Advances in Machine Learning Detecting Changeover Processes in Cyber Physical Production Systems

Abstract

The performance indicator, Overall Equipment Effectiveness (OEE), is one of the most important ones for production control, as it merges information of equipment usage, process yield, and product quality. The determination of the OEE is oftentimes not transparent in companies, due to the heterogeneous data sources and manual interference. Furthermore, there is a difference in present guidelines to calculate the OEE. Due to a big amount of sensor data in Cyber Physical Production Systems, Machine Learning methods can be used in order to detect several elements of the OEE by a trained model. Changeover time is one crucial aspect influencing the OEE, as it adds no value to the product. Furthermore, changeover processes are fulfilled manually and vary from worker to worker. They always have their own procedure to conduct a changeover of a machine for a new product or production lot. Hence, the changeover time as well as the process itself vary. Thus, a new Machine Learning based concept for identification and characterization of machine set-up actions is presented. Here, the issue to be dealt with is the necessity of human and machine interaction to fulfill the entire machine set-up process. Because of this, the paper shows the use case in a real production scenario of a small to medium size company (SME), the derived data set, promising Machine Learning algorithms, as well as the results of the implemented Machine Learning model to classify machine set-up actions.

Enhanced Changeover Detection in Industry 4.0 Environments with Machine Learning

“…In conclusion, it was pointed out that a detection of changeover phases with a heterogeneous sensor setup is feasible, and as a further research goal, the refinement of the classification approach was proposed. On one hand, this implied improvement of the boundary conditions for the machine learning model i.e., the sensor setup, and on the other hand, it indicated the need to make use of more detailed subphases for the general “Changeover” phases [ 3 ].…”

“…To add more flexibility and transparency, the entire programming project was shifted from the MATLAB “Classification Learner” environment to Python Jupyter notebooks. It was possible to reproduce the results from [ 3 ] in a Python environment. All but one of the algorithms were successfully replicated in Python using the scikit-learn library, while the RUSBoosted Decision Tree was built using the imbalanced-learn library, which is fully compatible with scikit-learn API.…”

“…In the former research, the research task was to distinguish between “Changeover” and “Production” with a binary classification in a supervised learning approach [ 3 ]. The supervision is conducted by a researcher who oversees the complete changeover and production process and records timestamps for the specific phases.…”

“…With the intention to support and digitalize the system landscape of these metalworking companies, the OBerA project (OBerA: “Optimization of processes and machine tools through provision, analysis, and target/actual comparison of production data”) was founded. The consortium is formed by five production-oriented companies from the region of Franconia in northern Bavaria, Siemens as a technology partner, and the University of Applied Sciences Würzburg-Schweinfurt as an academic partner (for a detailed description of the consortium, please see [ 3 ]).…”

Cyber-Physical Production Systems (CPPS): Introduction

“…Engelmann et al [1] presented a technical CPPS setup that aims at supporting the operation of production machines-more transparency is achieved here through an automated state recognition with machine learning as a base for OEE (overall equipment effectiveness) calculation.…”