Fouad Bahrpeyma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Fouad Bahrpeyma has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Dirk Reichelt
HTW Dresden
9 shared publications - David Heik
Faculty of Informatics/Mathematics
6 shared publications - Andrew McCarren
Dublin City University
2 shared publications Hassan Haghighi
2 shared publications- Ingolf Gehrhardt
HTW Dresden University of Applied Sciences
2 shared publications - Mark Roantree
Dublin City University
2 shared publications
Latest Publications
- 4 June 2025
- Published by Springer Nature
- p. 172-183
A Reference Architecture for Advanced QoS-Based Service Selection in SOA-Based SoS Architectures
- 4 June 2025
- Published by Springer Nature
- p. 172-183
A Reference Architecture for Advanced QoS-Based Service Selection in SOA-Based SoS Architectures
- Open Access
- 19 October 2024
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 77, 525-557
Study on the application of single-agent and multi-agent reinforcement learning to dynamic scheduling in manufacturing environments with growing complexity: Case study on the synthesis of an industrial IoT Test Bed
- 19 October 2024
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 77, 525-557
Open AccessStudy on the application of single-agent and multi-agent reinforcement learning to dynamic scheduling in manufacturing environments with growing complexity: Case study on the synthesis of an industrial IoT Test Bed
- Open Access
- 9 July 2024
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 20 (3) , 2291-2303
A Meta-learner approach to multistep-ahead time series prediction
The utilization of machine learning has become ubiquitous in addressing contemporary challenges in data science. Moreover, there has been significant interest in democratizing the decision-making process for selecting machine learning algorithms, achieved through the incorporation of meta-features and automated machine learning techniques for both classification and regression tasks. However, this paradigm has not been readily applied to multistep-ahead time series prediction problems. Unlike regression and classification problems, which utilize independent variables not derived from the target variable, time series models typically rely on past values of the series to forecast future outcomes. The structure of a time series is often characterized by features such as trend, seasonality, cyclicality and irregularity. In our study, we illustrate how time series metrics representing these features, in conjunction with an ensemble-based regression Meta-Learner, were employed to predict the standardized mean square error of candidate time series prediction models. Our experiments utilized datasets covering a broad feature space, facilitating the selection of the most effective model by researchers. A rigorous evaluation was conducted to assess the performance of the Meta-Learner on both synthetic and real time series data.
- 9 July 2024
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 20 (3) , 2291-2303
Open AccessA Meta-learner approach to multistep-ahead time series prediction
The utilization of machine learning has become ubiquitous in addressing contemporary challenges in data science. Moreover, there has been significant interest in democratizing the decision-making process for selecting machine learning algorithms, achieved through the incorporation of meta-features and automated machine learning techniques for both classification and regression tasks. However, this paradigm has not been readily applied to multistep-ahead time series prediction problems. Unlike regression and classification problems, which utilize independent variables not derived from the target variable, time series models typically rely on past values of the series to forecast future outcomes. The structure of a time series is often characterized by features such as trend, seasonality, cyclicality and irregularity. In our study, we illustrate how time series metrics representing these features, in conjunction with an ensemble-based regression Meta-Learner, were employed to predict the standardized mean square error of candidate time series prediction models. Our experiments utilized datasets covering a broad feature space, facilitating the selection of the most effective model by researchers. A rigorous evaluation was conducted to assess the performance of the Meta-Learner on both synthetic and real time series data.
- 19 April 2024
- Published by Springer Nature
- p. 413-432
Solving a Dynamic Scheduling Problem for a Manufacturing System with Reinforcement Learning
- 19 April 2024
- Published by Springer Nature
- p. 413-432
Solving a Dynamic Scheduling Problem for a Manufacturing System with Reinforcement Learning
- 15 February 2024
- Published by Springer Nature
- p. 237-254
Application of Multi-agent Reinforcement Learning to the Dynamic Scheduling Problem in Manufacturing Systems
- 15 February 2024
- Published by Springer Nature
- p. 237-254
Application of Multi-agent Reinforcement Learning to the Dynamic Scheduling Problem in Manufacturing Systems
- 3 June 2023
- Published by Springer Nature
- p. 523-533
Dynamic Job Shop Scheduling in an Industrial Assembly Environment Using Various Reinforcement Learning Techniques
- 3 June 2023
- Published by Springer Nature
- p. 523-533
Dynamic Job Shop Scheduling in an Industrial Assembly Environment Using Various Reinforcement Learning Techniques
- Open Access
- 1 March 2023
- Published by HTWK Leipzig
Anwendung von Reinforcement Learning in industriellen cyberphysischen Systemen
Die wachsende Besorgnis über den Klimawandel bewirkt, dass bei der Festlegung von Meilensteinen und Produktionsparametern für Fertigungssysteme ein größerer Fokus auf Energieeinsparung, Ressourcenschonung und Wandlungsfähigkeit gelegt wird. Dies ist auch notwendig, denn die Energiepreise steigen kontinuierlich an und durch den weltweiten Ausbruch von COVID-19 wurden die Lieferketten immer wieder gestört. Gleichzeitig hat die hohe Volatilität und Dynamik innerhalb der globalen Wertschöpfungsnetzwerke in letzter Zeit zu einer spürbaren Verkürzung der Produkt- und Technologiezyklen geführt. Darüber hinaus wandeln sich die Wünsche der Kunden, diese werden zunehmend anspruchsvoller und spezifischer. Um diesen Marktanforderungen gerecht zu werden, mussten und müssen Unternehmen ihre Fertigungsprozesse immer wieder anpassen und abwandeln. Ein wachsendes Portfolio und sinkende Auftragsvolumina (bis hin zur Losgröße 1) ergeben sich im Resultat. Um einen effektiven und effizienten Produktionsablauf zu realisieren, bedarf es ein dynamisches Regelwerk. Dieses bestimmt für die bestehenden Montagelinien und deren Fertigungszellen, unter Be- rücksichtigung des aktuellen Zustandes, auf welchen Stationen ein konkretes Produkt gefertigt werden soll. Gegenwärtig geschieht dies meist statisch über ein Manufacturing Execution System, welches für ganze Chargen entscheidet und üblicherweise nicht (oder nur schwerfällig) auf Unsicherheiten wie den Ausfall einer Operation, den Schwankungen in den Operationszeiten oder in der Qualität des Rohmaterials reagieren kann. Mit einem besonderen Fokus auf die Reduzierung der Gesamtproduktionszeit (und dem damit einhergehenden Energieverbrauch) eines Auftrags stellen wir in diesem Beitrag eine Simulationsumgebung vor, die eine Montagelinie des Industrial IoT Test Bed (an der HTW Dresden) nachbildet. Auf diese wenden wir vielversprechende Reinforcement Learning Methoden an, um für unterschiedliche Szenarien automatisiert eine kosteneffiziente Ressourcenzuweisung zu realisieren.
- 1 March 2023
- Published by HTWK Leipzig
Open AccessAnwendung von Reinforcement Learning in industriellen cyberphysischen Systemen
Die wachsende Besorgnis über den Klimawandel bewirkt, dass bei der Festlegung von Meilensteinen und Produktionsparametern für Fertigungssysteme ein größerer Fokus auf Energieeinsparung, Ressourcenschonung und Wandlungsfähigkeit gelegt wird. Dies ist auch notwendig, denn die Energiepreise steigen kontinuierlich an und durch den weltweiten Ausbruch von COVID-19 wurden die Lieferketten immer wieder gestört. Gleichzeitig hat die hohe Volatilität und Dynamik innerhalb der globalen Wertschöpfungsnetzwerke in letzter Zeit zu einer spürbaren Verkürzung der Produkt- und Technologiezyklen geführt. Darüber hinaus wandeln sich die Wünsche der Kunden, diese werden zunehmend anspruchsvoller und spezifischer. Um diesen Marktanforderungen gerecht zu werden, mussten und müssen Unternehmen ihre Fertigungsprozesse immer wieder anpassen und abwandeln. Ein wachsendes Portfolio und sinkende Auftragsvolumina (bis hin zur Losgröße 1) ergeben sich im Resultat. Um einen effektiven und effizienten Produktionsablauf zu realisieren, bedarf es ein dynamisches Regelwerk. Dieses bestimmt für die bestehenden Montagelinien und deren Fertigungszellen, unter Be- rücksichtigung des aktuellen Zustandes, auf welchen Stationen ein konkretes Produkt gefertigt werden soll. Gegenwärtig geschieht dies meist statisch über ein Manufacturing Execution System, welches für ganze Chargen entscheidet und üblicherweise nicht (oder nur schwerfällig) auf Unsicherheiten wie den Ausfall einer Operation, den Schwankungen in den Operationszeiten oder in der Qualität des Rohmaterials reagieren kann. Mit einem besonderen Fokus auf die Reduzierung der Gesamtproduktionszeit (und dem damit einhergehenden Energieverbrauch) eines Auftrags stellen wir in diesem Beitrag eine Simulationsumgebung vor, die eine Montagelinie des Industrial IoT Test Bed (an der HTW Dresden) nachbildet. Auf diese wenden wir vielversprechende Reinforcement Learning Methoden an, um für unterschiedliche Szenarien automatisiert eine kosteneffiziente Ressourcenzuweisung zu realisieren.
- Open Access
- 1 December 2022
- Published by Frontiers Media SA in Frontiers in Robotics and AI
- Vol. 9, 1027340
A review of the applications of multi-agent reinforcement learning in smart factories
The smart factory is at the heart of Industry 4.0 and is the new paradigm for establishing advanced manufacturing systems and realizing modern manufacturing objectives such as mass customization, automation, efficiency, and self-organization all at once. Such manufacturing systems, however, are characterized by dynamic and complex environments where a large number of decisions should be made for smart components such as production machines and the material handling system in a real-time and optimal manner. AI offers key intelligent control approaches in order to realize efficiency, agility, and automation all at once. One of the most challenging problems faced in this regard is uncertainty, meaning that due to the dynamic nature of the smart manufacturing environments, sudden seen or unseen events occur that should be handled in real-time. Due to the complexity and high-dimensionality of smart factories, it is not possible to predict all the possible events or prepare appropriate scenarios to respond. Reinforcement learning is an AI technique that provides the intelligent control processes needed to deal with such uncertainties. Due to the distributed nature of smart factories and the presence of multiple decision-making components, multi-agent reinforcement learning (MARL) should be incorporated instead of single-agent reinforcement learning (SARL), which, due to the complexities involved in the development process, has attracted less attention. In this research, we will review the literature on the applications of MARL to tasks within a smart factory and then demonstrate a mapping connecting smart factory attributes to the equivalent MARL features, based on which we suggest MARL to be one of the most effective approaches for implementing the control mechanism for smart factories.
- 1 December 2022
- Published by Frontiers Media SA in Frontiers in Robotics and AI
- Vol. 9, 1027340
Open AccessA review of the applications of multi-agent reinforcement learning in smart factories
The smart factory is at the heart of Industry 4.0 and is the new paradigm for establishing advanced manufacturing systems and realizing modern manufacturing objectives such as mass customization, automation, efficiency, and self-organization all at once. Such manufacturing systems, however, are characterized by dynamic and complex environments where a large number of decisions should be made for smart components such as production machines and the material handling system in a real-time and optimal manner. AI offers key intelligent control approaches in order to realize efficiency, agility, and automation all at once. One of the most challenging problems faced in this regard is uncertainty, meaning that due to the dynamic nature of the smart manufacturing environments, sudden seen or unseen events occur that should be handled in real-time. Due to the complexity and high-dimensionality of smart factories, it is not possible to predict all the possible events or prepare appropriate scenarios to respond. Reinforcement learning is an AI technique that provides the intelligent control processes needed to deal with such uncertainties. Due to the distributed nature of smart factories and the presence of multiple decision-making components, multi-agent reinforcement learning (MARL) should be incorporated instead of single-agent reinforcement learning (SARL), which, due to the complexities involved in the development process, has attracted less attention. In this research, we will review the literature on the applications of MARL to tasks within a smart factory and then demonstrate a mapping connecting smart factory attributes to the equivalent MARL features, based on which we suggest MARL to be one of the most effective approaches for implementing the control mechanism for smart factories.