Statistical Thinking to Improve Quality

This posting describes the final phase of the resin output variation example that illustrates the Hoerl-Snee process improvement strategy.   This example appears in Britz et al (2000) and in Hoerl and Snee (2002).   The Ricoh team is focusing on reducing the variability in resin output quantity.   The previous post stated that the next step for the team was to investigate the weighing processes.

The overall process had two weighing processes.  The first was an in-process manual method, and the second method was a final, automatic scale.   The manual method had individuals reading a line on a scale.  They observed that individuals of different heights read the line from different viewpoints.  Thus, they produced different readings.   The team changed the presentation of the line so people of different heights had the same view point.   This change reduced in-process measurement variation.

Next the team investigated the automatic scale and found significant measurement errors.   They reduced these errors by:

1.     Redesigning the scales protective cover.

2.     Establishing procedures for checking the alignment on a periodic basis.

The following figure presents a control chart showing the results for this project.  The difference between the final upper and lower control limits was less than the team objective of ± 5 kg.  However, the resulting average was 4292 kg which is less than the original target of 4300 kg.   Given the reduction in output variability, management regarded the results as more than adequate.   The improvement also resulted in reduction in the variation of resin viscosity.   This verified the team’s motivation to reduce variation of finished product quality by reducing the output quantity variation. To maintain the results, the team created procedure manuals and established a schedule adjusting the automatic weighing process.

The overall improvement process consisted of four Plan-Do-Check-Act (PDCA) cycles.   This posting describes the last one, the previous post describes two of them and the posting on March 21 (Hoerl-Snee Example) describes the first one.   That one focused on finding and correcting special causes.   This process is different than that suggested by a serial DMAIC process.   Our next posting will present the Hoerl-Snee process improvement strategy which has an overall PDCA approach.

References
1.     Britz, G. C., D. W. Emerling, et al. (2000). Improving Performance Through Statistical Thinking. Milwaukee, WI, ASQ Quality Press.
2.     Hoerl, R. and R. D. Snee (2002). Statistical Thinking - Improving Business Performance. Pacific Grove, CA, Duxbury.

 

Posted by Gordon M. Clark at 01:50 PM | Permalink | Comments (0)

This posting continues the resin output variation example described to illustrate the Hoerl-Snee process improvement strategy.   This example appears in Britz et al (2000) and in Hoerl and Snee (2002).   The Ricoh team is focusing on reducing the variability in resin output quantity.   The previous post ended with a description of cause & effect diagram the team constructed to list potential sources of variability.
Based this diagram the team regarded the following potential causes as most likely to be the largest contributors to output variation:

1. The procedure for dividing the resin after phase 2 (potentially the cause of differences between line A and B).
2. The solvent feed ratio.
3. The weighing process, i.e., final (automatic) and in-process (manual).

After attacking the first potential cause, the team found that some resin remained in the reaction tank after sending the materials to the two lines.  That meant that line B had less input and therefore less output.   After changing the dividing procedure, the team found no significant difference between the outputs of the two lines.

The output quantities still had too much variation.   The team turned to the second potential cause, i.e., the solvent feed ratio.  The following figure shows a scatter plot indicating that increasing solvent feed ratio is correlated with increasing output.   In calculating the regression line the team regarded the high output occurring at a feed ratio slightly less than 1, as an outlier.   This correlation violated the team’s knowledge of the underlying process.   They investigated the measurement of the feed ratio, and they found that the ratio measurement was affected by the length of time the solvent was in the tank.   They changed the procedure to insure that the solvent had stabilized prior to measurement.   They collected more data to measure the impact of this change and found less variation in the measured feed ratio and no correlation between the measured feed ratio and the output quantity.

The output variation still did not meet their targets shown in the previous posting.   The next posting will present their analysis of the weighing process.   This posting shows two cycles of PDCA which differs from the sequential process suggested by DMAIC.   From a DMAIC viewpoint the team went through two cycles of Analyze-Improve.

References

1. Britz, G. C., D. W. Emerling, et al. (2000). Improving Performance Through Statistical Thinking. Milwaukee, WI, ASQ Quality Press.
2. Hoerl, R. and R. D. Snee (2002). Statistical Thinking - Improving Business Performance. Pacific Grove, CA, Duxbury.
   

Posted by Gordon M. Clark at 10:17 AM | Permalink | Comments (0)

This posting continues the resin output variation example described to illustrate the Hoerl-Snee process improvement strategy.   We take this example from Britz et al (2000).   It also appears in Hoerl and Snee (2002).

Having removed the special cause, the Ricoh team focused on output quantity variability.   A histogram displays this variability, and the following figure shows recent output data.  This histogram displays an unexpected pattern indicating a combination of two underlying distributions for the output quantity.   Notice the peaks at 4284 and 4308 kg.

The process flowchart appearing in the previous posting suggested that these two component distributions were due to the split after phase 2 into two separate lines, i.e., lines A and B.   The following histograms shown below confirmed this difference.   The output from line B was consistently lower than line A.   Based on the needs of their customers, the team established the limits shown in the histograms, i.e., 4300 kg ± 5 kg.

Clearly, the variation in output quantity is excessive.   Next the team conducted a brainstorming session to document their collective thinking on potential causes of excessive variation and differences between the two lines.   The following cause and effect diagram shows the result of this session.

The next posting will describe the investigation based on the potential causes shown above.  
Note that the improvement process is iterative. Gather data, identify special cause, gather more data, notice differences, and then conduct brainstorming session.   This improvement strategy looks more like Shewhart’s Plan-Do-Check-Act (PDCA) than the DMAIC steps recommended for Six-Sigma projects.   Also, the team didn’t adopt a specified target until after two data analysis steps.   That is, their Define step occurred in their second PDCA cycle.

References

1. Britz, G. C., D. W. Emerling, et al. (2000). Improving Performance Through Statistical Thinking. Milwaukee, WI, ASQ Quality Press.
2. Hoerl, R. and R. D. Snee (2002). Statistical Thinking - Improving Business Performance. Pacific Grove, CA, Duxbury.
   

Posted by Gordon M. Clark at 02:06 PM | Permalink | Comments (2)

This posting starts presentation of an example application that illustrates the Hoerl-Sneel process improvement strategy.  We take the example from Britz et all (2000), and they reference Imai (1986).

Resin Output Variation Example (Part A)

Ricoh’s Numazu plant made raw material used as ingredients for copy machine toner.   The tolerances for the raw material ingredients were measured in ten-thousands of a gram.   They had a team which continually monitored the process to achieve continuous improvement.  The team noticed a problem with actual output results compared to theoretical output quantities.   The yield ratio frequently exceeded 1.0.   That is, the actual output quantity for a batch divided by the theoretical output quantity sometimes exceeded 1.0.   The following figure shows a run chart displaying these results.  These values were technically impossible, so the team attributed these results to undesirable variation somewhere in the process.   Their experience indicated that this variation would degrade finished product quality.   They wanted to know the source of the variation and how to eliminate it.

 

 

The figure on the left gives a flowchart of the process.  Notice that after the second phase the process splits into two lines supposedly with identical sub-processes.

The next step in their investigation was to examine the run chart for stability.   The time period in the middle of the run chart had the greatest concentration of values above 1.0.  Further analysis showed that a drop in air pressure was the root cause (special cause) for the excessive variations above 1.0.   They verified this conclusion, but removing this cause did not completed eliminate the outputs greater than theoretical predictions.

Notice that their first steps were to document (understand) the problem and look for special causes.

 

 

References

1. Britz, G. C., D. W. Emerling, et al. (2000). Improving Performance Through Statistical Thinking. Milwaukee, WI, ASQ Quality Press.
2. Imai, M. (1986). Kaizen, The Key to Japan’s Competitive Success.  New York: Random House Business Edition.
   

Posted by Gordon M. Clark at 11:33 AM | Permalink | Comments (0)