Conveyor equipment is commonly used in packaging systems, and operators must stand for long periods and perform a similar, monotonous task. This might increase the operator’s physical and mental fatigue. A hypothesis is, that the speed, noise, and direction of material flow influence the operator’s workload. A work system optimization could be proposed using response surface methodology. The Central Composite Design was proposed as an experimental design. Both physical and mental workload are assessed as the response. The results showed that the optimal combination of physical workload response parameters was a speed of 25.8579 cm/s, a noise of 89.3569 dB, and a flow direction from right to left. Meanwhile, the optimal parameter combination for mental workload response was a speed of 25.8579 cm/s and a noise level of 77.9289 dB, with the most optimal flow direction from right to left.
The paper compares the results of protective measures used to maintain business continuity of Polish production enterprises during a crisis situation. It examines how these results differ depending on the type and form of production organization. A mix-model CATI/CAWI survey was performed on a representative group of 600 enterprises. Over 40 individual protection measures were identified and assigned to one of five predefined groups. The research specified three production types (unit/job, batch, mass) and two production forms (line production, job-shop production). A chi-square test was used for the statistical analysis. The results indicate that production volume, defined as the number of manufactured, repetitive products, affects the selection of protection measures. The production form, reflecting the way a product flows through the production process, is also significant. The findings provide a basis for future preparations aimed at ensuring business continuity in the event of similar disruptions to production processes.
Quality assessment of manufactured products is vital to ensure performance, safety, and customer satisfaction across industries. Defects in items such as bottle caps, cables, capsules, leather, and metal components can affect functionality and durability. Traditional inspection methods relying on manual visual checks are time-consuming and error-prone. This study proposes an AI-driven framework using the Probabilistic U-Net integrated with a Conditional Variational Autoenco der (CVAE) for automated defect detection. The model introduces stochastic latent variables to generate multiple plausible segmentation maps, enhancing accuracy under ambiguous or noisy conditions. Using the MVTec Anomaly Detection dataset, which includes defects such as scratches and discoloration, the system applies preprocessing steps including resizing, normalization, and data augmentation to enhance the robustness and consistency of the input data. A hybrid loss combining cross-entropy and Kullback-Leibler divergence improves segmentation precision and latent space alignment. Experimental results confirm robust and reliable defect detection across diverse product categories, demonstrating the model's potential for automated manufacturing quality assurance.
The article addresses the challenges associated with the Product-as-a-Service (PaaS) concept in the context of the circular economy. The use of simulation technologies within this concept can accelerate the adoption of circular economy principles in companies. It enables faster modelling and optimisation of value-retention processes such as remanufacturing and refurbishment. The aim of the paper is to propose a simulation approach that allows engineers to develop simulation models (digital twins) in the same way they design and redesign closed-loop manufacturing systems. A further objective is to highlight the specific requirements for testing such simulation models and for sizing planned digital-twin projects. The article presents the development of a hybrid model that combines discrete piece-flow for remanufacturing with mass flow (in kilograms) for recycling, using household appliances as an example and identifying sources of uncertainty that affect the organisation of PaaS processes, particularly remanufacturing and recycling. The main contribution is the development of a hybrid simulation approach based on operations in the PaaS domain, i.e. closed-loop manufacturing with the 6R framework, and the application of the COSMIC methodology to size simulation projects. The article also discusses methods for simulation modelling of individual sources of uncertainty and uses an operation-based simulation method instead of conventional discrete-event simulation.
Three-dimensional (3D) printing is an attractive method for creating gel geometries for specific applications. In this research, the 3D printing parameters for direct and indirect extrusion of alginate-based hydrogels are optimized using the Taguchi and Analysis of Variance methods for output responses, including printed shape retention and the height of the 3-printed layers. The results show that for an indirect extrusion, the optimal values of extrusion speed, printing speed, and nozzle distance for the shape retention ability are 13 steps/mm, 50 mm/s, and 0.5 mm, respectively, while those values for the printed height are 13 steps/mm, 60 mm/s, and 0.5 mm, respectively. For direct extrusion, the optimal parameter set for former response is 35 steps/mm, 60 mm/s, and 0.6 mm, respectively, while the set for the latter is 35 steps/mm, 40 mm/s, and 0.6 mm, respectively. The findings in this report can be used as data for related research.
This study investigates the impact of Industry 4.0 on sustainable supply chain practices in Ghana's beverage industry, emphasizing the role of managerial support. Data were collected from 200 beverage firms in Greater Kumasi using a structured questionnaire and analyzed with SPSS 23 and SmartPLS SEM. Results show that Industry 4.0 positively influences both sustainable supply chain practices and managerial support. Managerial support also enhances sustainability and partially mediates the relationship between Industry 4.0 and supply chain sustainability. The findings suggest that the adoption of Industry 4.0 technologies, coupled with active managerial involvement, strengthens sustainable practices in the beverage sector. The study contributes to understanding how digital transformation and leadership jointly promote sustainability in developing economies. Future research should examine other manufacturing sectors and the role of Industry 4.0 in driving creativity and innovation.
The aim of this study was to analyse the impact of technological modifications in the production cycle of flat components on reducing production costs. The research involved a detailed analysis of production processes, identification of key areas requiring improvement, and the development of new technological solutions. As part of the study, changes were implemented in the production process, and their effect on production costs was thoroughly evaluated. The technological modification in the manufacturing of the "metal support bracket" was designed to enhance efficiency by reducing unit production time and assessing the effect of this change on overall production performance. The main innovations included the introduction of a machining centre that integrated drilling, chamfering, and threading operations into a single process. This significantly reduced both unit production time and cost while eliminating machine downtime. Additionally, powder coating was replaced with electroplating, which resolved issues related to hole narrowing and ensured the maintenance of precise technical dimensions. The implemented changes resulted in a shorter production cycle, improved product accuracy and quality, and reduced machine downtime. The analysis demonstrated that these modifications positively influenced the enterprise's competitiveness, generating substantial cost savings. This work provides a practical example of the application of industrial innovation, contributing to cost reduction, shorter production cycle times, and enhanced precision and quality of products.
The design of machine control systems requires the correct selection of safety logic devices to ensure functional safety and compliance with international standards such as EN ISO 13849-1 and IEC 62061. This process is typically based on expert knowledge and manual evaluation of design parameters, which can be time-consuming and error-prone. In this study, machine learning techniques are applied to automate and improve the selection of safety logic devices using real industrial data originating from various types of machinery and automated manufacturing real-world projects. This work introduces a significantly extended industrial dataset comprising 670 labelled machine configurations derived from real anonymized engineering projects and performs a comprehensive comparison of ten representative ML algorithms implemented in WEKA. The main novelty of the study is a unified large-scale comparative evaluation of heterogeneous machine learning classifiers on real industrial decision data, enabling joint assessment of scalability, generalization, interpretability, and computational efficiency under identical experimental conditions. The results demonstrate that increasing dataset size considerably enhances model stability and generalization. The Averaged 2Dependence Estimator (A2DE) achieved the highest performance with an accuracy of 86% and Kappa = 0.81, followed by REPTree and Random Forest classifiers. Rule-based methods such as PART and NNge maintained strong interpretability with competitive predictive power. The findings confirm that probabilistic and ensemble algorithms provide reliable and practically applicable solutions for data-driven decision support in industrial safety engineering, paving the way for deployable, explainable, and adaptive decision-support tools in smart manufacturing environments.
The rapid development of Industry 4.0 has introduced advanced technologies such as IoT, cyber-physical systems (CPS), and industrial IoT into manufacturing environments. However, traditional production management systems remain largely reactive, operating in discrete modes with fragmented interfaces. This paper presents a concept for a production management support system that integrates large language models (LLMs) that enable natural language interaction. This solution concept addresses the key challenge of data fragmentation by creating an intelligent digital twin that acts as a production expert capable of contextual reasoning, information synthesis from multiple sources, and real-time decision support. This concept demonstrates the potential to transform production management from a reactive to a proactive operating model by leveraging LLM's capabilities in pattern recognition, predictive analysis, and automated recommendation generation. Future development directions focus on optimizing business intelligence integration, improving automated recommendation mechanisms, and standardizing natural language user interfaces for industrial applications.
The study presents an approach to predicting the completion time of production operations using supervised machine learning techniques. The analysis was conducted on a database extracted from ERP and MES systems, comprising over 150,000 records containing technological and production completion times, operation numbers, and textual descriptions of operations. The dataset preprocessing involved cleaning, feature encoding, and text vectorisation using the TF-IDF method to represent semantic patterns within operation descriptions. Regression models, including Linear Regression, Random Forest, and XGBoost, were trained and evaluated using Go ogle Colab. Model performance was assessed using standard evaluation metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the coefficient of determination (R2). Experimental results showed that ensemble-based methods achieved the highest predictive accuracy, outperforming the baseline model based solely on technological completion time. In addition, the study examines the sensitivity of selected models to hyperparameter settings and analyses the impact of alternative categorical feature encoding methods on prediction accuracy. The proposed approach enables more accurate estimation of production durations and supports data-driven decision-making in manufacturing environments.
In the automotive industry, sustainable manufacturing involves integrating the triple bottom line of economic, environmental, and social aspects into manufacturing operations. However, the automotive industry faces challenges in prioritizing sustainability due to its interdependence and complexity, where effective decision-making requires identifying influential factors and understanding their relationship. To address the challenge, a hybrid method combining Interpretive Structural Modeling (ISM) and MICMAC analysis is utilized. ISM establishes connections between specific criteria, enabling a comprehensive understanding of their interdependencies. MICMAC analysis then helps the prioritization process by classifying factors according to their driving and dependency power. This approach helps stakeholders identify the most crucial factors and develop action plans to reduce or eliminate obstacles hindering the adoption of sustainable manufacturing practices. This study addresses the sustainability issues in the automotive sector in Kerala, India. Furthermore, the study suggests the potential expansion by conducting a large-scale survey to include additional criteria, thereby enhancing the understanding of sustainable practices in the automotive sector. The results indicate the proposed ISM-MICMAC model outperforms existing methods in several areas, including accuracy of prioritization (92.5% vs. 70% for AHP), resource efficiency (85% vs. 60% for Carbon Footprint Analysis), emission reduction (30% vs. 20% for LCA), and stakeholder engagement (85% vs. 80% for LCA). sustainability, Interpretive structural modeling, Structural Self Interaction Matrix, Cronbach Alpha, Statistical Package for Social Science.
The aim of this publication is to synthesize dynamic models for selected innovation diffusion models. For this purpose, modeling in the system dynamics (SD) convention was applied to represent the flow between the stock of potential and current users of innovations in three selected diffusion models: the Bass model, the source model, and the contact model. The AnyLogic software was used as the simulation environment. As a result of the study, simulation models were developed that enable forecasting the behavior of participants in a given population depending on predefined coefficients. This solution is particularly useful for the cost optimization of promotional activities in enterprise departments responsible for marketing innovative products, as well as for diffusion understood as the dissemination of modern organizational and process methods among employees of an organization.
Excavated material transportation is crucial in mining operations, requiring optimal efficiency. Since the early 2000s, various aspects of transportation network optimization have been researched, often producing methods with overlapping objectives and outcomes. This work consolidates and analyzes existing methods and artifacts related to decision support for optimizing ore transportation networks. A systematic literature review, following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, was conducted using sources such as Scopus, Web of Science, and Google Scholar. Out of 170 initial research papers, 46 were selected for detailed analysis. The review highlights the current state of decision support in ore transportation, focusing on supported decisions, optimized processes, and applied methods. It also identifies research gaps and future trends in this field.
Today, the development of a scientific and methodological approach to modelling the impact of digital transformation on enterprise management is highly relevant. This approach should be based on the rules of fuzzy logic and be adaptable to environmental changes. Consequently, the purpose of this study is to develop an optimal tool for modelling the decision-making process in enterprise management under the influence of digital transformation. The study’s outcome is a model for presenting fuzzy knowledge, demonstrated through examples of models designed to assess the impact of digital transformation on enterprise management, based on input from expert assessments. The developed model interprets the scored expert points for a loosely structured or unstructured task, thereby revealing the subjectivity of experts and providing a quantitative assessment for non-formalised tasks.
Green sand mould quality plays a pivotal role in casting reliability and dimensional accuracy, yet mould properties degrade over time due to environmental exposure and production delays. This study examines the time-dependent behaviour of critical mould characteristics - green compressive strength (GCS), mould hardness, and permeability - under varying process conditions. Using a Taguchi L27 orthogonal array, the effects of moisture content, ramming time, and holding time were systematically evaluated across 27 experimental setups, with sequential moulding under realistic foundry conditions. Regression models were developed to predict property degradation based on mould age, which is defined as the cumulative moulding and holding time. Results highlight that optimal property retention occurs at moderate ramming times, higher moisture content, and shorter holding periods. Case 2 (4% moisture, 5-second ramming, 10-minute holding) demonstrated the most favourable balance of strength, hardness, and gas permeability. The predictive models exhibited high accuracy (R-2 > 0.94), supporting their data-driven mould quality control application. These findings offer practical insights to improve maintainability, reduce casting defects, and enhance process reliability in sand casting operations. The research contributes to the broader goals of sustainable manufacturing and production system optimisation through statistically guided process management.
In the industrial sector, Supervisory Control and Data Acquisition (SCADA) systems are essential for managing Industrial Internet of Things (IIoT) networks. However, these systems have become increasingly exposed to cyberattacks targeting the communication layers embedded in industrial processes. Such vulnerabilities can cause severe disruptions in manufacturing and production environments. The ongoing digitalization of Industrial Control Systems (ICS) has further amplified these risks, emphasizing the need for robust security mechanisms such as Intrusion Detection Systems (IDS). This research aims to develop a high-precision AI-based IDS capable of protecting SCADA systems from evolving cyber threats. To achieve this, three categories of machine learning algorithms were evaluated: Deep Learning models (CNN, RNN, LSTM), Boosting algorithms (XGBoost, GBoost, AdaBoost), and classical methods (RF, DT, KNN). Extensive experiments were conducted using two benchmark SCADA datasets, WUSTL-IIoT-2018 and WUSTL-IIoT-2021. The results demonstrated outstanding detection performance, with all models achieving accuracy rates above 99.91%. Specifically, RF, DT, KNN, and XGBoost reached perfect accuracy (100%) on the WUSTL-IIoT-2018 dataset, while XGBoost, LSTM, and CNN achieved 99.99% accuracy on WUSTL-IIoT-2021. Additional evaluation metrics, including precision, recall, and F1-score, confirmed the robustness of the models. The findings highlight the potential of AI-driven IDS solutions to enhance the security and resilience of industrial SCADA infrastructures.
This study evaluates the implementation of the Lean, Agile, Resilient, and Green (LARG) approach in the electric motorcycle industry in Indonesia using the Bayesian Best Worst Method (BWM). The main focus of the study is to identify and determine the weight of the most relevant LARG indicators to improve the competitiveness and sustainability of the industry. From the analysis results, the Resilience indicator has the highest weight, while the Lean indicator has the lowest weight. Important sub-indicators identified include the ability to take corrective action when disruptions occur, waste management according to regulations, flexibility in collaboration with industry partners, and component quality testing. Recommended priority strategies include developing a standards-based safety system, consistent technology transfer, and provision of adequate infrastructure. The results of this study provide data-based strategic guidance to improve efficiency, flexibility, durability, and environmental sustainability in the electric automotive industry in Indonesia.
This study addresses the high defect rate (up to 99%) and the heavy dependence on operator experience in the Direct Memory Access (DMA) injection molding process. To overcome these limitations, the study applies the Taguchi method using L18 orthogonal arrays to systematically optimize six key process parameters: nozzle size, mold temperature, ejection pressure, binder chemical concentration, mold material weight, and molding time. Signal-to-noise (S/N) analysis and ANOVA were used to identify the most influential factors and determine the optimal settings. The results show a significant reduction in defect rates: delamination defects decreased from 20% to 4%, flat wire defects from 4% to 0.5%, and corrugated plastic defects from 0.6% to 0.1%. Notably, the integration of computer vision inspection and process optimization improved product quality and reduced production time. The novelty of this study lies in the systematic application of the Taguchi method to high-precision semiconductor processes and the combination with advanced testing technologies, opening up a new direction in process optimization for the industry.
Operating within a complex and dynamic global ecosystem, organizations are subject to continual evolution in response to shifting market demands. Adaptability has long been a crucial attribute of modern organizations. However, in recent years, resilience has become equally essential, shaping the future trajectory of their development. The aim of this study is to contribute empirically by proposing simulation as a method for enhancing operational resilience. Designing workflows within this context necessitates a multifaceted approach. While a comprehensive understanding of the individual process steps is essential, it is equally crucial to consider the broader system context and the factors that influence the successful execution and desired outcomes. This paper presents simulation research as a tool for performance improvement, and investment decisions. Crucially, simulation modelling facilitates a forwardlooking approach, enabling organizations to not only withstand and recover from challenges but to emerge strengthened and transformed, rather than merely reverting to pre-crisis conditions. The article emphasizes the strategic value of simulation modelling in enhancing an organization's operational resilience.
This study optimizes the Friction Stir Welding (FSW) process for aluminum alloys AA6061 and AA7075, crucial for the automotive and shipbuilding industries. The Taguchi method combined with Grey Relational Analysis (GRA) was employed to determine optimal process parameters: rotational speed, travel speed, and pin depth in the Z-axis. Experiments revealed that a rotational speed of 1000 RPM, travel speed of 20 mm/min, and pin depth of 0.16 mm achieved the highest tensile strength (166.68 MPa) and hardness (97.86 HV). Analysis of Variance (ANOVA) confirmed the significant impact of rotational speed on mechanical properties. The study demonstrates the efficacy of combining Taguchi and GRA methods for FSW optimization, providing a framework for improving material performance in lightweight, highstrength applications. Future research should explore broader material scopes, advanced control systems, and environmental impacts.
