ArticlePublisher preview available

Machine Learning-Based Classification of Productive Systems: A Framework for Operational Optimisation

Authors:
Wendell de Queiróz Lamas
Wendell de Queiróz Lamas
Wendell de Queiróz Lamas
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Leonardo Calache
Leonardo Calache
Leonardo Calache
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

The classification of productive systems is essential for optimising industrial operations, enabling organisations to align production processes with strategic goals. This study introduces a computational framework for classifying productive systems—continuous process, mass production, batch production, and project-based production—using machine learning techniques. Input parameters, including production volume, product variety, flexibility, and workforce qualification, were utilised to train decision tree and random forest classifiers. The models were evaluated on synthetic and real-world datasets to ensure accuracy and generalisability. Decision trees provided interpretable classification rules, while random forest models enhanced robustness by aggregating predictions across multiple decision trees. The framework incorporated visualisation tools to highlight decision boundaries and feature importance, offering valuable insights into the underlying classification logic. Results showed that continuous processes are characterised by high production volumes and low flexibility, whereas batch and project-based production systems exhibit greater adaptability and product variety. Despite the models’ effectiveness, feature importance analysis revealed limited differentiation among input parameters, suggesting opportunities for dataset enrichment and feature engineering. The study also identifies potential improvements, such as integrating advanced machine learning algorithms and real-time operational data for dynamic classification. This research provides a scalable and interpretable tool for classifying productive systems, bridging theoretical foundations with practical applications. The proposed methodology aids decision-makers in designing, managing, and optimising production processes, contributing to enhanced operational efficiency and adaptability in diverse industrial contexts. By offering a robust classification framework, this study establishes a foundation for future advancements in production system analysis and optimisation.
Figures - available from: SN Operations Research Forum
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Springer Nature.
Learn more
Content available from SN Operations Research Forum
This content is subject to copyright. Terms and conditions apply.
Operations Research Forum (2025) 6:31
https://doi.org/10.1007/s43069-025-00426-z
RESEARCH
Machine Learning-Based Classification of Productive
Systems: A Framework for Operational Optimisation
Wendell de Queiróz Lamas1,2 ·Leonardo Calache1
Received: 2 December 2024 / Accepted: 10 February 2025 / Published online: 24 February 2025
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025
Abstract
The classification of productive systems is essential for optimising industrial
operations, enabling organisations to align production processes with strategic
goals. This study introduces a computational framework for classifying productive
systems—continuous process, mass production, batch production, and project-based
production—using machine learning techniques. Input parameters, including produc-
tion volume, product variety, flexibility, and workforce qualification, were utilised
to train decision tree and random forest classifiers. The models were evaluated on
synthetic and real-world datasets to ensure accuracy and generalisability. Decision
trees provided interpretable classification rules, while random forest models enhanced
robustness by aggregating predictions across multiple decision trees. The frame-
work incorporated visualisation tools to highlight decision boundaries and feature
importance, offering valuable insights into the underlying classification logic. Results
showed that continuous processes are characterised by high production volumes and
low flexibility, whereas batch and project-based production systems exhibit greater
adaptability and product variety. Despite the models’ effectiveness, feature impor-
tance analysis revealed limited differentiation among input parameters, suggesting
opportunities for dataset enrichment and feature engineering. The study also identi-
fies potential improvements, such as integrating advanced machine learning algorithms
and real-time operational data for dynamic classification. This research provides a
scalable and interpretable tool for classifying productive systems, bridging theoretical
foundations with practical applications. The proposed methodology aids decision-
makers in designing, managing, and optimising production processes, contributing
to enhanced operational efficiency and adaptability in diverse industrial contexts. By
offering a robust classification framework, this study establishes a foundation for future
advancements in production system analysis and optimisation.
Keywords Decision trees ·Feature importance ·Production management ·
Random forest ·System optimisation
Extended author information available on the last page of the article
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
ResearchGate has not been able to resolve any citations for this publication.
Advanced machine learning artificial neural network classifier for lithology identification using Bayesian optimization
Article
Full-text available
  • Nov 2024
Identifying lithology is crucial for geological exploration, and the adoption of artificial intelligence is progressively becoming a refined approach to automate this process. A key feature of this strategy is leveraging population search algorithms to fine-tune hyperparameters, thus boosting prediction accuracy. Notably, Bayesian optimization has been applied for the first time to select the most effective learning parameters for artificial neural network classifiers used for lithology identification. This technique utilizes the capability of Bayesian optimization to utilize past classification outcomes to enhance the lithology models performance based on physical parameters calculated from well log data. In a comparison of artificial neural network architectures, the Bayesian-optimized artificial neural network (BOANN) demonstrably achieved the superior classification accuracy in validation and significantly outperformed a non-optimized wide, bilayer, and tri-layer network configurations, indicating that incorporating Bayesian optimization can significantly advance lithofacies recognition, thus offering a more accurate and intelligent solution for identifying lithology.
View
Effective Machine Learning Solution for State Classification and Productivity Identification: Case of Pneumatic Pressing Machine
Article
Full-text available
  • Oct 2024
The fourth industrial revolution (Industry 4.0) brought significant changes in manufacturing, driven by technologies like artificial intelligence (AI), Internet of Things (IoT), 5G, robotics, and big data analytics. For industries to remain competitive, the primary goals must be the improvement of the efficiency and safety of machinery, the reduction of production costs, and the enhancement of product quality. Predictive maintenance (PdM) utilizes historical data and AI models to diagnose equipment’s health and predict the remaining useful life (RUL), providing critical insights for machinery effectiveness and product manufacturing. This prediction is a critical strategy to maximize the useful life of equipment, especially in large-scale and important infostructures. This study focuses on developing an unsupervised machine state-classification solution utilizing real-world industrial measurements collected from a pneumatic pressing machine. Unsupervised machine learning (ML) models were tested to diagnose and output the working state of the pressing machine at each given point (offline, idle, pressing, defective). Our research contributes to extracting valuable insights regarding real-world industrial settings for PdM and production efficiency using unsupervised ML, promoting operation safety, cost reduction, and productivity enhancement in modern industries.
View
Active monitoring of production status in discrete manufacturing workshops driven by digital twins
Article
Full-text available
  • Oct 2024
  • INT J ADV MANUF TECH
Traditional manufacturing workshop processes are challenging to change, and passive monitoring methods, such as quality traceability and completion rate monitoring, pose challenges for discrete workshops in achieving real-time and accurate production control. Therefore, combined with the digital twin system, a multi-level production status monitoring model is established to monitor discrete manufacturing workshops actively. This article focuses on discrete manufacturing workshops’ modeling and self-evolution production processes. It also proposes an active monitoring architecture for production status in discrete workshops using digital twin technology. A full-information model is constructed for a discrete manufacturing workshop. Then, a multi-node interaction logic is established for the discrete manufacturing workshop. Finally, based on the fuzzy C-means clustering algorithm, production anomalies are monitored, and real-time monitoring of disturbances caused by various production factors, including material status, equipment status, and work-in-progress status, is performed to assess their impact on production. Based on deep Bayesian network learning algorithms, we can monitor production performance and analyze workshop capacity, equipment utilization, bottleneck rhythm, etc. This allows us to actively monitor production performance through the self-evolution of digital twin models. This article aims to proactively predict changes in workshop production status and provide support for further plan completion rate improvement, capacity prediction, and capacity optimization.
View
Probabilistic Ensemble Framework for Injury Narrative Classification
Article
Full-text available
  • Sep 2024
In this research, we analyzed narratives from the National Electronic Injury Surveillance System (NEISS) dataset to predict the top two injury codes using a comparative study of ensemble machine learning (ML) models. Four ensemble models were evaluated: Random Forest (RF) combined with Logistic Regression (LR), K-Nearest Neighbor (KNN) paired with RF, LR combined with KNN, and a model integrating LR, RF, and KNN, all utilizing a probabilistic likelihood-based approach to improve decision-making across different classifiers. The combined KNN + LR ensemble achieved an accuracy of 90.47% for the top one prediction, while the KNN + RF + LR model excelled in predicting the top two injury codes with a very high accuracy of 99.50%. These results demonstrate the significant potential of ensemble models to enhance unstructured narrative classification accuracy, particularly in addressing underrepresented cases, and the potential of the proposed probabilistic ensemble framework ML models in improving decision-making in public health and safety, providing a foundation for future research in automated clinical narrative classification and predictive modeling, especially in scenarios with imbalanced data.
View
Fuzzy logic approach for controlling uncertain and nonlinear systems: a comprehensive review of applications and advances
Article
Full-text available
  • Aug 2024
This paper presents a comprehensive review of the latest developments in fuzzy logic (FL) applications across critical domains which include energy harvesting (EH), ambient conditioning systems (ACS), and robotics and autonomous systems (RAS), highlighting FL's capability to address nonlinearities and uncertainties in diverse technological environments. Through a detailed comparative analysis of research trends over the past decade, it underscores the increasing significance of FL in EH and RAS, contrasting with the sustained interest in ACS. Furthermore, the evaluation of different fuzzy inference systems across domains provides valuable insights into their specific strengths and limitations, aiding researchers and practitioners in making informed decisions aligned with their application needs. Additionally, the paper explores advanced modifications and hybridizations of FL, such as swarm intelligence and integrations with other control strategies, emphasizing the necessity for robust and adaptive FL systems. The review also identifies key open problems and potential research directions, such as the demand for adaptive FL systems in EH and advanced optimization techniques in ACS and RAS. Overall, this state-of-the-art review not only summarizes the current state of FL applications but also outlines a roadmap for future research, offering valuable insights for advancing FL's role in handling uncertainties and nonlinearities in complex systems.
View
Enhancing Machine Learning Models and Classification Accuracy with Advanced Attention Mechanisms
Preprint
Full-text available
  • Jul 2024
This paper provides a detailed discussion of multiple machine learning algorithms and pays close attention to their use, advantages, and disadvantages. Specifically, the Random Forest classifier is highlighted for being more effective with a classification accuracy of 93% being achieved in a binary classification problem. The current method proves superior to known methods and preserves the spatial relationships, thus solving the vanishing gradient problem with the help of two kinds of attention mechanisms. This paper also examines various techniques, such as convolutional neural networks, k-means clustering, and collaborative filtering, explaining how these methods can be applied and optimized. Thus, the rationale of the paper lies in comparison of the above-mentioned methods, emphasizing the significance of modern approaches to ensemble learning for the improvement of model accuracy and stability. Moreover, the paper highlights areas for future research to explore, such as hyper parameters tuning, integration with deep learning frameworks, and use cases in practice. As a result, the presented results can be useful for more advanced studies in the field of machine learning as well as for practical applications for various domains when it is necessary to develop more effective approaches to the use of big data.
View
Enhanced Feature Selection Using Genetic Algorithm for Machine-Learning-Based Phishing URL Detection
Article
Full-text available
  • Jul 2024
In recent years, the importance of computer security has increased due to the rapid advancement of digital technology, widespread Internet use, and increased sophistication of cyberattacks. Machine learning has gained great interest in securing data systems because it offers the capability of automatically detecting and responding to security threats in real time, which is crucial for maintaining the security of computer systems and protecting data from malicious attacks. This study concentrates on phishing attack detection systems, a prevalent cyber-threat. These systems assess the features of the incoming requests to identify whether they are malicious or not. Although the number of features is increasing in these systems, feature selection has become an essential pre-processing phase that identifies the most important features of a set of available features to prevent overfitting problems, improve model performance, reduce computational cost, and decrease training and execution time. Leveraging genetic algorithms, known for simulating natural selection to identify optimal solutions, we propose a novel feature selection method, based on genetic algorithms and locally optimized, that is applied to a URL-based phishing detection system with machine learning models. Our research demonstrates that the proposed technique offers a promising strategy for improving the performance of machine learning models.
View
Enhancing deep neural network training efficiency and performance through linear prediction
Article
Full-text available
  • Jul 2024
Deep neural networks have achieved remarkable success in various fields. However, training an effective deep neural network still poses challenges. This paper aims to propose a method to optimize the training effectiveness of deep neural networks, with the goal of improving their performance. Firstly, based on the observation that parameters (weights and bias) of deep neural network change in certain rules during training process, the potential of parameters prediction for improving training efficiency is discovered. Secondly, the potential of parameters prediction to improve the performance of deep neural network by noise injection introduced by prediction errors is revealed. And then, considering the limitations comprehensively, a deep neural network Parameters Linear Prediction method is exploit. Finally, performance and hyperparameter sensitivity validations are carried out on some representative backbones. Experimental results show that by employing proposed Parameters Linear Prediction method, as opposed to SGD, has led to an approximate 1% increase in accuracy for optimal model, along with a reduction of about 0.01 in top-1/top-5 error. Moreover, it also exhibits stable performance under various hyperparameter settings, shown the effectiveness of the proposed method and validated its capacity in enhancing network’s training efficiency and performance.
View
View
View