Juraj Kováč
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Juraj Kováč has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Peter Malega
Faculty of Mechanical Engineering, Technical University of Košice, 04200 Košice, Slovakia
9 shared publications - Jozef Svetlík
Department of Production Technology and Robotics, Faculty of Mechanical Engineering, Technical University of Kosice, Letná 9, 040 01 Kosice, Slovakia
8 shared publications - Vladimír Rudy
Tech Univ Kosice, Fac Mech Engn, Inst Management Ind & Digital Engn, Kosice, Slovakia
4 shared publications - Matúš Leščinský
Department of Industrial and Digital Engineering, Faculty of Mechanical Engineering, Technical University of Košice, Park Komenského 9, 042 00 Košice, Slovakia
2 shared publications - Miroslav Štofa
Department of Manufacturing Machinery, Faculty of Mechanical Engineering, The Technical University of Košice, Letná 9, Košice 04001, Slovakia
2 shared publications - Adriana Sekerakova
Faculty of manufacturing technologies with the seat in Presov, Technical University of Kosice, Sturova 31, 080 01 Presov, Slovakia
1 shared publication
Latest Publications
- Open Access
Decision Stability in Operator-Dependent Dimensional Inspection: A Process-Oriented Framework for Data-Driven Quality Control
Operator-dependent dimensional measurements can produce inconsistent categorical quality decisions, particularly near classification boundaries. This study presents an integrated framework combining measurement-system analysis, inter-operator agreement, analytical-threshold sensitivity, and machine learning to characterize decision instability. Six operators each measured the width and height of 25 Charpy V-notch specimens three times using one digital micrometer (900 measurements). Four exploratory symmetric thresholds around 10 mm generated 1200 operator–specimen–dimension–threshold decisions; disagreement was defined relative to the consensus of the other five operators, not certified dimensional ground truth. Overall, 114 decisions (9.5%) disagreed with consensus, with higher disagreement for height (93/600; 15.5%) than width (21/600; 3.5%). Under nested specimen-grouped cross-validation, the extended Random Forest model achieved AUROC 0.908 (95% specimen-cluster bootstrap confidence interval, 0.878–0.940) and PR-AUC 0.565 (0.448–0.668); a transparent boundary-distance baseline achieved 0.889 and 0.445, respectively. Feature-ablation analysis indicated additional predictive information beyond categorical decision and boundary distance, although target-related dependence remained. Lowering the model-output cutoff from 0.50 to 0.15 increased recall from 0.439 to 0.851 at the cost of more verification warnings. Findings support exploratory risk-based screening, while small specimen numbers, shared instrumentation, and lack of independent reference dimensions limit claims about error detection and industrial transferability.
Open AccessDecision Stability in Operator-Dependent Dimensional Inspection: A Process-Oriented Framework for Data-Driven Quality Control
Operator-dependent dimensional measurements can produce inconsistent categorical quality decisions, particularly near classification boundaries. This study presents an integrated framework combining measurement-system analysis, inter-operator agreement, analytical-threshold sensitivity, and machine learning to characterize decision instability. Six operators each measured the width and height of 25 Charpy V-notch specimens three times using one digital micrometer (900 measurements). Four exploratory symmetric thresholds around 10 mm generated 1200 operator–specimen–dimension–threshold decisions; disagreement was defined relative to the consensus of the other five operators, not certified dimensional ground truth. Overall, 114 decisions (9.5%) disagreed with consensus, with higher disagreement for height (93/600; 15.5%) than width (21/600; 3.5%). Under nested specimen-grouped cross-validation, the extended Random Forest model achieved AUROC 0.908 (95% specimen-cluster bootstrap confidence interval, 0.878–0.940) and PR-AUC 0.565 (0.448–0.668); a transparent boundary-distance baseline achieved 0.889 and 0.445, respectively. Feature-ablation analysis indicated additional predictive information beyond categorical decision and boundary distance, although target-related dependence remained. Lowering the model-output cutoff from 0.50 to 0.15 increased recall from 0.439 to 0.851 at the cost of more verification warnings. Findings support exploratory risk-based screening, while small specimen numbers, shared instrumentation, and lack of independent reference dimensions limit claims about error detection and industrial transferability.
- Open Access
- 28 July 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (15) , 7502
Artificial Intelligence-Driven Quality Control in Mechanical Manufacturing: Vibration-Based Multiclass Gear Fault Detection Using LightGBM
Artificial intelligence is increasingly used to improve industrial quality control, but its practical value depends on whether models remain accurate under different operating conditions and fault classes. This study evaluates an artificial-intelligence-based workflow for gear quality control using vibration signals measured on a real two-stage reduction gearbox. Two orthogonal vibration channels were analyzed for six health states, three shaft speeds, and two load levels. Because the time-series data were only partly stationary, the dataset was divided chronologically into training and test segments. A 54-feature representation was built from rolling-window statistics and operating variables, and six classifiers were compared: Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBM), random forest, decision tree, multilayer perceptron (MLP), and logistic regression. LGBM achieved the best overall accuracy (0.9728) while maintaining substantially lower training time than several competing nonlinear models. Class-wise precision, recall, and F1-score ranged from 0.95 to 1.00, and the nominal response time for most operating-condition transitions was approximately 0.0998 s. The results show that vibration-based machine learning can support robust, near-real-time fault identification in mechanical manufacturing environments. The study also highlights the importance of chronological validation, feature engineering over multiple time windows, and the trade-off between predictive performance and deployment efficiency. Because the validation dataset originates from one gearbox platform, the results should be interpreted as promising internal evidence rather than as proof of universal industrial robustness.
- 28 July 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (15) , 7502
Open AccessArtificial Intelligence-Driven Quality Control in Mechanical Manufacturing: Vibration-Based Multiclass Gear Fault Detection Using LightGBM
Artificial intelligence is increasingly used to improve industrial quality control, but its practical value depends on whether models remain accurate under different operating conditions and fault classes. This study evaluates an artificial-intelligence-based workflow for gear quality control using vibration signals measured on a real two-stage reduction gearbox. Two orthogonal vibration channels were analyzed for six health states, three shaft speeds, and two load levels. Because the time-series data were only partly stationary, the dataset was divided chronologically into training and test segments. A 54-feature representation was built from rolling-window statistics and operating variables, and six classifiers were compared: Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBM), random forest, decision tree, multilayer perceptron (MLP), and logistic regression. LGBM achieved the best overall accuracy (0.9728) while maintaining substantially lower training time than several competing nonlinear models. Class-wise precision, recall, and F1-score ranged from 0.95 to 1.00, and the nominal response time for most operating-condition transitions was approximately 0.0998 s. The results show that vibration-based machine learning can support robust, near-real-time fault identification in mechanical manufacturing environments. The study also highlights the importance of chronological validation, feature engineering over multiple time windows, and the trade-off between predictive performance and deployment efficiency. Because the validation dataset originates from one gearbox platform, the results should be interpreted as promising internal evidence rather than as proof of universal industrial robustness.
- Open Access
Development and Laboratory Feasibility Validation of a Virtual Reality Simulation Model for Robotic End-Effector Assembly Training
Virtual reality can support the preparation and rehearsal of assembly tasks by providing a safe and repeatable digital representation of workstations. This study presents the development and laboratory feasibility validation of a geometry- and procedure-oriented VR simulation model for the assembly and disassembly of end-effectors on an industrial robot. The workflow was implemented using the Almega AX-V6 robotic workstation as a case study and included geometric acquisition of the real robot, CAD modelling in SolidWorks, redesign of the original end-effector connection using a quick-change flange concept, creation of two alternative end-effector models, modelling of the laboratory workspace in SketchUp, and scene enhancement in Twinmotion. The resulting robot and environment models were integrated in Pixyz Review and deployed through an Oculus Rift-based VR setup. Compared with the original flange concept, which required twelve screws, the redesigned training concept used two screws and two nuts, reducing the number of fastening elements by 66.7% and the number of screw positions by 83.3%. The VR implementation supported visual inspection, controller-based placement and alignment, and symbolic confirmation of fastening steps; it did not include force feedback, threaded fastening physics, automatic error scoring, or quantified transfer-of-training evaluation. Laboratory feasibility validation confirmed correct asset integration, spatial correspondence with the physical workplace, and functional executability of the target exchange sequence. The results show that the workflow is useful as a case-study pipeline for CAD-to-VR modelling and assembly rehearsal, while controlled user studies are still required before claims about training effectiveness can be made.
Open AccessDevelopment and Laboratory Feasibility Validation of a Virtual Reality Simulation Model for Robotic End-Effector Assembly Training
Virtual reality can support the preparation and rehearsal of assembly tasks by providing a safe and repeatable digital representation of workstations. This study presents the development and laboratory feasibility validation of a geometry- and procedure-oriented VR simulation model for the assembly and disassembly of end-effectors on an industrial robot. The workflow was implemented using the Almega AX-V6 robotic workstation as a case study and included geometric acquisition of the real robot, CAD modelling in SolidWorks, redesign of the original end-effector connection using a quick-change flange concept, creation of two alternative end-effector models, modelling of the laboratory workspace in SketchUp, and scene enhancement in Twinmotion. The resulting robot and environment models were integrated in Pixyz Review and deployed through an Oculus Rift-based VR setup. Compared with the original flange concept, which required twelve screws, the redesigned training concept used two screws and two nuts, reducing the number of fastening elements by 66.7% and the number of screw positions by 83.3%. The VR implementation supported visual inspection, controller-based placement and alignment, and symbolic confirmation of fastening steps; it did not include force feedback, threaded fastening physics, automatic error scoring, or quantified transfer-of-training evaluation. Laboratory feasibility validation confirmed correct asset integration, spatial correspondence with the physical workplace, and functional executability of the target exchange sequence. The results show that the workflow is useful as a case-study pipeline for CAD-to-VR modelling and assembly rehearsal, while controlled user studies are still required before claims about training effectiveness can be made.
- Top Cited PapersOpen Access
- 24 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2175
Augmented Reality as a Tool for Training Assembly Line Workers
Augmented reality (AR) is increasingly adopted in industrial environments as a tool for improving employee training and supporting complex assembly operations. The purpose of this study was to investigate the design, implementation, and strategic potential of AR-based work instructions using Microsoft HoloLens 2 in a real manufacturing environment. The study proposes and applies an integrated evaluation framework combining direct observation, performance evaluation, semi-structured interviews, quantitative SWOT analysis, PDCA-based process assessment, and economic cost analysis to assess AR-based training in a real manufacturing environment. AR training was implemented through Microsoft Dynamics 365 Guides for a standardized assembly procedure and evaluated with respect to training efficiency, user interaction, and feasibility of deployment. The results indicate improved task guidance consistency and descriptive performance indicators, suggesting enhanced training support under real production conditions. The SWOT analysis identified a favorable SO strategic position, highlighting strong internal capabilities and promising external opportunities for further deployment. The cost analysis shows that AR-based training becomes economically advantageous when applied to a larger number of trainees, despite high initial investment costs. Overall, the study demonstrates that AR-based training, when evaluated through a structured strategic and economic framework, represents a promising and strategically advantageous approach for industrial education, provided that ergonomic challenges, user adaptation, and financial constraints are systematically addressed.
- 24 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2175
Top Cited PapersOpen AccessAugmented Reality as a Tool for Training Assembly Line Workers
Augmented reality (AR) is increasingly adopted in industrial environments as a tool for improving employee training and supporting complex assembly operations. The purpose of this study was to investigate the design, implementation, and strategic potential of AR-based work instructions using Microsoft HoloLens 2 in a real manufacturing environment. The study proposes and applies an integrated evaluation framework combining direct observation, performance evaluation, semi-structured interviews, quantitative SWOT analysis, PDCA-based process assessment, and economic cost analysis to assess AR-based training in a real manufacturing environment. AR training was implemented through Microsoft Dynamics 365 Guides for a standardized assembly procedure and evaluated with respect to training efficiency, user interaction, and feasibility of deployment. The results indicate improved task guidance consistency and descriptive performance indicators, suggesting enhanced training support under real production conditions. The SWOT analysis identified a favorable SO strategic position, highlighting strong internal capabilities and promising external opportunities for further deployment. The cost analysis shows that AR-based training becomes economically advantageous when applied to a larger number of trainees, despite high initial investment costs. Overall, the study demonstrates that AR-based training, when evaluated through a structured strategic and economic framework, represents a promising and strategically advantageous approach for industrial education, provided that ergonomic challenges, user adaptation, and financial constraints are systematically addressed.
- Open Access
- 11 May 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (10) , 5358
Reduction of Carbon Footprint in Mechanical Engineering Production Using a Universal Simulation Model
The paper presents the design and development of a universal simulation model named SustainSIM, intended for optimizing the carbon footprint in mechanical engineering production. The objective of this model is to enable enterprises to accurately quantify, monitor, and simulate CO2 emissions generated during various manufacturing processes, thereby identifying and evaluating effective reduction strategies. The paper thoroughly examines methodologies for data collection and processing, determination of emission factors, and categorization of emissions (Scope 1 and Scope 2), utilizing standards such as the GHG Protocol and associated databases. Through a digital simulation environment created in Unity Engine, the model interactively visualizes the impacts of implementing green technologies—such as solar panels, electric vehicles, and heat pumps—on reducing the overall carbon footprint. The practical applicability of the model was validated using a mechanical engineering company as a case study, where simulations confirmed the model’s potential in supporting sustainable decision-making and production process optimization. The findings suggest that the implementation of such a tool can significantly contribute to environmentally responsible management and the reduction of industrial emissions. In comparison to existing methods such as SimaPro/OpenLCA (detailed LCA) and the Corporate Calculator (GHG Protocol), SustainSIM achieves the same accuracy in calculating Scopes 1/2, while reducing the analysis time to less than 15% and decreasing the requirements for expertise. Unlike simulation packages like Energy Plus, users can modify parameters without scripting, and they can see the immediate impact in CO2e.
- 11 May 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (10) , 5358
Open AccessReduction of Carbon Footprint in Mechanical Engineering Production Using a Universal Simulation Model
The paper presents the design and development of a universal simulation model named SustainSIM, intended for optimizing the carbon footprint in mechanical engineering production. The objective of this model is to enable enterprises to accurately quantify, monitor, and simulate CO2 emissions generated during various manufacturing processes, thereby identifying and evaluating effective reduction strategies. The paper thoroughly examines methodologies for data collection and processing, determination of emission factors, and categorization of emissions (Scope 1 and Scope 2), utilizing standards such as the GHG Protocol and associated databases. Through a digital simulation environment created in Unity Engine, the model interactively visualizes the impacts of implementing green technologies—such as solar panels, electric vehicles, and heat pumps—on reducing the overall carbon footprint. The practical applicability of the model was validated using a mechanical engineering company as a case study, where simulations confirmed the model’s potential in supporting sustainable decision-making and production process optimization. The findings suggest that the implementation of such a tool can significantly contribute to environmentally responsible management and the reduction of industrial emissions. In comparison to existing methods such as SimaPro/OpenLCA (detailed LCA) and the Corporate Calculator (GHG Protocol), SustainSIM achieves the same accuracy in calculating Scopes 1/2, while reducing the analysis time to less than 15% and decreasing the requirements for expertise. Unlike simulation packages like Energy Plus, users can modify parameters without scripting, and they can see the immediate impact in CO2e.
- Open Access
- 1 December 2024
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 14 (6) , 17555-17563
Enhancing Filter Assembly and Printing Efficiency through Automation and Virtual Reality Integration
This paper presents a comprehensive approach to improving the assembly and printing process of filters in a selected company. Currently, the company's process involves significant manual labor, which leads to inefficiencies and potential health risks for employees. The proposed solution integrates the use of pneumatic cylinders to automate the assembly and printing process, reducing manual effort and improving production efficiency. The fixture and filter were modeled using SOLIDWORKS 2022 and simulated in virtual reality with Pixyz Review software. The virtual reality simulation serves as an innovative training tool for new employees, enhancing understanding and accuracy in the assembly process. The study compares the current state, where manual handling results in a time-consuming process with a lower production rate, to the proposed automated system. The proposed solution eliminates unnecessary intermediate storage and manual fixture transfers, significantly reducing the total assembly and printing time from 24 s to 22 s per unit. This results in a 14.16 % increase in production efficiency, allowing the company to produce 1227.3 filters per shift compared to the current 1075 filters.
- 1 December 2024
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 14 (6) , 17555-17563
Open AccessEnhancing Filter Assembly and Printing Efficiency through Automation and Virtual Reality Integration
This paper presents a comprehensive approach to improving the assembly and printing process of filters in a selected company. Currently, the company's process involves significant manual labor, which leads to inefficiencies and potential health risks for employees. The proposed solution integrates the use of pneumatic cylinders to automate the assembly and printing process, reducing manual effort and improving production efficiency. The fixture and filter were modeled using SOLIDWORKS 2022 and simulated in virtual reality with Pixyz Review software. The virtual reality simulation serves as an innovative training tool for new employees, enhancing understanding and accuracy in the assembly process. The study compares the current state, where manual handling results in a time-consuming process with a lower production rate, to the proposed automated system. The proposed solution eliminates unnecessary intermediate storage and manual fixture transfers, significantly reducing the total assembly and printing time from 24 s to 22 s per unit. This results in a 14.16 % increase in production efficiency, allowing the company to produce 1227.3 filters per shift compared to the current 1075 filters.
- Open Access
- 6 June 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (11) , 4933
Smart Electric Three-Wheeled Unit for the Manufacturing Industry
This article presents the design of a smart three-wheeled unit for the manufacturing industry with the aim of optimizing and automating internal logistical processes. It presents an innovative solution that combines the advantages of mobility, intelligent transportation technology, and smart devices to ensure the efficient movement of materials and raw materials in manufacturing facilities. The article describes the design, production, and testing of the tricycle in a real manufacturing environment of the production system and the testing of the proposed smart devices. It evaluates the advantages of the electric smart tricycle, including increased efficiency, reduced costs, and more flexible production processes. The results of this study suggest that the intelligent three-wheeled unit represents a promising technological innovation with the potential to increase competitiveness and productivity in manufacturing enterprises.
- 6 June 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (11) , 4933
Open AccessSmart Electric Three-Wheeled Unit for the Manufacturing Industry
This article presents the design of a smart three-wheeled unit for the manufacturing industry with the aim of optimizing and automating internal logistical processes. It presents an innovative solution that combines the advantages of mobility, intelligent transportation technology, and smart devices to ensure the efficient movement of materials and raw materials in manufacturing facilities. The article describes the design, production, and testing of the tricycle in a real manufacturing environment of the production system and the testing of the proposed smart devices. It evaluates the advantages of the electric smart tricycle, including increased efficiency, reduced costs, and more flexible production processes. The results of this study suggest that the intelligent three-wheeled unit represents a promising technological innovation with the potential to increase competitiveness and productivity in manufacturing enterprises.
- Open Access
- 23 January 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (3) , 1496
Vumark’s Method of Production Layout Designing
The paper deals with the issue of the mixed-reality usage in the design of production systems, its changes during expansion, or technological changes in the production, where it is necessary to flexibly and quickly verify the integration of a new machine into the existing layout and eliminate collision situations even before the installation of a physical machine in production. This is realized through Vumark’s design methodology, which was verified and applied in the conditions of the production environment of the Innovation and Prototyping Centre in the Faculty of Mechanical Engineering at the Technical University of Kosice. The individual parts of the paper describe the Vumark deployment procedure in the production area and the software and hardware resources that the user can apply. Three production machines and one robotic device selected from the Factory design software database were chosen for the experiment. The chosen method enables us to verify during the experiment several variants of deployment the machines in the layout and thus to find the optimal location in a very short time. The experiment showed that the chosen method is applicable in practice and brings savings of time, costs, and energy especially when changing the layout or replacing the machine in the production hall.
- 23 January 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (3) , 1496
Open AccessVumark’s Method of Production Layout Designing
The paper deals with the issue of the mixed-reality usage in the design of production systems, its changes during expansion, or technological changes in the production, where it is necessary to flexibly and quickly verify the integration of a new machine into the existing layout and eliminate collision situations even before the installation of a physical machine in production. This is realized through Vumark’s design methodology, which was verified and applied in the conditions of the production environment of the Innovation and Prototyping Centre in the Faculty of Mechanical Engineering at the Technical University of Kosice. The individual parts of the paper describe the Vumark deployment procedure in the production area and the software and hardware resources that the user can apply. Three production machines and one robotic device selected from the Factory design software database were chosen for the experiment. The chosen method enables us to verify during the experiment several variants of deployment the machines in the layout and thus to find the optimal location in a very short time. The experiment showed that the chosen method is applicable in practice and brings savings of time, costs, and energy especially when changing the layout or replacing the machine in the production hall.
- Open Access
- 10 November 2021
- Published by Walter de Gruyter GmbH in Open Engineering
- Vol. 11 (1) , 1086-1092
Design and construction of a scanning stand for the PU mini-acoustic sensor
The main objective of this work was to design and construct a device that would facilitate and reduce the intensity of work in the process of scanning surfaces with a Microflown PU Mini acoustic probe. Our proposed device will shorten the total time to scan the surface and improve the quality of data measurements. The device was designed to be able to perform movement along the X and Y axes, thus achieving scanning of the entire sound field. Some components of the device are manufactured strictly according to particular specifications to suit our requirements. The most important step in the design of the scanning device we prepared for the particular acoustic sensor was the design for the linear connection that fulfills a supporting function for the entire device as well as provides movement along the X and Y axes. The final step was the design for the automation of the entire system, for which we selected an Arduino Uno (ATmega328) controller. The actual device was tested in the final stages. The device is designed and tested as functional and able to work on what was intended for and is ready for further modifications or use in practice.
- 10 November 2021
- Published by Walter de Gruyter GmbH in Open Engineering
- Vol. 11 (1) , 1086-1092
Open AccessDesign and construction of a scanning stand for the PU mini-acoustic sensor
The main objective of this work was to design and construct a device that would facilitate and reduce the intensity of work in the process of scanning surfaces with a Microflown PU Mini acoustic probe. Our proposed device will shorten the total time to scan the surface and improve the quality of data measurements. The device was designed to be able to perform movement along the X and Y axes, thus achieving scanning of the entire sound field. Some components of the device are manufactured strictly according to particular specifications to suit our requirements. The most important step in the design of the scanning device we prepared for the particular acoustic sensor was the design for the linear connection that fulfills a supporting function for the entire device as well as provides movement along the X and Y axes. The final step was the design for the automation of the entire system, for which we selected an Arduino Uno (ATmega328) controller. The actual device was tested in the final stages. The device is designed and tested as functional and able to work on what was intended for and is ready for further modifications or use in practice.
- Open Access
Application of Innovative Methods of Predictive Control in Projects Involving Intelligent Steel Processing Production Systems
This paper describes the enhancement of the existing predictive system of quality management in the processes of metallurgic manufacturing. Specifically, it addresses steel-strip manufacturing. The main quality management innovation is the transition from the current methodological process of a single-step defect evaluation to a two-step evaluation. A two-step defect check of the strip’s surface involves checking for defects during the hot-rolling process first, and double-checking it during the process of pickling. These defects are detected in a well-established process of camera imaging in the production process. The recorded image is then processed mathematically to find the degree of defect correlation in those processes. The two-step evaluation enables a more detailed focus on a particular defect and its position on the strip. Decisions concerning further processing are based on defect evaluation, for instance, whether a rework is necessary to maximize the product utilization and minimize the eventual negative impact of the defect on production equipment. A crucial aspect is also the reduced probability of failures in the manufacturing process.
Open AccessApplication of Innovative Methods of Predictive Control in Projects Involving Intelligent Steel Processing Production Systems
This paper describes the enhancement of the existing predictive system of quality management in the processes of metallurgic manufacturing. Specifically, it addresses steel-strip manufacturing. The main quality management innovation is the transition from the current methodological process of a single-step defect evaluation to a two-step evaluation. A two-step defect check of the strip’s surface involves checking for defects during the hot-rolling process first, and double-checking it during the process of pickling. These defects are detected in a well-established process of camera imaging in the production process. The recorded image is then processed mathematically to find the degree of defect correlation in those processes. The two-step evaluation enables a more detailed focus on a particular defect and its position on the strip. Decisions concerning further processing are based on defect evaluation, for instance, whether a rework is necessary to maximize the product utilization and minimize the eventual negative impact of the defect on production equipment. A crucial aspect is also the reduced probability of failures in the manufacturing process.