Design of Experiments
From race cars to retail Web pages, almost anything can be improved by running designed experiments in JMP. What was thought to be impossible all of a sudden becomes real. And success becomes the rule rather than
Featured Article
Split-Plot Designs: What, Why, and How
Learn what split-plot designs are, why they are important and how to use them. Download a copy (PDF) of this award-winning design of experiments article by Bradley Jones, PhD, Principal Research Fellow in the JMP Division of SAS, and Christopher J. Nachtsheim, PhD, Carlson School of Management at the University of Minnesota. Published in the Journal of Quality Technology in October 2009, this paper on split-plot designs won the prestigious Brumbaugh Award from ASQ.
the exception when experiments are powered by JMP.
The software’s Custom Designer lets you create designs that can’t be matched by any compilation of pre-formulated designs offered by other software vendors. JMP offers novices and DOE experts alike the power to design experiments that meet the most complex conditions.
Built into JMP, Custom Designer lets you control the design process, meet complex experimental design requirements, and address optimal split plot design where one or more factors are held constant within a group but vary between groups. JMP also provides nonlinear design capabilities to generate and augment optimal experimental designs for fitting models that are nonlinear in the parameters. Custom Designer permits even the novice user to produce reasonable designs with only two days of training and no prior experimental design knowledge.
JMP Custom Designer meets your toughest challenges.
- Are your experimental challenges more complex than the basic textbook examples?
- Do you have a large number or combination of continuous, categorical, blocking or mixture factors?
- Do you ever need to specify a number of runs equal or greater than the minimum possible?
- Do you wish you could include covariate factors with values known in advance of the design but not controllable?
- Can you run your experiment in a convenient order by doing groups of runs where some factor or factors do not change within the group?
- Do you want to apply prior knowledge of your systems to rule out certain factor combinations in three different ways?
- Do you wish you could augment prior data with new data to resolve any remaining questions?
- Can you supply linear inequality constraints on a cubic region, specify disallowed combinations, or design within a sphere rather than a cube?
- Are you able to specify any lower order polynomial model as the model you want to be able to fit?
More on DOE
JMP® Design of Experiments (DOE) Points of Interest
(PDF, 970K)
White Paper: Interactive Data Mining and DOE: the JMP Partition and Custom Design Platforms
(PDF, 600K)
Related Data (Zip File, 21K)
Research by JMP Authors
JMP staff are active in publishing leading-edge research on design of experiments and other fields. Browse a partial list of recently published papers.
