Issue link: https://iconnect007.uberflip.com/i/1415897
OCTOBER 2021 I SMT007 MAGAZINE 59 inspection solutions should be able to collect and analyze a large amount of data to produce traceable results. Our approach to the analysis solution is to collect all inspection and measurement data from all equipment via a hub, and then provide the data anywhere within the network through a web-based application. Big data is a foun- dation for Industry 4.0, so advanced inspec- tion systems must evolve from simply judging "pass/fail" tools into highly intuitive, dynam- ic decision-making systems, which emphasizes the need for reliable, traceable data. is traceable data can then ensure the high- est levels of transparency by showing all con- ditions of the lines, including machine config- uration and soware version, while provid- ing the required documentation for changes to the job file, package, part, and more. Users can quickly verify whether all lines are with- in the ideal conditions. If a variance occurs, the user can instantly upload optimized pro- grams and inspection conditions without fine tuning with the Library Manager module (Fig- ure 2). e soware module provides a com- plete central management solution for com- ponent libraries, programs, inspection condi- tions, and more. Library Manager combines all equipment into a single centralized library. All changes are traceable and manageable by user level Identification. Such controlled data man- agement allows continuous analysis of the raw data and helps guide experts towards the right direction. Eliminates the Bottleneck Of course, maintaining quality, repeatable measurement data is not enough to realize a smart factory. Instead, analyzed data needs to be instantly visualized with relevant indica- tors like yield rate, NG analysis, PPM analysis, Gage R&R, offset analysis, and more to allow users to compare board performance and iden- tify process deviations. Using the real-time sta- tistical process control module, users can iden- tify the exact defect origin by checking false calls and NG parts from the dashboard, as well as evaluate, and optimize default settings. For instance, if the thickness was the major problem in a worst-case part, users can click the part to view analysis result and find the root cause. An X-bar chart of measured thick- ness of the part will be shown across time with average, minimum, and maximum val- ues, plus tolerance levels (Figure 3). If values frequently deviated from average values and tolerances are too tight, users can adjust the tolerance levels to minimize false calls. On the other hand, if the process was stable, operators can tighten the tolerance to prevent Figure 2: KSMART Library Manager module on web interface.