Issue link: https://iconnect007.uberflip.com/i/1511625
30 PCB007 MAGAZINE I NOVEMBER 2023 to uniquely identify each panel. is can be achieved by the application of machine-read- able 2D barcodes—allowing processes to auto- matically set up for the panels as they arrive. Learning From the Data Data represents raw facts resulting from observation or measurement. Information is data arranged or interpreted to provide mean- ing, while knowledge is information that has been understood and can be used to make decisions. By adding interpretation and con- text to data, it can be transformed into mean- ingful information and, ultimately, knowledge. Wisdom is the ability to use knowledge to take action. In a Smart manufacturing sys- tem, data is captured and analyzed to provide action- able insights and to push cor- rective actions into the pro- duction data highway. How- ever, it's essential to under- stand that not all data will lead to the correct action, and incorrect data can lead to incorrect deci- sions. As Bill Gates said, "e first rule of any tech- nology used in business is that automation applied to an efficient operation will magnify the efficiency. e second is that automation applied to an inefficient operation will magnify the inefficiency." Data Quality and Integrity For a successful Smart factory implementa- tion, establishing a robust framework for data quality and integrity is paramount for mitigat- ing risks associated with existing operational processes. Inconsistencies and inaccuracies in data can erode trust rapidly, leading to a loss of confidence in the information being used. e value of data is intrinsically linked to its quality, and without proper data quality assur- ance, its potential remains untapped. Fur- thermore, when data is not integrated reli- ably, organizations are oen compelled to fall back on traditional methods and institutional knowledge, undermining the potential for innovation and efficiency that well-managed data offers. It is not uncommon for a deluge of data to find its way into projects and reports without undergoing the necessary validation. Robust process controls are vital from the start to ensure data quality instead of correct- ing mistakes. Measurement machines must be maintained and calibrated to prevent noise or shi within the data captured. e envi- ronment around the machines should be monitored and controlled to prevent changes in temperature or humid- ity being reflected in the data captured. Changes in data quality should be monitored with daily checks using known standards, and statistical con- trols should determine when maintenance or calibration is required rather than waiting on predetermined schedules. Measurement systems analysis will deter- mine the accuracy and precision of the mea- surement processes by verifying the differ- ences in data are due to actual differences in the product being measured and not due to variation in measurement methods or the machine operator. It is critical the machine is set up consistently for each batch, no mat- ter who is performing the setup. Gauge R&R methods should be used to check repeatability and reproducibility. Enriching the Data Within the PCB manufacturing facility, a plethora of design and manufacturing data is scattered across various locations and stored in a variety of formats. By combining data from multiple sources, it is possible to derive addi- tional information which will aid in our under- standing of the process. Wisdom is the ability to use knowledge to take action.