Design Expert 7.1

Nearly two years ago, StatEase released a major version upgrade of their flagship DoE (Design of Experiments) software to DesignExpert (or DX) 7. Earlier this year the first decimal increment arrived, and by June had matured to the 7.1.3 which has been under road test since.

In a mature, well-established product which dominates its market, upgrades should be expected to reflect developing practice and display evolutionary growth rather than piling on attention-seeking gimmicks; and so it is here. The shift from 6 to 7.0 brought solidly useful additions (3D surface plots for categoric factors, for example, and a t-value effects Pareto chart), capacity expansions (including 512 runs by 21 factors for two-level fractional factorials) and numerous productivity or other subtle enhancements from ignoring faulty data at single cell level to point colour coding as a dimensional display extension. Incremental update to 7.1 continues the same philosophical process.

Core functions have been extended usefully. Main effects screening for two-level factorial designs, for example, saw MR4 (Min Run Res IV) for between 5 and 50 factors in the first push to version 7·0 and with 7·1 much the same (from 6 to 50 factors) has been done for MR5, which significantly increases economic balanced resolution access. For real world application (which includes a majority of users who have no specialist understanding of DoE or even statistical methods in general) a significant addition is the FDS line graph. This shows the relationship between areas having high and low prediction error levels, and serves as a relatively simple alarm measure across design types while also providing enough subtlety to allow graded user responses.

The design editor in DX has long been a strong point, providing a central cockpit from which to intuitively refine and test most aspects, but various new control features have now been added to the package, small in themselves but significantly enhancing productive control. Placing power calculations in front of the user at build time, display of mixture constraints in different coding schemes, expanded “right click” context controls and the ability to type in variable constraints directly to the design in ratio form, are examples. Other usability improvements include the improved visual reading accuracy from the addition of gridlines on the back planes of 3D displays, reduced “hunt and peck” scrolling through solutions and bookmarked report selection.

Talking to other software, normally one of the less strong areas in this market segment as a whole, is enhanced with the extension of import from older files, including DesignEase format, and the ability to write into Taylor VarTran files (as well as text, native versions 7 or 6, and XML). Graphics can be exported to EMF, EPS and a range of bitmaps for external use.

All in all, a very worthwhile upgrade has been made even more attractive in 7.S.


Robert Roe reports on developments in AI that are helping to shape the future of high performance computing technology at the International Supercomputing Conference


James Reinders is a parallel programming and HPC expert with more than 27 years’ experience working for Intel until his retirement in 2017. In this article Reinders gives his take on the use of roofline estimation as a tool for code optimisation in HPC


Sophia Ktori concludes her two-part series exploring the use of laboratory informatics software in regulated industries.


As storage technology adapts to changing HPC workloads, Robert Roe looks at the technologies that could help to enhance performance and accessibility of
storage in HPC


By using simulation software, road bike manufacturers can deliver higher performance products in less time and at a lower cost than previously achievable, as Keely Portway discovers