Pavel Stasa
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
Research Subjects
Pavel Stasa has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
- Filip Benes
VŚB-Technical University of Ostrava
24 shared publications - Jiri Svub
VŚB-Technical University of Ostrava
22 shared publications Vladimir Kebo
The Institute of Economics and Control Systems, VŠB - Technical University of Ostrava, Faculty of Mining and Geology, 17. listopadu 15, 708 33 Poruba, Czech Republic
6 shared publications- Ganjar Alfian
Universitas Gadjah Mada
5 shared publications Jakub Unucka
GABEN, spol. s r.o., Hajkova 558/1, Ostrava, Czech Republic
4 shared publications- Jongtae Rhee
Industrial Artificial Intelligence Researcher Center, Dongguk University-Seoul, 30 Pildongro 1-gil, Jung-gu, Seoul 04620, Republic of Korea
4 shared publications
Latest Publications
- 19 September 2026
- Published by Springer Nature
- p. 303-322
Design of an AirSim-Based Virtual Testbed for Pre-programmed UAV Waypoint Missions in Industrial Indoor Environments
- 19 September 2026
- Published by Springer Nature
- p. 303-322
Design of an AirSim-Based Virtual Testbed for Pre-programmed UAV Waypoint Missions in Industrial Indoor Environments
- 19 September 2026
- Published by Springer Nature
- p. 195-211
Model-Based CO2-Driven Environmental Control for Proactive Safety in Raw Material Industry Indoor Workplaces
- 19 September 2026
- Published by Springer Nature
- p. 195-211
Model-Based CO2-Driven Environmental Control for Proactive Safety in Raw Material Industry Indoor Workplaces
- 26 May 2026
- Published by SAGE Publications in International Journal of RF Technologies
- Vol. 16 (2) , 136-161
A Novel Lightweight RFID Reader for UAV-Based Inventory: Design and Performance Evaluation
This paper presents the design, integration, and experimental validation of a novel lightweight RFID reader specifically developed for Unmanned Aerial Vehicles (UAVs). The system was engineered to meet the strict constraints of aerial platforms, including weight, power consumption, and communication reliability. The resulting 1 kg reader, integrated with a DJI Matrice 600 drone, maintained full flight endurance and demonstrated robust performance at altitudes up to 10 meters. A series of 50 experimental flyovers in a warehouse-style corridor confirmed that tag orientation significantly affects read success: top-mounted tags achieved nearly 30 reads per pass, while side-mounted and partially covered tags showed reductions of approximately 50% and 65%, respectively. These findings highlight the critical role of tag visibility and provide practical guidance for flight-path planning and tag placement in logistics environments. Unlike previous studies that focused on isolated technical aspects or specific applications, this work offers a comprehensive evaluation of UAV-based RFID systems, combining hardware innovation, rigorous performance testing, and broad applicability. The results confirm the feasibility of aerial RFID inventory and lay the groundwork for scalable, autonomous asset tracking in complex industrial settings.
- 26 May 2026
- Published by SAGE Publications in International Journal of RF Technologies
- Vol. 16 (2) , 136-161
A Novel Lightweight RFID Reader for UAV-Based Inventory: Design and Performance Evaluation
This paper presents the design, integration, and experimental validation of a novel lightweight RFID reader specifically developed for Unmanned Aerial Vehicles (UAVs). The system was engineered to meet the strict constraints of aerial platforms, including weight, power consumption, and communication reliability. The resulting 1 kg reader, integrated with a DJI Matrice 600 drone, maintained full flight endurance and demonstrated robust performance at altitudes up to 10 meters. A series of 50 experimental flyovers in a warehouse-style corridor confirmed that tag orientation significantly affects read success: top-mounted tags achieved nearly 30 reads per pass, while side-mounted and partially covered tags showed reductions of approximately 50% and 65%, respectively. These findings highlight the critical role of tag visibility and provide practical guidance for flight-path planning and tag placement in logistics environments. Unlike previous studies that focused on isolated technical aspects or specific applications, this work offers a comprehensive evaluation of UAV-based RFID systems, combining hardware innovation, rigorous performance testing, and broad applicability. The results confirm the feasibility of aerial RFID inventory and lay the groundwork for scalable, autonomous asset tracking in complex industrial settings.
- Open Access
- 30 April 2026
- Published by Walter de Gruyter GmbH in Management Systems in Production Engineering
- Vol. 34 (2) , 282-289
The Proactive Maintenance Strategy and its impact on the Performance in Public Transport
The availability of public transport is one of the most important strategic goals of transport companies in terms of mobility, quality, safety, and environmental sustainability. Satisfying customer requirements for mobility and sustainability, while also achieving an acceptable level of environmental protection and quality of life, is the main goal of public transport. The breakdown of public transport vehicles represents a fundamental barrier to meeting the goals of public transport sustainability. The main goal of the article is to optimize the maintenance process in a transport company in Slovakia with an orientation to cost reduction from the perspective of the ineffective maintenance process. An important role in the elimination of faults is played by the maintenance process of public transport vehicles and the setting of a maintenance strategy in terms of minimizing costs and the number of vehicle faults. The results of economic analyses pointed to increasing costs for the maintenance process, increasing vehicle kilometers, increasing maintenance costs per vehicle kilometer, and increasing spare parts costs for public transport vehicles. Improving the economic and technical parameters of public transport vehicles makes it possible to increase the overall performance of the transport company in Slovakia. The proactive maintenance enriched with autonomous and predictive elements creates the basis for maintenance strategy in public transport with environmental, social, and economic sustainability.
- 30 April 2026
- Published by Walter de Gruyter GmbH in Management Systems in Production Engineering
- Vol. 34 (2) , 282-289
Open AccessThe Proactive Maintenance Strategy and its impact on the Performance in Public Transport
The availability of public transport is one of the most important strategic goals of transport companies in terms of mobility, quality, safety, and environmental sustainability. Satisfying customer requirements for mobility and sustainability, while also achieving an acceptable level of environmental protection and quality of life, is the main goal of public transport. The breakdown of public transport vehicles represents a fundamental barrier to meeting the goals of public transport sustainability. The main goal of the article is to optimize the maintenance process in a transport company in Slovakia with an orientation to cost reduction from the perspective of the ineffective maintenance process. An important role in the elimination of faults is played by the maintenance process of public transport vehicles and the setting of a maintenance strategy in terms of minimizing costs and the number of vehicle faults. The results of economic analyses pointed to increasing costs for the maintenance process, increasing vehicle kilometers, increasing maintenance costs per vehicle kilometer, and increasing spare parts costs for public transport vehicles. Improving the economic and technical parameters of public transport vehicles makes it possible to increase the overall performance of the transport company in Slovakia. The proactive maintenance enriched with autonomous and predictive elements creates the basis for maintenance strategy in public transport with environmental, social, and economic sustainability.
- Open Access
- 1 February 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15 (1)
Web-based geothermal drilling stuck pipe prediction using decision tree algorithm
In geothermal drilling operations, data from rig-mounted sensors play a crucial role in maintaining operational efficiency and preventing drilling failures. However, sensor uncertainties and complex subsurface conditions can lead to stuck pipe incidents, causing significant non-productive time and financial losses. This study proposes web-based drilling monitoring system integrated with machine learning (ML) to predict stuck pipe occurrences in geothermal drilling. Several ML algorithms—decision tree (DT), random forest (RF), naïve Bayes (NB), multilayer perceptron (MLP), and support vector machine (SVM)—were evaluated using geothermal drilling data from an Indonesian geothermal project conducted in 2023. To address class imbalance, the synthetic minority oversampling technique (SMOTE) was applied to the training dataset. Feature selection was performed using the correlation coefficient method, and predictions were generated using a 5 minute sliding window. Among the evaluated models, the DT consistently demonstrated superior performance across multiple prediction horizons (PH), achieving an accuracy of 97.4%, precision of 98.6%, recall of 72.9%, and a ROC-AUC of 0.729 using the top five selected features. The trained model was integrated into web-based monitoring platform that provides visualization and predictive alerts. This system enables early detection and better decision-making, helping improve drilling efficiency, reduce stuck pipe risks, and enhance operational safety.
- 1 February 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15 (1)
Open AccessWeb-based geothermal drilling stuck pipe prediction using decision tree algorithm
In geothermal drilling operations, data from rig-mounted sensors play a crucial role in maintaining operational efficiency and preventing drilling failures. However, sensor uncertainties and complex subsurface conditions can lead to stuck pipe incidents, causing significant non-productive time and financial losses. This study proposes web-based drilling monitoring system integrated with machine learning (ML) to predict stuck pipe occurrences in geothermal drilling. Several ML algorithms—decision tree (DT), random forest (RF), naïve Bayes (NB), multilayer perceptron (MLP), and support vector machine (SVM)—were evaluated using geothermal drilling data from an Indonesian geothermal project conducted in 2023. To address class imbalance, the synthetic minority oversampling technique (SMOTE) was applied to the training dataset. Feature selection was performed using the correlation coefficient method, and predictions were generated using a 5 minute sliding window. Among the evaluated models, the DT consistently demonstrated superior performance across multiple prediction horizons (PH), achieving an accuracy of 97.4%, precision of 98.6%, recall of 72.9%, and a ROC-AUC of 0.729 using the top five selected features. The trained model was integrated into web-based monitoring platform that provides visualization and predictive alerts. This system enables early detection and better decision-making, helping improve drilling efficiency, reduce stuck pipe risks, and enhance operational safety.
- 1 February 2026
- Published by Springer Nature
- p. 143-159
Application of the DJI Mavic 3 Drone for Photogrammetric Modeling of Surface Mines
- 1 February 2026
- Published by Springer Nature
- p. 143-159
Application of the DJI Mavic 3 Drone for Photogrammetric Modeling of Surface Mines
- 1 January 2026
- Published by Rzeszow University of Technology in Technologia i Automatyzacja Montażu
- Vol. 132 (2) , 49-57
Experimental Evaluation of Vision-Based Inspection Performance under Different Illumination Conditions in Automated Manufacturing
Machine vision systems are of paramount importance in the context of contemporary industrial quality control, wherein the reliability of object detection is contingent upon the conditions of image acquisition. It is evident that illumination is a pivotal factor in the context of inspection performance, exerting a substantial influence on factors such as image contrast, feature visibility, and detection accuracy. The present study investigates the influence of multispectral illumination on the detection of coloured polymer samples using a Keyence intelligent vision system. The experimental investigation was conducted using geometrically identical samples manufactured from the same material but differing in surface colour. Four colour variants, namely white, black, violet, and orange, were evaluated under seven illumination conditions, including white, red, green, blue, infrared (IR), ultraviolet (UV), and ambient mode illumination. The acquired images were then subjected to analysis with respect to the visibility of geometric features, contour sharpness, and overall detection reliability. The findings of the present study demonstrated that the illumination wavelength exerts a significant influence on the detectability of objects and the quality of the resulting images. White illumination was found to provide the most consistent performance across all the samples that were tested. In addition, specific combinations of object colour and illumination spectrum were found to produce enhanced feature visibility. Infrared illumination exhibited favourable outcomes for surfaces of a darker hue, whilst ultraviolet illumination enhanced the visibility of specific features on samples of a coloured nature. The findings presented herein corroborate the significance of illumination selection in industrial machine vision applications and furnish pragmatic recommendations for the optimisation of inspection conditions when employing intelligent vision sensors. The findings may inform the development of more reliable automated quality control systems in Industry 4.0 and Industry 5.0 environments.
- 1 January 2026
- Published by Rzeszow University of Technology in Technologia i Automatyzacja Montażu
- Vol. 132 (2) , 49-57
Experimental Evaluation of Vision-Based Inspection Performance under Different Illumination Conditions in Automated Manufacturing
Machine vision systems are of paramount importance in the context of contemporary industrial quality control, wherein the reliability of object detection is contingent upon the conditions of image acquisition. It is evident that illumination is a pivotal factor in the context of inspection performance, exerting a substantial influence on factors such as image contrast, feature visibility, and detection accuracy. The present study investigates the influence of multispectral illumination on the detection of coloured polymer samples using a Keyence intelligent vision system. The experimental investigation was conducted using geometrically identical samples manufactured from the same material but differing in surface colour. Four colour variants, namely white, black, violet, and orange, were evaluated under seven illumination conditions, including white, red, green, blue, infrared (IR), ultraviolet (UV), and ambient mode illumination. The acquired images were then subjected to analysis with respect to the visibility of geometric features, contour sharpness, and overall detection reliability. The findings of the present study demonstrated that the illumination wavelength exerts a significant influence on the detectability of objects and the quality of the resulting images. White illumination was found to provide the most consistent performance across all the samples that were tested. In addition, specific combinations of object colour and illumination spectrum were found to produce enhanced feature visibility. Infrared illumination exhibited favourable outcomes for surfaces of a darker hue, whilst ultraviolet illumination enhanced the visibility of specific features on samples of a coloured nature. The findings presented herein corroborate the significance of illumination selection in industrial machine vision applications and furnish pragmatic recommendations for the optimisation of inspection conditions when employing intelligent vision sensors. The findings may inform the development of more reliable automated quality control systems in Industry 4.0 and Industry 5.0 environments.
- 21 August 2024
- Published by Springer Nature
- p. 141-152
Creating a Virtual Scene Using Photogrammetric 3D Models and Simulating a Security Task
- 21 August 2024
- Published by Springer Nature
- p. 141-152
Creating a Virtual Scene Using Photogrammetric 3D Models and Simulating a Security Task
- 21 August 2024
- Published by Springer Nature
- p. 261-272
The Realization of a Low-Cost System for Entry into Restricted Areas
- 21 August 2024
- Published by Springer Nature
- p. 261-272
The Realization of a Low-Cost System for Entry into Restricted Areas