PPG INDUSTRIES: STATISTICAL QUALITY CONTROL
PPG INDUSTRIES: STATISTICAL QUALITY CONTROL
INSTRUCTION
PPG Cleveland was the largest automotive-coating manufacturing plant in the world with annual production exceeding 18 million gallons. The plant was a converter-type operation, which assembled and processed raw materials into hundreds of primers and topcoats for application at automotive assembly plants throughout the United States.
When a sample of paint from the production process was received in the Cleveland quality-control lab, it was tested for physical and visual properties. Important physical properties included viscosity, nonvolatile solids, VOC, and film cure.
OBJECTIVE OF THIS ASSIGNMENT
Given the exhibits of some of the physical properties, Charles Orange has to interpret the control charts for E5657 and also develop and interpret control charts for other property data such as E5657 Grind Time, DPX 1715 Solids and VOC, DCT8914 Package viscosity, DCT8914 Spray viscosity.
ANALYSIS
1. Interpretation of Figure 2
1) Objective of PGG
The control charts were applied to the grinding process in an effort to reduce extremely high batch-to-batch grind-time differences. So what we should focus on is the average ranges of the sample.
2) Data Analysis
From X-bar chart of Figure 2, there are only five points (12.5833, 18.6667, 12.9167, 17.9167, and 12.9167) are close to the UCL or LCL, others are fluctuating around the average of sample mean. Given that X-bar chart monitors the changes in the mean of a process, we can conclude that the tendencies of the average grind time are centralized.
More importantly, we have to pay attention on the R-chart of the grind time. We can see that there are up to 16/37 points close to the UCL or LCL. Given that R-chart monitors the dispersion or variability of the process, if some points are far away from the average range, it indicates that among the randomly selected samples, the gap between the largest amount and the smallest amount is
INSTRUCTION
PPG Cleveland was the largest automotive-coating manufacturing plant in the world with annual production exceeding 18 million gallons. The plant was a converter-type operation, which assembled and processed raw materials into hundreds of primers and topcoats for application at automotive assembly plants throughout the United States.
When a sample of paint from the production process was received in the Cleveland quality-control lab, it was tested for physical and visual properties. Important physical properties included viscosity, nonvolatile solids, VOC, and film cure.
OBJECTIVE OF THIS ASSIGNMENT
Given the exhibits of some of the physical properties, Charles Orange has to interpret the control charts for E5657 and also develop and interpret control charts for other property data such as E5657 Grind Time, DPX 1715 Solids and VOC, DCT8914 Package viscosity, DCT8914 Spray viscosity.
ANALYSIS
1. Interpretation of Figure 2
1) Objective of PGG
The control charts were applied to the grinding process in an effort to reduce extremely high batch-to-batch grind-time differences. So what we should focus on is the average ranges of the sample.
2) Data Analysis
From X-bar chart of Figure 2, there are only five points (12.5833, 18.6667, 12.9167, 17.9167, and 12.9167) are close to the UCL or LCL, others are fluctuating around the average of sample mean. Given that X-bar chart monitors the changes in the mean of a process, we can conclude that the tendencies of the average grind time are centralized.
More importantly, we have to pay attention on the R-chart of the grind time. We can see that there are up to 16/37 points close to the UCL or LCL. Given that R-chart monitors the dispersion or variability of the process, if some points are far away from the average range, it indicates that among the randomly selected samples, the gap between the largest amount and the smallest amount is