With the increased integration of advances in robotics, automaton, sensors, the Internet-of-Things (IoT) and high-speed computing performance manufacturing industry is entering a period of substantial innovation and change. These technologies make products and processes become smarter and connected, but at the same time also lead to pervasive digitalization of factories and challenges manufacturing enterprises to re-consider and re-evaluate their current operations and future strategic directions. Over the past few decades, the topic of Artificial Intelligence (AI) in manufacturing has attracted widespread attention in the scientific community because of its huge potential in improving the efficiency of various processes. AI plays a vital role from big data to full automation. It can be explored to solve problems with uncertain, time-variant, and complex features, which in turn can help manufacturers to improve design, manufacturing, and handling of processes and products, and thus significantly improve productivity, quality, flexibility, safety, and cost. However, though a wide array of AI tools has already been implemented to address a diverse set of manufacturing problems, many gaps still exist that should be addressed to ensure that AI can be seamlessly integrated into factory operations.
This special issue aims to bring together researchers and scientists from artificial intelligence and advanced manufacturing from various application areas to discuss problems and solutions in the area, to identify new issues, and to shape future directions for research. Seven papers are included in this special issue.
The first paper Smart Seru Production System for Industry 4.0: A Conceptual Model Based on Deep Learning for Real-Time Monitoring and Controlling by Torkul et al. proposes a six-component smart seru production system model by adding advanced technologies and deep learning techniques in order to overcome the deficiencies of the seru production system. When the assembly process is monitored, and the worker performs an operation in the wrong order, it warns the worker and prevents the assembly of the product in the next stages and prevents the reassembly process, thus eliminating unnecessary time from reprocessing. In addition, through real-time data capture and collection, intelligent data processing based on deep learning models and intelligent decision- making processes, seru production system is shown to have the advanced intelligence capabilities.
The second paper Sequential Delay-Free Strategy for Sliding Mode Position Observer Considering Analog and Digital Implementations of PMSM Drives by Li et al. proposes a sequential delay-free scheme for sliding mode observer (SMO) to improve the estimation accuracy of the high-speed permanent magnet synchronous motor (PMSM) drives aiming at the shortcomings of the existing delay handling schemes in SMO. Both the digital delay and analog delay effect on estimation accuracy are analysed theoretically, which are not comprehensively focused on in the previous studies. In detail, current-based, position-based and speed-prediction- based compensation strategies are adopted to deal with the analog delay, LPF delay and execution delay, respectively. Compared with traditional methods, the proposed algorithm is easier to understand and implement in engineering.
The third paper Non-contact surface roughness evaluation of Milling surface using CNN-Deep Learning models by Bhandari & Park propose a contactless surface roughness evaluation system implemented using deep learning models that have the potential for industrial applications. The roughness classes were based on the experiment surface roughness value (Ra) employing Taguchi design of experiments (DOE). The proposed system has the potential to be used without human intervention, eliminate surface damage, and be fast and reliable, which is well suited for smart factories. The proposed Deep Learning (DL) model features a highly accurate and slim structure, potentially substituting human quality control procedures that employ expensive surface roughness measuring devices. Furthermore, DL models’ prediction and classification performance are evaluated through the Precision and Recall curve, ROC curve, and confusion matrix, in addition to the validation and test accuracy. The developed model can be integrated into the machining equipment and perform a real-time inspection of the machining results.
The fourth paper Utilization of acoustic signals with generative Gaussian and autoencoder modeling for condition-based maintenance of injection moulds by Rønsch et al. explores two different generative models for binary classification of machinery into faultless or faulty. These two different generative methods model faultless machinery and, in turn, are based on Gaussian and autoencoder modelling. The authors proposed to perform model adaptation for improved performance/generalization across different production units and under data scarcity conditions. The proposed framework requires minimal computational resources and is highly feasible for implementation in real-world industrial settings with numerous analogous machines.
The fifth paper A Connective Framework to Minimize the Anxiety of Collaborative Cyber-Physical System by Islam et al. proposes a connective framework that intelligently safeguards a Cyber Physical Systems (CPS’s) physical and psychological safety. The approach combines knowledge, human intelligence, and AI to improve the decision-making of the CPS encountering complex and dynamic situations. AI-based methods supported by visual and IR sensor cues are employed to produce a flexible system in which the cooperating physical elements can detect and react well to better conceived situations. Regardless of the specific case, the proposed method’s effectiveness and applicability have been validated in a real-world industrial scenario. The technique has an edge in developing safe mechanisms for intelligent factories to safeguard costly physical assets and involve human workers to conserve and optimize the desired goal’s efficiency and productivity. The developed methodology can be applied to various industrial scenarios.
The sixth paper Adaptive neuro-fuzzy inference system with analytic hierarchy process: an application for drawworks’ failure mode and effect analysis by Olugu et al. propose a fuzzy multi-criteria decision-making (MCDM) approach integrated with an artificial intelligence algorithm for sustainable maintenance decision-making in the oil and gas industry. The proposed model incorporates a combination of quantitative intelligent-based technique and qualitative objective accuracy-oriented analyses to evaluate participants experience on maintenance of offshore drilling and production platforms operated in Turkmenistan. The applicability of the proposed model was validated by a case study of a drawworks system on an offshore production and drilling platform. The findings demonstrated that the model’s accuracy is satisfactory, making it suitable for assessing sustainable maintenance practices.
The seventh paper Machine learning algorithms applied to intelligent tyre manufacturing by Acosta et al. examines the predictive performance of six machine learning algorithms for tyre weight modeling in intelligent tyre manufacturing from real data. The main contribution of this research is developing a scheme solution that uses machine learning algorithms to industrial processes in stored data large manufacturing processes, allowing the process engineer to manage the finished products and the process parameters. The proposed relevance vector machine is compared with other algorithms such as support vector machine, artificial neural network, k-nearest neighbors, random forest, and model trees. A Relevant Vector Machine (RVM) algorithm presented the smallest measures of squared error and better performance than the other algorithms.
Finally, we would like to all the reviewers who gave their significant comments and suggestions for improving the published papers in this special issue. Thank you to all the authors in contributing to the publication of this special issue. Special thanks should be given to Professor Stephen T Newman, Editor-in-Chief of the International Journal of Computer Integrated Manufacturing who gave his great support. We hope that this special issue will bridge the academic and practitioners to enhance the advanced manufacturing in the future.