value may go up, or it may go down (or stay the same). But by receiving an option the employee can wait and see. They can eliminate purchasing the stock in any scenario where the stock doesn’t appreciate. There is added value in having this choice. It is certainly true that in a fully predictable universe the stock option has no more value than just purchasing the stock outright, but we don’t live in that universe.
Option value is not just related to finances. Imagine a rush hour commute in a large metropolitan area. Would you rather have your commute pre-programmed ahead of time for every highway and street, or would you rather choose your path at the time, perhaps while listening to traffic updates on the radio? There is option value in the latter because we don’t live in a predictable world. And test strategists certainly don’t either.
Imagine setting test strategy for the production floor of an electronic manufacturer. You need to choose the number and type of each of your test stations, since each station type is optimized for a different product. Do you know exactly the order ramp of a newly introduced product? Of course not. Marketing’s forecast may be using the best analytic methods they have, but they can’t be precise. And what about the real time product mix? Even less so. OK, how about big orders from key customers and when they will occur? Not that either. How about the date when a piece of equipment will fail and need to be replaced? Nope. And remember that corporate decision to consolidate manufacturing? You didn’t get that memo well in advance?
Stuff happens. You simply can’t predict a commercial product mix exactly, and sometimes not even close. But you can set strategies that maximize your flexibility, and therefore give you option value.
I recently discussed test flexibility with Jack Trautman. Jack was Vice President and General Manager of the Manufacturing Test Division of Hewlett Packard in the mid-90s. Jack told me the story of the HP 3070 Board Test System. Soon after its launch, HP added a “Pay For Use” feature where fully loaded HP 3070 systems were deployed at a low price, but with software restrictions that limited the pin count and functionality essentially to a system you would expect at that price (perhaps at a modest premium). However, the customer had the option to “pay for use” to expand the pin count and functionality when needed. It was the ultimate flexibility. The concept was a tremendous hit and contract manufacturers adopted the system in droves. The reason is obvious. Contract manufacturers have the least influence and warning about the product mix, but must deliver results in a very competitive environment.
The HP 3070 depended heavily on pin electronics, being chiefly an In Circuit Tester (ICT). You can imagine how ICT is adaptable to this type of option strategy. But what if you’re deploying internal functional testers? Here, you must be vigilant in adopting strategies and products that will give you maximum flexibility.
One obvious strategy is to create a common core for all testers. This allows quicker redeployment of test assets when the mix changes. A common controller and operator interface is a start. Perhaps standardizing on an intermediate fixture interface that connects to the DUT (Device Under Test).
It won’t be much of a surprise that I believe modular instrumentation can play a big role here. Vendors invented modular products because it gave their customers (and themselves) options to mix and match and to scale channel counts. That’s a lot of option value right there. Open system modular systems (such as VXI, PXI, or AXIe) expanded this concept to include products from multiple vendors. This brings flexibility, and therefore option value, to a whole new level. Not just for the choices of when you first create a system- but for reconfiguring systems and redeploying assets later.
But even within the open system modular domain, it behooves a test system strategist to think through which products maximize the flexibility that modular can bring. Here’s one example. Two weeks ago I wrote about ADLINK’s latest PXI introduction, and their new 9-slot PXI mainframe that maximized the number of hybrid slots. Hybrid slots allow a user to deploy either an older PXI module or a newer PXIe module into any given slot. As the core of a test system, PXI chassis that maximize hybrids slots offer phenomenal option value because you can deploy or redeploy any number of PXI or PXIe modules into the slots of your choosing. Think about it- if you don’t even know your product mix in the future, how could you possibly know what type of module will be located in which slot if you choose to redeploy assets? You can’t, that’s why you need options.
There’s many more examples, but I think you get the picture. It’s all about maximizing flexibility for an unknown future. So, when you choose test strategies, test architectures, and the individual products, spend some time asking this: which solutions also maximize my option value?