PCB007 Magazine

PCB007-Nov2024

Issue link: https://iconnect007.uberflip.com/i/1529411

Contents of this Issue

Navigation

Page 25 of 105

26 PCB007 MAGAZINE I NOVEMBER 2024 ing with defective units, they will use statistical analyses based on a binomial probability model. e engineer checks for the presence of over- dispersion or underdispersion using the P Chart Diagnostic test for the I/L, MHC, and O/L pro- cesses. e data points (defective proportions) should align close to a diagonal line, indicating a good fit. e origin of the red diagonal line comes from the binomial distri- bution parameters n and p. All three FTY diagnostic tests lack evidence to support overdisper- sion or underdispersion. Fig- ure 1 shows the P Chart Diag- nostic test for the I/L process. e variation ratio (104.4%) is significantly less than the 95% upper limit of 243.7%, corrobo- rating no overdispersion. Next, the engineer generates P Charts for the I/L, MHC, and O/L processes. e cen- Figure 2: P chart of I/L defectives. ter line is the average proportion of defectives. e control limits are set at a distance of three standard deviations above and below the cen- ter line and vary due to unequal sample sizes. All points are in control. Figure 2 shows the P Chart for the I/L process. Figure 1: P chart diagnostic test of I/L defectives.

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB007-Nov2024