Talks

Design for Six Sigma:  The Key to Efficient Innovation - Talk presented at ISPE meeting, Sept. 16, 2008.  An overview of DFSS based on the IDOV (Identify, Design, Optimize, Validate) framework.

Classification of Breast Cancer Cells Using JMP - This paper illustrates some of the features of JMP that support classification and data mining.  It is based on the Wisconsin Breast Cancer Diagnostic Data Set, a set of data used to classify breast lumps as malignant or benign based on the values of thirty potential predictors, obtained by measuring the nuclei of fluid removed using a fine needle aspirate.  The paper illustrates some visualization techniques that help build an understanding of the data set.  Then, four models are fit to the data: a logistic model, a partition model, and two neural net models. 

Data sets in JMP format (Download a free trial version of JMP here)

Visualization and the Improvement of Anodized Parts Using JMP -  Using JMP's visualization and Custom Design capabilities, a team was able to substantially improve the performance of a low yield anodize process. Visualization techniques allowed the team to set specifications for the four quality characteristics. The Custom Design platform allowed the team to design an experiment in five process factors that estimated the effects of interest, subject to constraints on the number of runs. Then, using the Profiler, the team found settings for the process factors that greatly improved yield.

Multiple Optimization Using the JMP® Statistical Software - The challenge in multiple response optimization is to find settings for multiple input variables that achieve desirable performance levels for one or more responses. This powerpoint presentation discusses multiple optimization using JMP.

Six Sigma, Data Mining, and the JMP® Partition Platform - Data mining refers to a collection of techniques used to analyze large observational data sets. This talk discusses JMP’s partition platform, which is a version of classification and regression tree analysis.

Interactive Data Mining and DOE – This powerpoint addresses the synergy between data mining and design of experiments in Six Sigma projects. An example is presented where a team uses the JMP Partition platform to narrow the list of variables and settings for a designed experiment. The team then uses JMP6’s Custom Design platform to design the experiment.
Data set in JMP format
(Download a free trial version of JMP here)

 

Interactive Data Mining and DOE – This document summarizes the powerpoint file of the same name, but provides more detail on the JMP partition platform. The paper addresses the synergy between data mining and design of experiments in Six Sigma projects. An example is presented where a team uses the JMP Partition platform to narrow the list of variables and settings for a designed experiment. The team then uses JMP6’s Custom Design platform to design the experiment.
Data set in JMP format
(Download a free trial version of JMP here)

 

 

North Haven Group, LLC
2 Senter Drive
Brookline, NH 03033
(P) 603.315.3518
© Copyright

Website Design and Maintenance by JAZ Design Company