Data Mining for Design and Manufacturing

The productivity of individual companies as well as the efficiency of the global economy can be dramatically affected by Engineering Design and Manufacturing decisions and processes. Powerful data acquisition systems (such as minicomputers, microprocessors, transducers, and analog-to-digital converters) that collect, analyze, and transfer data are in use in virtually all mid-range and large companies. Over time, more and more current, detailed, and accurate data are accumulated and stored in databases at various stages of design and production. This data may be related to designs, products, machines, materials, processes, inventories, sales, marketing, and performance data and may include patterns, trends, associations, and dependencies. There is valuable information in the data. For instance, understanding the data and the quantitative relationships among product design, product geometry and materials, manufacturing process, equipment capabilities, and related activities could be considered strategic information. Extracting, organizing, and analyzing such useful information could be utilized to improve and optimize company planning and operations. The large amount of data, which was generated and collected during daily operations and which contain hundreds of attributes, needs to be simultaneously considered in order to accurately model the system's behavior. It is the abundance of data, however, that has impeded the ability to extract useful knowledge. Moreover, the large amount of data in many design and manufacturing databases make it impractical to manually analyze for valuable decision-making information. This complexity calls for new techniques and tools that can intelligently and (semi)automatically turn low-level data into high-level and useful knowledge. The need for automated analysis and discovery tools for extracting useful knowledge from huge amounts of raw data suggests that Knowledge Discovery in Databases (KDD) and Data Mining methodologies may become extremely important tools in realizing the above objectives. Data mining is primarily used in business retail. Applications to design and manufacturing are still under utilized and infrequently used on a large scale. Data Mining is often defined as the process of extracting valid, previously unknown, comprehensible information from large databases in order to improve and optimize business decisions. Some researchers use the term KDD to denote the entire process of turning low-level data into high-level knowledge.