Reliability Growth: Enhancing Defense System Reliability

This report (pdf) from the National academies on reliability growth is interesting. There's a lot of good stuff on design for reliability, physics of failure, highly accelerated life testing, accelerated life testing and reliability growth modeling. Especially useful is the discussion about the suitability of assumptions underlying some of the different reliability growth models.

The authors provide a thorough critique of MIL-HDBK-217, Reliability Prediction of Electronic Equipment, in Appendix D, which is probably worth the price of admission by itself. If you're concerned with product reliability you should read this report (lots of good pointers to the lit).

Guidelines for Planning and Evidence for Assessing a Well-Designed Experiment

This paper is full of great guidance for planning a campaign of experimentation, or assessing the sufficiency of a plan that already exists. The authors break up the effort into four phases:

1. Plan a Series of Experiments to Accelerate Discovery
2. 1. Design Alternatives to Span the Factor Space
   2. Decide on a Design Strategy to Control the Risk of Wrong Conclusions
3. Execute the Test
4. Analyze the Experimental Design

They give a handy checklist for each phase (reproduced below). The checklists are comprehensive (significantly more than my little list of questions) and I think they stand-alone, but the whole paper is well worth a read. Design of experiments is more than just math, as this paper stresses it is a strategy for discovery.

No Interactions? OFAT is still a Bad Idea

PDF version

Suppose you are trying to estimate the effect that 6 factors have on a response, and you know that none of the factors influence the effect of the others, so that a simple model like this

$$Y=b_1{X}_1+b_2{X}_2+b_3{X}_3+b_4{X}_4+b_5{X}_5+b_6{X}_6$$

(1)

is the perfect choice. How should you get the data you need to estimate the $b_i$’s? You may be tempted to design a test to estimate each of these factors by changing one factor at a time (OFAT). There are no interaction terms (e.g. $b_7{X}_1{X}_4$) in equation 1. So there’s no need to perform any runs that change several of the $X$’s at once, right? Wrong.

DMLS Wind Tunnel Models

Additive manufacturing, sometimes called direct digital fabrication or rapid prototyping, has been in the news quite a bit lately. I wrote a post recently for Dayton Diode about the many additive manufacturing options available for fabricating functional parts or tooling in response to comments on a piece in the Economist, and commented recently on Armed and Dangerous in a discussion about 3D printed handguns. There are just so many exciting processes and materials available for direct digital parts production today. Some of the work I've been doing recently to qualify one particular additive process for fabricating high-speed wind-tunnel models (abstract) was accepted for presentation at next year's Aerospace Sciences Meeting.

We used Direct Metal Laser Sintering (DMLS) to fabricate some proof-of-concept models in 17-4 stainless steel. DMLS is a trade name for the selective laser melting process developed by EOS. The neat thing about DMLS (and additive processes in general) is that complicated internal features like pressure tap lines can be printed in a single-piece model. Being able to reduce the parts count on a model to one while incorporating 20 or so instrumentation lines (limited only by the base area of our particular model) is really great, because one part is much faster and less expensive to design and fabricate than a multi-component model with complicated internal plumbing. The folks down at AEDC are also exploring the use of DMLS to fabricate tunnel force and moment balances for much the same reason we like it for models and others like it for injection mold tooling: intricate internal passages, in their case, for instrumentation cooling and wiring.

Fooling Yourself is Easy

Common problems from an interesting set of slides:

• Confounding in experimental design
• Mixing up the sample labels
• Mixing up the group labels
• Incomplete documentation

"Unfortunately, we suspect, The most simple mistakes are common."

AAAS, Feb 19: pursuing reproducibility audio / slides

Things we look for:

• Data
• Provenance
• Code
• Descriptions of nonscriptable steps
• Descriptions of Planned Design, if Used