You have to keep in mind that creating a spreadsheet is much like a creating a software program. You wouldn't trust a program without running test data through it, and you really have to do the same with a spreadsheet.
Thanks for the study, Brent. Was this an Hypothesis-Driven study or a Data-Driven study? I'm a scientist, not an engineer, but I do use spreadsheet calculations to do quick statistics on very small collections of preliminary data. To analyze experimental results I go a bit deeper and use either MathCAD or Mathematica depending on the preference of my collaborators. For measurements and signal processing I use test-driven software development guidelines to produce code in National Instruments LabVIEW. I use a spreadsheet to document quick calculations that I used to perform with a handheld calculator.
I would hypothesize the use of spreadsheets to analyze mission-critical data would be widespread in business and financial applications. Is it that widespread in science and engineering?
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.