CJSteele Probabilistic Services
MEMBERS AREA LOGIN
It's always worth reviewing the basics. On the articles page you will find a collection of articles that you can read to keep yourself well grounded in quality and its relationship to design and risk management.
6 sigma is probably something that you have at least heard of, if you are in anyway involved with quality. Given that probabilistic methods make an ideal tool for quality improvement the question of the relationship between the two (6s and probabilistic methods) is one that is likely to be raised.
When quantifying the effects of uncertainty on a system there is a dilemma that many face: the selection of most appropriate distribution for each of the input variables.
Robustification allows us to make systems less sensitive to various sources of random variability. This is done by adjusting the nominal values of the design variables; not by restricting the actual random variability. Because this reduces the need to install more costly manufacturing equipment or put extra effort into the tight control of various operations, it is a more cost effective approach. Many not familiar with robustification find this difficult to believe (if something sounds too good to be true…).
To be able to take advantage of Robustica you must create a spreadsheet model of the system that you are investigating. For business systems, this is often fairly easy; the spreadsheet environment is where this is typically done. It is also relatively easy for many engineering, scientific and other technical system; the spreadsheet environment is familiar to many professionals in these fields. However, there are times when the system of interest can only be modeled in dedicated software (CFD, FEA and other dedicated simulations packages) or when a physical prototype must be made of the system if the operation and performance are to be investigated. In these cases, it is not immediately obvious how you can use Robustica to robustify your system’s design.