Benny Drescher
Research Subjects
Benny Drescher has a total of 140 co-authors. Here are the top co-authors by number of shared publications:
- Chang Liu
School of Information Science and Engineering, Shenyang Ligong University, Shenyang, 110159, China
2 shared publications - Muhammad Muaz
College of Aeronautical Engineering, National University of Sciences and Technology, Islamabad, Pakistan
2 shared publications - Wai Lam Sung
Hong Kong Industrial and Artificial Intelligence Centre (FLAIR), Hong Kong, China
2 shared publications - Yan Nei Law
Hong Kong Industrial Artificial Intelligence and Robotics Centre Limited, Kowloon Tong, Hong Kong
2 shared publications Achim Seibertz
Tag des Systems Engineering Paderborn 7. -9
1 shared publication- Akashdeep Roy
Center Connected Industry (CCI), Aachen, Germany
1 shared publication
Latest Publications
- Open Access
- 9 February 2026
- Published by Springer Nature in Process Science
- Vol. 3 (1) , 4
Cargo distribution analysis: a process mining approach
Air cargo terminals are indispensable hubs in the global trade network, yet inefficiencies in their cargo handling processes can trigger significant disruptions, manifesting as costly delays, inflated operational expenses, and a decline in service quality, thereby jeopardizing intricate supply chains. To gain a comprehensive understanding of operational efficiency, we leverage process mining to analyze event data that track the transportation journey of each cargo item from event-driven systems. Following the process mining project methodology, we present an end-to-end pipeline that diagnoses bottlenecks within the cargo distribution process and predicts future distribution flows based on historical event data. Real-world event data are often characterized by inherent complexities, presenting challenges in data quality and scalability. Addressing these challenges, our standardized effort focuses on processing event data from supporting information systems to automatically detect and diagnose the root causes of inefficiencies to deliver actionable, data-driven insights. These insights are integrated into an interactive GUI that supports decision-making, enabling proactive measures and streamlining operations.
- 9 February 2026
- Published by Springer Nature in Process Science
- Vol. 3 (1) , 4
Open AccessCargo distribution analysis: a process mining approach
Air cargo terminals are indispensable hubs in the global trade network, yet inefficiencies in their cargo handling processes can trigger significant disruptions, manifesting as costly delays, inflated operational expenses, and a decline in service quality, thereby jeopardizing intricate supply chains. To gain a comprehensive understanding of operational efficiency, we leverage process mining to analyze event data that track the transportation journey of each cargo item from event-driven systems. Following the process mining project methodology, we present an end-to-end pipeline that diagnoses bottlenecks within the cargo distribution process and predicts future distribution flows based on historical event data. Real-world event data are often characterized by inherent complexities, presenting challenges in data quality and scalability. Addressing these challenges, our standardized effort focuses on processing event data from supporting information systems to automatically detect and diagnose the root causes of inefficiencies to deliver actionable, data-driven insights. These insights are integrated into an interactive GUI that supports decision-making, enabling proactive measures and streamlining operations.
- Open Access
- 26 September 2025
- Published by Springer Nature in Discover Robotics
- Vol. 1 (1) , 1-17
Evaluating mobile robot navigation behavior in flexible assembly systems through digital twin and real-world experiments
- 26 September 2025
- Published by Springer Nature in Discover Robotics
- Vol. 1 (1) , 1-17
Open AccessEvaluating mobile robot navigation behavior in flexible assembly systems through digital twin and real-world experiments
- 1 May 2025
- Published by Elsevier in Journal of Manufacturing Processes
- Vol. 145, 371-380
Explainable AI for correct root cause analysis of product quality in injection moulding
- 1 May 2025
- Published by Elsevier in Journal of Manufacturing Processes
- Vol. 145, 371-380
Explainable AI for correct root cause analysis of product quality in injection moulding
- Top Cited PapersOpen Access
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 44650-44661
LLM-Enhanced Human–Machine Interaction for Adaptive Decision-Making in Dynamic Manufacturing Process Environments
Modern production systems generate vast amounts of process data that hold valuable insights for optimizing manufacturing processes. However, production personnel often face the challenge of interpreting this information, especially when dealing with unexpected anomalies or when insights beyond standard reports are required. This challenge arises both from the complex data structures in which the data is provided, and the lack of analytical expertise. This research introduces an approach that leverages Large Language Models (LLMs) to facilitate natural language queries and flexible data visualization, allowing production personnel to interact effortlessly with complex datasets. Tested on process data from an industrial extrusion process that have been enhanced using data augmentation techniques, the proposed concept demonstrates the capability to retrieve relevant data and present tailored visualizations based on simple user prompts. The results demonstrate that LLM-driven data exploration can support production personnel and help overcome the challenges described, which arise from the complex nature of manufacturing data and the specialized domain knowledge required. Future work will concentrate on improving accuracy, robustness, and further integration of domain-specific knowledge, aiming to provide a more reliable and accessible tool for various industrial environments.
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 44650-44661
Top Cited PapersOpen AccessLLM-Enhanced Human–Machine Interaction for Adaptive Decision-Making in Dynamic Manufacturing Process Environments
Modern production systems generate vast amounts of process data that hold valuable insights for optimizing manufacturing processes. However, production personnel often face the challenge of interpreting this information, especially when dealing with unexpected anomalies or when insights beyond standard reports are required. This challenge arises both from the complex data structures in which the data is provided, and the lack of analytical expertise. This research introduces an approach that leverages Large Language Models (LLMs) to facilitate natural language queries and flexible data visualization, allowing production personnel to interact effortlessly with complex datasets. Tested on process data from an industrial extrusion process that have been enhanced using data augmentation techniques, the proposed concept demonstrates the capability to retrieve relevant data and present tailored visualizations based on simple user prompts. The results demonstrate that LLM-driven data exploration can support production personnel and help overcome the challenges described, which arise from the complex nature of manufacturing data and the specialized domain knowledge required. Future work will concentrate on improving accuracy, robustness, and further integration of domain-specific knowledge, aiming to provide a more reliable and accessible tool for various industrial environments.
- 1 September 2023
- Published by Elsevier in Journal of Manufacturing Processes
- Vol. 103, 238-247
A multitask encoder–decoder model for quality prediction in injection moulding
- 1 September 2023
- Published by Elsevier in Journal of Manufacturing Processes
- Vol. 103, 238-247
A multitask encoder–decoder model for quality prediction in injection moulding
- Open Access
- 1 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 95782-95795
A GAN-Based Multi-Sensor Data Augmentation Technique for CNC Machine Tool Wear Prediction
The prediction of CNC machine tool wear is limited due to the scarcity of data in industry. A sufficient amount of experimental and real-live factory data is missing to train accurate supervised machine learning prediction models. Considering that collecting a large amount of data in real industrial environments can be a great challenge, this paper investigates the applicability of data augmentation techniques for generating synthetic data. Our paper focuses on a multi-sensor approach for the classification of tool wear states in the milling process. Multi-sensor data fusion is performed in the frequency domain. Extracted features are then used for tool wear classification. In order to increase the amount of training data, a generative adversarial network (GAN) is designed for data augmentation purposes. An early stopping strategy is designed to improve the effectiveness of the proposed GAN. The experimental results show that the proposed GAN helps to significantly improve the performance of tool wear classification models. With the inclusion of GAN, the number of required real-environment industrial data can be reduced. The research shows promising results to supplement experimental data by GAN-based synthetic data for predicting tool wear.
- 1 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 95782-95795
Open AccessA GAN-Based Multi-Sensor Data Augmentation Technique for CNC Machine Tool Wear Prediction
The prediction of CNC machine tool wear is limited due to the scarcity of data in industry. A sufficient amount of experimental and real-live factory data is missing to train accurate supervised machine learning prediction models. Considering that collecting a large amount of data in real industrial environments can be a great challenge, this paper investigates the applicability of data augmentation techniques for generating synthetic data. Our paper focuses on a multi-sensor approach for the classification of tool wear states in the milling process. Multi-sensor data fusion is performed in the frequency domain. Extracted features are then used for tool wear classification. In order to increase the amount of training data, a generative adversarial network (GAN) is designed for data augmentation purposes. An early stopping strategy is designed to improve the effectiveness of the proposed GAN. The experimental results show that the proposed GAN helps to significantly improve the performance of tool wear classification models. With the inclusion of GAN, the number of required real-environment industrial data can be reduced. The research shows promising results to supplement experimental data by GAN-based synthetic data for predicting tool wear.
- 25 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-12
Feature Clustering for Open-Set Recognition in LCD Manufacturing
Inspecting defects in LCD manufacturing is of uttermost importance to ensure customer’s satisfaction and reduce time and money losses. Deep learning classification methods rely on closed-set assumption that the classes to predict during operation are the same as the training ones. However, in real-world settings, new unseen classes (defects) often arise. In this work we evaluate the capabilities of state-of-the-art deep learning methods of classifying known and unknown defects on LCD images. Given the limited performance of such methods, we here propose a novel Cluster Error (CE) classifier and a strong-repulsive (SR) training loss for feature clustering to enhance the classification accuracy both on known and unknown defects. Our results on two real-world industrial datasets show the challenges of such task and how our classifier outperforms the other methods.
- 25 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-12
Feature Clustering for Open-Set Recognition in LCD Manufacturing
Inspecting defects in LCD manufacturing is of uttermost importance to ensure customer’s satisfaction and reduce time and money losses. Deep learning classification methods rely on closed-set assumption that the classes to predict during operation are the same as the training ones. However, in real-world settings, new unseen classes (defects) often arise. In this work we evaluate the capabilities of state-of-the-art deep learning methods of classifying known and unknown defects on LCD images. Given the limited performance of such methods, we here propose a novel Cluster Error (CE) classifier and a strong-repulsive (SR) training loss for feature clustering to enhance the classification accuracy both on known and unknown defects. Our results on two real-world industrial datasets show the challenges of such task and how our classifier outperforms the other methods.