Statistical process control (SPC) involves using statistical techniques to analyze and monitor the variation in manufacturing processes and maintain processes to fixed targets. The use of SPC will greatly enhance the in-mill quality control program. To demonstrate the benefits of applying SPC in general panelboard products, the current member mill application of SPC was first reviewed in this study. Mill visits and a survey were conducted to identify and prioritize the areas for improvement in plywood manufacturing. Key process variables were determined in terms of product performance, productivity and material recovery. Subsequently, different SPC statistics/control charts were reviewed and effective tools for process control were selected. A two-step sampling and statistical analysis method was established for panelboard quality control with a given confidence level. Coupled with this panelboard quality control module, an integrated computer software program, PanelSPC®, was developed for mill data acquisition, data analysis and decision-assistance. The software helps establish histograms and X-bar and Range (R) (or standard deviation, s) control charts for a given process variable and can perform process capability analysis.
To address the No. 1 issue, i.e., panel delamination in plywood manufacturing, a practical SPC approach was established. A cause-and-effect diagram was first constructed to identify the checkpoints and key variables involved in the manufacturing process. A histogram chart (Pareto) was then established to: 1) find the root causes of panel delamination due to a low percent wood failure; and 2) identify potential process variables overlooked in the current practice. Mill and laboratory studies were conducted to investigate the effect of key variables on panel gluebond performance using an experimental design approach. The results revealed that panel pressing time and compression ratio (CR) had a tremendous effect on panel gluebond quality. This led to a new direction to reducing panel delamination.
As a case study, a production data set of dry Douglas-fir heart veneer width was collected and imported into the PanelSPC® software for statistical analysis. With this off-line SPC tool, the distribution, X-bar and R control charts of the dry veneer width were established. The trial control limits were computed and then revised for continuous production monitoring. The assignable causes were subsequently identified to maintain the dry veneer width under statistical control with less variability. However, in this case, the dry veneer width was still centered incorrectly with many sheets being out of the specification limit. This problem was ultimately tracked to the wider clipping width resulting from inaccurate green veneer sorting. It was demonstrated that with a proper application of SPC, the assignable causes and upstream (in this case) or downstream problems can be detected. By adjusting veneer drying control and green veneer moisture sorting, dry veneer width can be tightly controlled, resulting in approximately 1.9% recovery improvement or about $300,000~$450,000 annual savings for an average plywood mill.
With the off-line PanelSPC® tool, sources of process variability can be detected and the manufacturing process can be modified and better controlled to attain greater material recovery, increased product quality and productivity.