Thomas Gittler
Thomas Gittler has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
Konrad Wegener
ETH Zurich, Institute of Machine Tools and Manufacturing (IWF), Zurich, Switzerland
6 shared publicationsLukas Weiss
Inspire AG, Edison EG, Zürich, Switzerland
5 shared publications- Fabian Stoop
Institute of Machine Tools and Manufacturing (IWF), ETH Zürich, 8092, Zürich, Switzerland
2 shared publications - Konrad Wegener
Institute for Machine Tools and Manufacturing, Eidgenössische Technische Hochschule Zurich, 8092 Zurich, Switzerland
2 shared publications Adam Gontarz
1 shared publication- Alisa Rupenyan
ETH Switzerland
1 shared publication
Latest Publications
- 14 September 2023
- Published by Springer Nature
- p. 339-356
Service-Oriented Architecture for Driving Digital Transformation: Insights from a Case Study
- 14 September 2023
- Published by Springer Nature
- p. 339-356
Service-Oriented Architecture for Driving Digital Transformation: Insights from a Case Study
- 4 March 2023
- Published by Springer Nature
- p. 447-462
People, Process, Master Data, Technology: Data-Centric Engineering of Manufacturing Management Systems
- 4 March 2023
- Published by Springer Nature
- p. 447-462
People, Process, Master Data, Technology: Data-Centric Engineering of Manufacturing Management Systems
- Open Access
- 24 May 2021
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 117 (7-8) , 2213-2226
International Conference on Advanced and Competitive Manufacturing Technologies milling tool wear prediction using unsupervised machine learning
Degraded or defect machine components and consumables negatively impact manufacturing quality and productivity. Diagnosing and predicting the wear or degradation status of critical machine components or parts are therefore of general interest. To tackle this challenge, data-driven approaches based on supervised machine learning principles have demonstrated promising results. However, supervised learning models capable of degradation identification require large quantities of data. In practice, run-to-failure data in large amounts is usually not available and expensive to obtain. To overcome this issue, this study proposes an unsupervised learning approach for degradation prognostics of machine tool components and consumables. It uses time series of multi-sensor signal data, which are transformed into a feature representation. The features consist of various characterizations of the time series, allowing to make different signal measurements comparable, and cluster them according to their feature values. The herewith obtained density-based clustering model is used to diagnose and predict the degradation states of components and parts in unknown conditions. The novelty in the proposed approach lies within the identification of continuous component and part degradation states based on unsupervised learning principles. The proposal is verified and demonstrated on an exemplary data set containing a small sample of run-to-failure multi-sensor signals of milling inserts and their corresponding wear state. By the application of the proposed procedure on the exemplary data set, we demonstrate that an unsupervised clustering approach is capable of separating wear data such that meaningful and accurate estimations of the part condition are possible. The advantages are its ability to cope with scarce data sets, its limited engineering and hyperparameter tuning effort, and its straightforward implementation to a multitude of degradation and wear diagnostics scenarios.
- 24 May 2021
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 117 (7-8) , 2213-2226
Open AccessInternational Conference on Advanced and Competitive Manufacturing Technologies milling tool wear prediction using unsupervised machine learning
Degraded or defect machine components and consumables negatively impact manufacturing quality and productivity. Diagnosing and predicting the wear or degradation status of critical machine components or parts are therefore of general interest. To tackle this challenge, data-driven approaches based on supervised machine learning principles have demonstrated promising results. However, supervised learning models capable of degradation identification require large quantities of data. In practice, run-to-failure data in large amounts is usually not available and expensive to obtain. To overcome this issue, this study proposes an unsupervised learning approach for degradation prognostics of machine tool components and consumables. It uses time series of multi-sensor signal data, which are transformed into a feature representation. The features consist of various characterizations of the time series, allowing to make different signal measurements comparable, and cluster them according to their feature values. The herewith obtained density-based clustering model is used to diagnose and predict the degradation states of components and parts in unknown conditions. The novelty in the proposed approach lies within the identification of continuous component and part degradation states based on unsupervised learning principles. The proposal is verified and demonstrated on an exemplary data set containing a small sample of run-to-failure multi-sensor signals of milling inserts and their corresponding wear state. By the application of the proposed procedure on the exemplary data set, we demonstrate that an unsupervised clustering approach is capable of separating wear data such that meaningful and accurate estimations of the part condition are possible. The advantages are its ability to cope with scarce data sets, its limited engineering and hyperparameter tuning effort, and its straightforward implementation to a multitude of degradation and wear diagnostics scenarios.
- Open Access
- 2 September 2020
- Published by MDPI AG in Journal of Manufacturing and Materials Processing
- Vol. 4 (3) , 86
Machine Tool Component Health Identification with Unsupervised Learning
Unforeseen machine tool component failures cause considerable losses. This study presents a new approach to unsupervised machine component condition identification. It uses test cycle data of machine components in healthy and various faulty conditions for modelling. The novelty in the approach consists of the time series representation as features, the filtering of the features for statistical significance, and the use of this feature representation to train a clustering model. The benefit in the proposed approach is its small engineering effort, the potential for automation, the small amount of data necessary for training and updating the model, and the potential to distinguish between multiple known and unknown conditions. Online measurements on machines in unknown conditions are performed to predict the component condition with the aid of the trained model. The approach was exemplarily tested and verified on different healthy and faulty states of a grinding machine axis. For the accurate classification of the component condition, different clustering algorithms were evaluated and compared. The proposed solution demonstrated encouraging results as it accurately classified the component condition. It requires little data, is straightforward to implement and update, and is able to precisely differentiate minor differences of faults in test cycle time series.
- 2 September 2020
- Published by MDPI AG in Journal of Manufacturing and Materials Processing
- Vol. 4 (3) , 86
Open AccessMachine Tool Component Health Identification with Unsupervised Learning
Unforeseen machine tool component failures cause considerable losses. This study presents a new approach to unsupervised machine component condition identification. It uses test cycle data of machine components in healthy and various faulty conditions for modelling. The novelty in the approach consists of the time series representation as features, the filtering of the features for statistical significance, and the use of this feature representation to train a clustering model. The benefit in the proposed approach is its small engineering effort, the potential for automation, the small amount of data necessary for training and updating the model, and the potential to distinguish between multiple known and unknown conditions. Online measurements on machines in unknown conditions are performed to predict the component condition with the aid of the trained model. The approach was exemplarily tested and verified on different healthy and faulty states of a grinding machine axis. For the accurate classification of the component condition, different clustering algorithms were evaluated and compared. The proposed solution demonstrated encouraging results as it accurately classified the component condition. It requires little data, is straightforward to implement and update, and is able to precisely differentiate minor differences of faults in test cycle time series.
- Open Access
- 1 January 2020
- Published by Elsevier in Procedia CIRP
- Vol. 88, 358-363
Condition monitoring system for machine tool auxiliaries
- 1 January 2020
- Published by Elsevier in Procedia CIRP
- Vol. 88, 358-363
Open AccessCondition monitoring system for machine tool auxiliaries
- 1 November 2019
- Published by MM Publishing, s.r.o. in MM Science Journal
- Vol. 2019 (04) , 3192-3199
SENSORS AS AN ENABLER FOR SELF-OPTIMIZING GRINDING MACHINES
Today, an operator performs experiments to adaptively select grinding process parameters using observations, expert knowledge, and rules of thumb. Self-optimizing grinding machines cannot use operator observations and must therefore extract enough information out of the grinding process. In this study, a holistic sensor set-up as foundation for self-optimizing machines are presented exemplarily for cup wheel grinding machines. In-process detection of grinding burn, based on temperature and gas measurements, is tested and compared. Afterwards, the influence of input variables such as feed rate and cutting speed on grinding cost, grinding burn, and surface roughness are investigated.
- 1 November 2019
- Published by MM Publishing, s.r.o. in MM Science Journal
- Vol. 2019 (04) , 3192-3199
SENSORS AS AN ENABLER FOR SELF-OPTIMIZING GRINDING MACHINES
Today, an operator performs experiments to adaptively select grinding process parameters using observations, expert knowledge, and rules of thumb. Self-optimizing grinding machines cannot use operator observations and must therefore extract enough information out of the grinding process. In this study, a holistic sensor set-up as foundation for self-optimizing machines are presented exemplarily for cup wheel grinding machines. In-process detection of grinding burn, based on temperature and gas measurements, is tested and compared. Afterwards, the influence of input variables such as feed rate and cutting speed on grinding cost, grinding burn, and surface roughness are investigated.
- 1 January 2019
- Published by Inderscience Publishers in International Journal of Mechatronics and Manufacturing Systems
- Vol. 12 (3/4) , 344
Smart factory equipment integration through standardised OPC UA communication with companion specifications and equipment specific information models
Industry 4.0 promotes digitalisation of manufacturing organisations into smart factories. Communication and connectivity of an operational data collection are required principal. OPC Unified Architecture and standards like ISA95 are key resources to develop automated interfaces between equipment and manufacturing execution system (MES). The target is to integrate complex equipment in a convenient way through the use of a companion specification, a dynamic address space and a OPC UA client characterisation process. Challenges are the reconciliation of the right programming languages, the integration of generic, domain agnostic, domain specific and equipment specific information models. This study describes the generic approach and demonstrates its applicability in a complex manufacturing environment, serving as a blueprint for other environments. The outlined concept is based on C#/LabView source-code developed for semiconductor back-end equipment, which interacts with an MES. The proposed system is validated for a final-testing-unit in the production of kW-class high-power laser diodes.
- 1 January 2019
- Published by Inderscience Publishers in International Journal of Mechatronics and Manufacturing Systems
- Vol. 12 (3/4) , 344
Smart factory equipment integration through standardised OPC UA communication with companion specifications and equipment specific information models
Industry 4.0 promotes digitalisation of manufacturing organisations into smart factories. Communication and connectivity of an operational data collection are required principal. OPC Unified Architecture and standards like ISA95 are key resources to develop automated interfaces between equipment and manufacturing execution system (MES). The target is to integrate complex equipment in a convenient way through the use of a companion specification, a dynamic address space and a OPC UA client characterisation process. Challenges are the reconciliation of the right programming languages, the integration of generic, domain agnostic, domain specific and equipment specific information models. This study describes the generic approach and demonstrates its applicability in a complex manufacturing environment, serving as a blueprint for other environments. The outlined concept is based on C#/LabView source-code developed for semiconductor back-end equipment, which interacts with an MES. The proposed system is validated for a final-testing-unit in the production of kW-class high-power laser diodes.
- Open Access
- 1 January 2019
- Published by Elsevier in Procedia CIRP
- Vol. 79, 586-591
A fundamental approach for data acquisition on machine tools as enabler for analytical Industrie 4.0 applications
- 1 January 2019
- Published by Elsevier in Procedia CIRP
- Vol. 79, 586-591
Open AccessA fundamental approach for data acquisition on machine tools as enabler for analytical Industrie 4.0 applications
- Open Access
- 1 January 2019
- Published by Elsevier in Procedia CIRP
- Vol. 79, 125-130
Towards predictive quality management in assembly systems with low quality low quantity data – a methodological approach
- 1 January 2019
- Published by Elsevier in Procedia CIRP
- Vol. 79, 125-130
Open AccessTowards predictive quality management in assembly systems with low quality low quantity data – a methodological approach
- Open Access
- 1 January 2018
- Published by Elsevier in Procedia CIRP
- Vol. 77, 1-17
Dawn of new machining concepts:
- 1 January 2018
- Published by Elsevier in Procedia CIRP
- Vol. 77, 1-17
Open AccessDawn of new machining concepts: