A Multi-objective Optimization Approach for Feature Selection in Gentelligent Systems

1 A Multi-objective Optimization Approach for Feature Selection in Gentelligent Systems Mohammadhossein Ghahramani, Senior Member, IEEE, Yan Qiao, Senior Member, IEEE, NaiQi Wu, Fellow, IEEE, and Mengchu Zhou, Fellow, IEEE Abstract—The integration of advanced technologies, such as Artificial Intelligence (AI), into manufacturing processes is at- tracting significant attention, paving the way for the development of intelligent systems that enhance efficiency and automation. This paper uses the term ”Gentelligent system” to refer to systems that incorporate inherent component information (akin to genes in bioinformatics—where manufacturing operations are likened to chromosomes in this study) and automated mechanisms. By implementing reliable fault detection methods, manufacturers can achieve several benefits, including improved product quality, increased yield, and reduced production costs. To support these objectives, we propose a hybrid framework with a dominance- based multi-objective evolutionary algorithm. This mechanism enables simultaneous optimization of feature selection and clas- sification performance by exploring Pareto-optimal solutions in a single run. This solution helps monitor various manufacturing operations, addressing a range of conflicting objectives that need to be minimized together. Manufacturers can leverage such predictive methods and better adapt to emerging trends. To strengthen the validation of our model, we incorporate two real- world datasets from different industrial domains. The results on both datasets demonstrate the generalizability and effectiveness of our approach. Index Terms—Multi-objective Optimization, Artificial Intelli- gence, Feature Selection, Smart Manufacturing. I. INTRODUCTION M ORE recently, manufacturing has embraced the Indus- trial Internet of Things (IIoT), where digital sensors, network technologies, and gentelligent components are inte- grated into manufacturing processes. A gentelligent compo- nent, as defined in the Collaborative Research Centre 653 project [1], refers to components that intrinsically store infor- mation. The focus of that work is on encoding and preserving data within physical parts throughout the product lifecycle. Inspired by this concept, we extend the notion into what we define as a ”gentelligent system.” In our model, we draw an analogy to biological systems, where manufacturing operations are viewed as chromosomes, and their associated features act like genes–each contributing specific behavioral traits to the M. Ghahramani is with Birmingham City University, UK, (e-mail: moham- madhossein.ghahramani@bcu.ac.uk) Y. Qiao is with the Institute of Systems Engineering, Macau University of Science and Technology, Macau, (e-mail: yqiao@must.edu.mo) N. Q. Wu is with Institute of Systems Engineering, Macau University of Science and Technology, Macau. (e-mail: nqwu@must.edu.mo). M. C. Zhou is with the Helen and John C. Hartmann Department of Electrical and Computer Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA (e-mail: zhou@njit.edu). overall system. Just as gene expression determines biological functions, the selected features influence the decision-making capabilities and reasoning and efficiency of our model. By integrating this perspective into a smart manufacturing context, we develop a system that adapts, learns, and evolves by selecting the most informative ’genes’ (features) to optimize industrial performance. This interpretation not only reinforces the original idea of information-rich components, but also adds a novel layer of systemic intelligence and decision autonomy. We are currently witnessing a significant transformation in manufacturing, known as Industry 4.0. This revolution in- volves maximizing production yields through IIoT innovations and integrating intelligent and autonomous systems powered by emerging technologies such as Artificial Intelligence (AI) and Machine Learning (ML). Additionally, Cloud and edge computing are enabling data storage and processing from virtually anywhere. AI has been successfully applied in vari- ous domains, including engineering [2]–[4], urban planning [5], [6], and manufacturing [7]–[14], where it helps solve problems such as functional design, process planning, and product optimization. The application of AI in manufacturing can be categorized into decision support, data management, operations management, and life cycle management. The broad applicability of AI methods allows manufacturers to achieve numerous benefits, such as accurately estimating production costs, mitigating failures, and assessing the financial viability of proposed products. Integrating data from various sensors with state-of-the-art AI technologies can lead to an efficient industrial ecosystem. However, this integration poses several challenges, necessitat- ing practical solutions to address these issues. One of such critical challenges is the Value of Information (VoI) that we can extrac

This content is AI-processed based on open access ArXiv data.