1. STATISTICS AND OUTLIERS Aaron Saks Process Advancement Leader Boise, Inc 10/26/2011

2. Bio Graduated in 2007 – ChemE and PSE Started career with Boise Inc, Wallula WA as a process engineer October 2008 went to work for Envoy Development April 2010 returned to Boise as Process Advancement Leader, Wallula WA November 1 st 2011 move to Boise, ID to become Project Manager for Boise Packaging

3. Statistics and Data Why is it important? Everything we really know, we know because of data. As a new engineer, we know how to apply math and logic to solve problems. We don’t really know how anything works. By focusing on the data we can learn, solve problems, and teach others. There is a lot of data out there. Statistics = The language of data.

4. Thinking Statistically We can’t just think in terms of the “average”. Need to think in terms of the distribution of data, and the probability of events occurring. What are good statistical tools for a new engineer? Six Sigma methods Understanding the common probability distributions and their mean and variance Understanding histograms and pareto charts Excel skills

5. Tear Strength 4.1% represents ~ 1640 tons below current spec ~40 tons (0.1%) rejected for below-spec MD tear Current Target: 42

6. Ppkm: Run to Target and Reduce Variation Historically, papermakers would run “in the warning” all day, as long as the tests are within specification limits. This resulted in running off target, with different means run to run (Poor Quality). Needed a way to encourage ($$$) running to targets, and reducing variation within the specs. Created a variation on the classical Process Capability metric “Cpk”. Result  a financial incentive for improving Quality.

7. P pkm - Running to Target and Reducing Variation P pkm is an example of a process performance metric. P pkm captures both deviation from target and variation within specification limits. If P pkm >= 1, then the process is Capable: running to target and variation is well within specification limits To improve Quality we award operators for the number for Key Product Properties that have a Ppkm value above 1.0.

8. Ppkm = 1.02 Here is an example of a process that has a P pkm >1, meaning it is fully capable of meeting customer expectations. Process is running to target – i.e. the mean is equal to the target Variation is well within the Specification (Red) limits. 8

9. Ppkm =.65 9 Running to Target – Excessive variation Acceptable variation – but off Target

10. Outliers We spend most of our time working on the outliers: Process problems Lowering cost Increasing production Outliers exist in people too Top 20 Middle 70 Bottom 10 The future leaders of tomorrow will be from the top 20.

11. Thoughts for a New Engineer Understand what is expected Focus on independent learning Always speak from facts and data Effective communication creates results Focus on business results and accomplishments You don’t have to be at the top to be a leader

12. Questions: What has been your biggest challenge? If you could go back to school what additional classes would you take? What's a typical day like? What kind of products does Boise make? How long do projects last? What ethical issues have you encountered?