Big Data from Modern IT Technology

The application of modern IT technology represented by ERP, MES, SPC and other systems has brought tremendous help to the production, quality and operation management of enterprises. Not only that, these systems have also accumulated in the process of operation Another valuable asset: data. Today, with the increasingly widespread application of big data, how can enterprises fully discover the important information hidden in the data and help them improve the quality level in all dimensions of product, service and even management? Quality big data is not only about IT Famous fine management consulting brand Micro consulting In a large number of enterprises in the process of research found that many people think that all aspects of data collection, storage, acquisition, analysis and presentation is an IT problem: the data Acquisition can be measured by the computer software to achieve, Oracle, SQL Server and other professional database can provide us with enough powerful data storage capabilities, and we can easily get through the interface program from the massive data to the information we need, Statistical analysis software can help us analyze the data, build models, and present the results of the analysis with useful icons. However, why do not many well-established enterprises in many IT systems get help with mass data technology? What data really needs to be collected and stored in terms of data collection? Product yield data we want to collect? Pressure, temperature, weather, time and so on and so on, which ones are we need to collect? Should we use product pass / fail as the standard for product quality evaluation, or should we have a better measurement? For the data that needs to be collected, how can we reduce the error and avoid errors effectively? In terms of data acquisition and arrangement, what kind of data should we analyze to help solve the practical problems? We often have missing values, outliers or outliers in our data. How can we identify the authenticity of the data, especially those special values? How to properly fill in, correct or remove the data that has been prepared? analysis? The highlight is also data analysis. When we first analyze a data without any previous experience with the data, what analytical methods should we use to find out the most secret of the data? Hypothesis Testing, ANOVA, Simple / Generalized Linear Models, Cluster Analysis ... So many analytical methods, how do we make the right choice based on the situation, and how to find the best balance between Lack of fit and Over-fitting ? How to convert engineering problems into data analysis problems, and how to restore the data analysis results to the actual engineering application environment? Different analysis results in the end that way to show the most useful to others to understand the findings we get from the analysis? Obviously, all these are far beyond the scope of the IT system. The field of big data applications can be all aspects of our production and life, but in terms of quality management, although statistical quality management in the United States has long been a mature method of quality improvement, the more demanding in the process of the industry, data collection and analysis The higher the requirement is. However, considering the research on micro-consulting, quality big data has more extensive connotation, more diverse methodology and much greater value for the enterprise. "Mass Big Data" can be said to be a solution to a set of quantitative decision-making ideas, industry quality management experience, a reasonable industrial data collection plan, professional industrial data analysis (including but not limited to statistics) method! China's Quality Management Since it has not experienced the stage of genuine statistical quality management, it is often more effective to use the ideas and methodology of "mass big data" and "quantitative decision-making" in the process of refinement of enterprise management . In summary, we need to develop targeted data collection plans based on the characteristics of each line, including indicator formulation, variable selection, data structure design, sample size and effectiveness evaluation, experimental structure design and so on; We need to extract, clean and collate data that help solve our problems with modern IT technology. Sometimes we also need to integrate data from after-sales service, warranty and even customer satisfaction with R & D or production data to evaluate the actual improvement we are going to make. The impact of work on after sales and customers; on this basis, we can explore the data and find clues that are important for improving quality, product design or customer loyalty. Exploratory data analysis can help us do this well A job; the key determinants of the key quality indicators often need to be locked down after we have repeatedly tempered and analyzed the quality data, then we can guide us to develop a workable quality improvement plan that predicts the quality level ... " Time will lie, "we must also To screen. If the quality data (including research and development, production, after-sales, reliability, customer satisfaction, etc.) compared to the gold sand, IT systems can be seen as a finishing container, but we also need to have how to effectively gold out of gold Idea and method.