Issue link: https://iconnect007.uberflip.com/i/1529411
28 PCB007 MAGAZINE I NOVEMBER 2024 Next, the engineer generates Capability Analysis reports for the I/L, MHC, and O/L processes. e report displays four graphs: P Chart, Rate of Defectives, Cumulative % Defective, and Histogram. Figure 3 shows the Capability Analysis report for the I/L process. ere is much information contained in the report, and the process engineer extracts and infers the following: • P Chart: e I/L process %defective is sta- ble and in control. • Rate of defectives: e data fall randomly about the center line; the probability of a defective item is constant across different sample sizes. e data follow a binomial distribution. • Cumulative %defective: e %defective is stabilizing, shown by a flattening of the plotted points along the mean %defective line. ere are enough samples for a stable estimate of the %defective. • Histogram: e peak represents the most common values and approximates the center of the %defectives, about 6%. e spread varies about 3%. e engineer real- izes histograms need at least 50 data points for unambiguous interpretation. Finally, the engineer wants to chart the RTY. ey decide that a Run Chart is the most straightforward and efficient method. Run Charts expose patterns or trends that indi- cate the presence of special-cause variation 9 . e engineer noticed that the RTY trended up from week three through week seven and then dropped sharply in week eight. Investi- gation is warranted to determine what hap- pened (Figure 4). Unlike the FTY, the RTY exposes the "hid- den factory" 10 , meaning the unseen activities outside the normal processes that occur when a defect is discovered. ese unseen activi- ties add labor that eats away productivity and drives costs up. Using RTY can aid the engi- Figure 3. Capability analysis report for the I/L defectives.