Multi-domain, Advisory Computing System in Continuous Manufacturing Processes
Many decisions that must be made during the production process mean that limited human perception is not able to meet the growing requirements of keeping the parameters and constantly striving to increase the efficiency of current production lines. The main challenge is also the continuous increase of awareness about the process and the possibility of its modernization. This forces the expansion of the production with new elements, which are not directly related to the production line itself. And this in turn forces the expansion of knowledge, for example, cooperation with new elements. The theoretical knowledge that each employee must have from every issue going to be very general without going deeper into details. Simultaneous control of all mutual elements with the same coincidence becomes impossible. Traditional methods of failure analysis and finding reasons for its occurrence are inefficient and ineffective. This paper is attempting to create a system topology for all subsystems. A comprehensive production management system, its efficiency and failure predictive system will be discussed. The system should integrate and correlate many different databases, which are conducted according to different standards. This causes a necessity of choosing a method for seeking solutions for problems in such a large stored database. “Big data” which is popular today, is using neural networks which not always is the best choice. Especially when we don’t have enough knowledge about technology and connections inside the process. Maximum use of expert knowledge, experience of employees, data acquisition and usage of unfiltered data will be presented in this paper.
- 3.Bolte, C., Kurbel, K., Rautenstrauch, C.: Integration of knowledge-based modules into a distributed production planning and control system. In: Tjoa, A.M., Wagner, R. (eds.) Database and Expert Systems Applications. Springer, Vienna (1990). https://doi.org/10.1007/978-3-7091-7553-8_16CrossRefGoogle Scholar
- 7.Cwikła, G., Grabowik, C., Kalinowski, K., Paprocka, I., Banas, W.: The initial considerations and tests on the use of real time locating system in manufacturing processes improvement. In: IOP Conference Series: Materials Science and Engineering, vol. 400, p. 042013 (2018). https://doi.org/10.1088/1757-899x/400/4/042013CrossRefGoogle Scholar
- 14.Harańczyk, G.: Prediction of failures and quality problems. StatSoft Polska (2013, in Polish)Google Scholar
- 15.Heikkinen, J., Ghalamchi, B., Viitala, R., Sopanen, J., Juhanko, J., Mikkola, A., Kuosmanen, P.: Vibration analysis of paper machine’s asymmetric tube roll supported by spherical roller bearings. Mech. Syst. Signal Process. 104, 688–704 (2018). https://doi.org/10.1016/j.ymssp.2017.11.030CrossRefGoogle Scholar
- 17.Han, J., Kamber, M., Pei, J.: Data Mining Concepts and Techniques. The Morgan Kaufmann Series in Data Management Systems (2012). https://doi.org/10.1016/c2009-0-61819-5