Introduction:

Data from experiments, product testing, simulation, surveys, and statistical process and quality control must be appropriately analyzed before results can be determined and conclusions drawn.  Statistical analysis and modeling involves the appropriate application of statistical analysis techniques, each requiring certain assumptions be met, to perform hypothesis tests, interpret the data, and reach valid conclusions.  To be credible, results from experimentation or testing must be obtained following established statistical procedures, including experimental design and the appropriate use of statistical analysis and modeling techniques.  These results can then be reproduced, within sampling error, by repeating the experiment.

Statistical analysis and modeling requires careful selection of analytical techniques, verification of assumptions, and verification of data.  Descriptive statistics, graphs, and relational plots of the data should first be examined to evaluate the legitimacy of the data, identify possible outliers and assumption violations, and form preliminary ideas on variable relationships for modeling.  The many different statistical analysis and modeling techniques have different goals and are appropriate for different types of data.

Benefits:

• Application of appropriate statistical analysis techniques
• Development of appropriate conclusions and key learnings from the data
• Ensuring results address experimental objectives
• Maximizing information gained from the data
• Maximizing chances of the experiment being successful

Capabilities:

• Experimental design
• Selection and application of appropriate analysis methods
• Interpretation and presentation of analysis/modeling results
• Statistical analysis and modeling techniques
• Descriptive statistics
• Data graphs, plots, and exploratory data analysis
• Contingency table analysis
• Multiple linear regression analysis
• Logistic regression
• Analysis of variance and covariance
• General linear models
• Time series analysis
• Reliability/survival analysis
• Discriminant analysis
• Principal components analysis
• Factor analysis
• Cluster analysis
• Multivariate analysis
• Nonparametric analysis
• Statistical analysis and modeling software
• Minitab
• Systat
• Statistica
• SAS
• SPSS
• BMDP

Experience:

• Analyzed and modeled data for joint project with Motorola to test new circuit board low-residue soldering process
• Analyzed data from process characterization/qualification experiment for AC Rochester, a division of General Motors, for their new low-residue soldering line
• Analyzed and modeled data from Low-Residue Soldering Task Force project to test low-residue soldering process reliability across several manufacturers, equipment, and materials
• Analyzed data from experiment to determine the effects of smoke on electronic circuitry for NRC
• Analyzed and modeled simulation data to determine influential factors on customer satisfaction for Intel manufacturing line
• Analyzed and modeled data from Neutron tube subassembly manufacturing process characterization data
• Analyzed and modeled data from tests with SEHO USA, Inc. to evaluate new fluxless soldering process

References:

1. Tanaka, T. J., S. P. Nowlen, and D. J. Anderson (1996), "Circuit Bridging of Components by Smoke," NUREG/CR-6476, SAND96-2633, Sandia National Laboratories, Albuquerque, NM.
    
2. Iman, R. L., R. V. Burress, D. J. Anderson, et al (1995), "Evaluation of Low-Residue Soldering for Military and Commercial Applications: A Report form the Low-Residue Soldering Task Force," SAND95-1060, Sandia National Laboratories, Albuquerque, NM.
    
3. Anderson, D. J., R. M. Cranwell, R. L. Iman, and P. Van Buren (1994), "Evaluation and Qualification of Environmentally Conscious Manufacturing Processes for Commercial and Military Applications," International IEEE Workshop on Environmentally Conscious Manufacturing for the Electronics Industry:  Research on Advanced Materials and Processes, IBM Research Center, Yorktown Heights, NY.
    
4. Iman, R. L., D. J. Anderson, M. E. Armendariz, L. R. Lichtenberg, P. Van Buren, and M. T. Paffett (1992), "Evaluation of a No-Clean Soldering Process Designed to Eliminate the Use of Ozone Depleting Chemicals," SAND92-1776, Sandia National Laboratories, Albuquerque, NM.