Abstract
With the Industry 4.0 paradigm comes the convergence of the Internet Technologies and Operational Technologies, and concepts, such as Industrial Internet of Things (IIoT), cloud manufacturing, Cyber-Physical Systems (CPS), and so on. These concepts bring industries into the big data era and allow for them to have access to potentially useful information in order to optimise the Overall Equipment Effectiveness (OEE); however, most European industries still rely on the Computer-Integrated Manufacturing (CIM) model, where the production systems run as independent systems (i.e., without any communication with the upper levels). Those production systems are controlled by a Programmable Logic Controller, in which a static and rigid program is implemented. This program is static and rigid in a sense that the programmed routines cannot evolve over the time unless a human modifies it. However, to go further in terms of flexibility, we are convinced that it requires moving away from the aforementioned old-fashioned and rigid automation to a ML-based automation, i.e., where the control itself is based on the decisions that were taken by ML algorithms. In order to verify this, we applied a time series classification method on a scale model of a factory using real industrial controllers, and widened the variety of parts the production line has to treat. This study shows that satisfactory results can be obtained only at the expense of the human expertise (i.e., in the industrial process and in the ML process).
Highlights
Over the last few years, the manufacturing sector has been facing a new industrial revolution.This fourth industrial revolution—Industry 4.0—originates from Germany, and was initially a proposal for developing a new concept of German economic policy in 2011 [1]
We demonstrated that a specific time series classification (TSC) method, namely Word ExtrAction for the time Series cLassification (WEASEL), which is integrated into an automation architecture, is able to enhance the flexibility of the production lines
We showed that: (i) the TSC method can be as accurate as a static program, so the effectiveness of the classification is not decreased; (ii) the TSC method is able to handle changes in terms of number of different colours/parts with some adaptations to the preprocessing method used before training the model; and, (iii) a single model trained on all of the parts can only classify a sub-part of the total variety of products
Summary
Over the last few years, the manufacturing sector has been facing a new industrial revolution.This fourth industrial revolution—Industry 4.0—originates from Germany, and was initially a proposal for developing a new concept of German economic policy in 2011 [1]. Customisation/Individualisation of the products: the production needs to adapt to the customer’s requirements that tend to be increasingly precise and individual. This allows for the development of innovative business models. The multiplication of application areas is driven by, among other things, the increasing use of sensors and IoT devices in several domains, as denoted in the previous subsection (e.g., 61.8% of the UCR datasets are sensor data or simulated data) and the fact that increasing researchers and practitioners are using time series to model non-temporal observations [21] (e.g., 38.2% of the UCR datasets are data from images). The most common methods are 1-NN Euclidian distance (ED) and 1-NN Dynamic Time
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
