Interpretative identification of the faulty conditions in a cyclic manufacturing process

Abstract

The intensive development of information and communication technologies in recent years has led to an increase in data size and complexity. Conventional approaches, with associated methods of analysis based on descriptive and inductive statistics, may no longer be suitable for extracting the valuable information that is hidden in the available data.Decision Support, Big Data niques, tools, methods and concepts, which conjointly form efficient systems for extracting the value from large and complex data.Heuristic methods of analysis turn out to be very efficient for fault-diagnosis problems in complex manufacturing processes, where the states are described by complex combinations of many parameters. Recently, in the manufacturing 30 domain, a lot of research was carried out regarding fault-diagnostics problems by incorporating machine-learning methods, data mining, and other evolving techniques, targeting the problems of sensing and describing multiple and dynamic faults, monitoring, real-time processing, searching for the best features, failure-condition identification, early prediction of failures, and assessing the 35 severity of failures [11].Many of so-far untouched challenges and opportunities remain for the development and introduction of useful heuristic techniques in manufacturing systems that would improve the process quality through an intelligent use of the available data [12]. 40The paper presents a holistic data-analysis method for the interpretative identification of faulty conditions in a cyclic manufacturing process (IIFC). The objective here is to demonstrate the usability of large volumes of data generated during a widespread type of the so called cyclic manufacturing processes, such as plastic injection moulding, for extracting valuable information and new 45 knowledge models for better understanding of the manufacturing process and its faulty operating conditions. The proposed approach uses well known datamining methods and addresses the arising issues of increasing data size and complexity.Cyclic-manufacturing processes are defined as manufacturing processes in 50 which the parts are produced repeating the same sequence of steps that forms the manufacturing cycle. The data that is generated in these processes describes the state of the process for each cycle. This is suitable for applying machinelearning and other advanced techniques of data analysis, because minimal effort is required to define the common features for each cycle due to the repetitive na-55 ture of the cycle's steps. Although machine-learning techniques are well known,

Advancing manufacturing systems with big-data analytics: A conceptual framework

“…Predicted as the same sequence of operations No. Title, Description, and Mapping 1 Title: Interpretative identification of the faulty conditions in a cyclic manufacturing process (Kozjek et al 2017b) Description: The proposed data-analysis method integrates well-known heuristic algorithms, i.e., decision trees and clustering, with the purpose of identifying types of faulty operating conditions in a cyclic manufacturing process. The result of the analysis is an interpretable model for decision support that can be used for fault identification, to search for root causes, and to develop prognostic systems.…”

“…This section demonstrates the use of a conceptual framework on the selected studies of introducing data analytics in manufacturing systems. Five existing case studies of developing data-analytics solutions in manufacturing systems, i.e., (Kozjek et al 2017a(Kozjek et al , 2017b(Kozjek et al , 2018a(Kozjek et al , 2018bVrabič, Kozjek, and Butala 2017), are selected. Data-analytics solutions developed within these projects are either innovative ways of reducing the incompleteness of information and discovering new knowledge through additional use of data or they enable the more efficient reduction of information incompleteness than the conventional approaches.…”

Artificial Intelligence in Advanced Manufacturing: Current Status and Future Outlook

“…Data synthesis refers to generating synthetic data that highly resembles the real data by sufficiently learning data characteristics, which allows us to bypass the limitation of data availability. Traditionally, data synthesis often relies on interpolation (e.g., Synthetic Minority Over-sampling Technique or SMOTE), which cannot capture complex data characteristics [163]. A major breakthrough came with Generative Adversarial Networks (GAN) [164], a DL method that is able to learn salient features and synthesize data with high fidelity.…”

Machine Learning in Manufacturing towards Industry 4.0: From ‘For Now’ to ‘Four-Know’

“…This issue might be alleviated by the Synthetic Minority Over-sampling Technique (SMOTE) [162]. However, SMOTE cannot capture complex representative data, as it often relies on interpolation [163]. Data augmentation [139,164] or transfer learning [165] may address this problem.…”