Human-machine Collaboration Research Articles

Implementing and evaluating the quality 4.0 PMQ framework for process monitoring in automotive manufacturing

This study presents an applied integration of machine learning (ML) within the Process Monitoring for Quality (PMQ) framework to address persistent limitations in traditional quality control systems, particularly their inability to manage high-dimensional and real-time manufacturing data. This research enhances the PMQ framework with a novel Validate phase that introduces human oversight and interpretability into the ML decision-making loop. The modified framework has been implemented in a high-precision automotive component facility. The study relied on various ML algorithms, such as Decision Trees (DT), Random Forest (RF), Gradient Boosting Machine (GBM), Logistic Regression (LR), Support Vector Machine (SVM), and Artificial Neural Networks (ANN), to classify and predict defects in engine valves during manufacturing processes. The findings highlighted that GBM and RF provided the best performance, achieving an F1 score of 0.98 and an AUC of 0.99. Feature importance analyzes identified seat height and undercut diameter as key predictors, reinforcing the relevance of interpretable ML in industrial quality management. Beyond technical accuracy, this work demonstrates how structured human-machine collaboration can foster trust in AI-driven quality control, offering a scalable blueprint for Quality 4.0 adoption. The findings contribute to academic literature and industrial practice by bridging conceptual frameworks and real-world implementation strategies for AI-enhanced quality assurance.

Development Status and Trend of Mine Intelligent Mining Technology

Intelligent mining technology, as the core driving force for the digital transformation of the mining industry, integrates cyber-physical systems, artificial intelligence, and industrial internet technologies to establish a “cloud–edge–end” collaborative system. In this paper, the development trajectory of intelligent mining technology has been systematically reviewed, which has gone through four stages: stand-alone automation, integrated automation and informatization, digital and intelligent initial, and comprehensive intelligence. And the current development status of “cloud–edge–end” technologies has been reviewed: (i) The end layer achieves environmental state monitoring and precise control through a multi-source sensing network and intelligent equipment. (ii) The edge layer leverages 5G and edge computing to accomplish real-time data processing, 3D dynamic modeling, and safety early warning. (iii) The cloud layer realizes digital planning and intelligent decision-making, based on the industrial Internet platform. The three-layer collaboration forms a “perception–analysis–decision–execution” closed loop. Currently, there are still many challenges in the development of the technology, including the lack of a standardization system, the bottleneck of multi-source heterogeneous data fusion, the lack of a cross-process coordination of the equipment, and the shortage of interdisciplinary talents. Accordingly, this paper focuses on future development trends from four aspects, providing systematic solutions for a safe, efficient, and sustainable mining operation. Technological evolution will accelerate the formation of an intelligent ecosystem characterized by “standard-driven, data-empowered, equipment-autonomous, and human–machine collaboration”.

Application and Limitations of Large Language Models in the Translation of the Contract Book of the Civil Code

In recent years, the rapid advancement of Large Language Models (LLMs) has created new possibilities for legal translation. This study employs a combination of quantitative analysis and case studies to systematically compare the textual features of the authoritative human-translated version and the GPT-4o-translated version of the Contract Book of the Civil Code across three dimensions: terminological accuracy, syntactic complexity, and the handling of semantically ambiguous expressions. The findings are as follows: (1) In trans-lating the Contract Book of the Civil Code, the human version demon-strates su-perior rigor and precision in legal terminology, in contrast, the GPT-generated version displays deviations in term usage that may introduce ambiguities in le-gal interpretation. (2) Although the GPT version enhances readability by reduc-ing syntactic complexity, this simplification compromises the precision and formality required in legal discourse. (3) In addressing se-mantically ambiguous expressions, human translators employ terminological transformation, lexical supplementation, and tense modulation, while the GPT version tends to rely on literal translation, increasing the risk of omission or distortion in legal provi-sions. This study highlights the current limitations of LLMs in legal translation and underscores the importance of human-machine collaboration, offering in-sights and guidance for producing high-quality legal translations with the assis-tance of LLMs.

A Framework for Integrating Robotic Process Automation with Artificial Intelligence Applied to Industry 5.0

The transition to Industry 5.0 highlights the growing integration of Robotic Process Automation (RPA) and Artificial Intelligence (AI) in industrial ecosystems. However, adoption remains fragmented, lacking standardized frameworks to align intelligent automation with human-centric principles. While RPA improves operational efficiency and AI enhances cognitive decision-making, challenges such as organizational resistance, interoperability, and ethical governance hinder scalable and sustainable implementation. The envisioned scenario involves seamless RPA-AI integration, fostering human–machine collaboration, operational resilience, and sustainability. Expected outcomes include (1) hyperautomation for efficiency gains, (2) agile, data-driven decision-making, (3) sustainable resource optimization, and (4) an upskilled workforce focusing on innovation. This study proposes a structured five-stage framework for RPA-AI deployment in Industry 5.0, combining automation, cognitive enhancement, and human–machine symbiosis. A systematic literature review (PICO method) identifies gaps and supports the framework’s design, validated through operational, human-impact, and sustainability metrics. Incorporating ethical governance and continuous upskilling, the model ensures technological advancement aligns with societal and environmental values. Results demonstrate its potential as a roadmap for responsible digital transformation, balancing efficiency with human-centricity. Future research should focus on empirical validation and sector-specific adaptations.

Is control necessary for drivers? Exploring the influence of human-machine collaboration modes on driving behavior and subjective perception under different hazard visibility scenarios.

Is control necessary for drivers? Exploring the influence of human-machine collaboration modes on driving behavior and subjective perception under different hazard visibility scenarios.

Systematic literature review on the evolution of human-machine collaboration in the workplace

The inclusion of advanced technology within workplaces has significantly transformed human-machine collaboration. From the initial stage of industrial automation to the modern artificial intelligence-based decision-making process, business organizations have depended heavily on machines to improve their accuracy, efficiency, and productivity. Human-machine collaboration has extended beyond basic automation, facilitating synergy where human and intelligent tools work together to perform complex tasks. The rapid development of artificial intelligence, machine learning and robotics has redefined job responsibilities and has reshaped workforce dynamics. It has been observed that though technology promises to provide efficiency gains, it has also pointed out specific concerns regarding ethical considerations, the evolving nature of the work environment and job displacement.

Human-Centric IoT-Driven Digital Twins in Predictive Maintenance for Optimizing Industry 5.0

Predictive maintenance now heavily relies on digital twins and the Internet of Things (IoT), which allow industrial assets to be monitored and decisions made in real time. However, adding human components to conventional optimization processes creates new difficulties as Industry 5.0 moves toward human-centric systems. Existing frameworks frequently disregard human preferences, intuition, and safety considerations, which makes human operators distrustful and unwilling to accept them. To enable predictive maintenance, this paper presents a novel multi-objective optimization framework that incorporates human feedback into IoT-driven digital twins. The framework uses an enhanced particle swarm optimization (PSO) algorithm to reconcile competing goals, including maintaining operator safety, optimizing asset reliability, and minimizing maintenance costs. Furthermore, maintenance tasks are adaptively scheduled using built-in reinforcement learning (RL) and optimized model parameters are fine-tuned for improved predictive accuracy using Bayesian optimization. The latter is based on real-time operational data. In addition to promoting a safer working environment, the suggested approach shows a significant reduction in unplanned downtime and maintenance costs. This research contributes to the development of more resilient, adaptive, and collaborative industrial systems by aligning with the human-centric principles of Industry 5.0. The proposed model was tested using the maintenance duration and achieved an improvement of 10 to 100 hours. The model was further compared with the PSO algorithm, demonstrating its superiority with a 7.5% reduction in total maintenance cost and a 6.3% decrease in total downtime. These improvements contribute to enhanced operational efficiency and better human-machine collaboration by minimizing unnecessary interventions and optimizing resource allocation.

Innovation and entrepreneurship education guidance and optimization analysis based on deep learning optimization algorithm

According to the constructivist learning theory, learners need to actively construct a knowledge system in a practical situation, and deep learning, with intelligent data analysis, pattern recognition and predictive decision-making capabilities, just builds a practical platform for students to discover market opportunities and formulate innovative solutions, which effectively promotes the development of their innovation and entrepreneurship skills. At the same time, deep learning emphasizes the understanding, utilization and generation of knowledge, and further strengthens students’ innovation and entrepreneurship literacy by cultivating human-machine collaboration, which is in line with the concept of “learning by doing” in contextual cognitive theory. In addition, deep learning has the characteristics of thinking training, knowledge transfer and application, which echoes the requirements of higher-order thinking cultivation in Bloom’s educational goal classification theory, which can effectively improve the knowledge integration, transformation and application ability of college students, and enhance the ability of knowledge creation and innovation and entrepreneurship. In the application of practical education scenarios, taking the YOLOv5s network as an example, after the introduction of the SE module and the SIoU loss function, the average accuracy (mAP) of the feature detection and entrepreneurship scene recognition tasks of innovative projects is increased to 89.74% and 89.33%, respectively. This significant performance improvement intuitively demonstrates the ability of deep learning algorithms to accurately analyze education data. Through the efficient processing of educational data such as classroom teaching videos and project display images, the system can accurately identify students’ innovative thinking performance and entrepreneurial practice scenarios, and then provide data support for teachers to optimize teaching design and reform teaching methods, tailor personalized learning paths for students, truly realize data-driven educational innovation, and promote the deep integration of innovation and entrepreneurship education theory and practice.

AI, Optimization, and Human Values: Mapping the Intellectual Landscape of Industry 4.0 to 5.0

This study conducts a systematic bibliometric literature review to explore the conceptual and technological transition from Industry 4.0 to Industry 5.0, focusing on the roles of artificial intelligence (AI), optimization, and human values. Applying the PRISMA 2020 protocol, the analysis includes 53 peer-reviewed sources from the Scopus database, emphasizing the integration of advanced technologies such as cyber–physical systems, the Internet of Things, collaborative robotics, and explainable AI. While Industry 4.0 is marked by intelligent automation and digital connectivity, Industry 5.0 introduces a human-centric paradigm emphasizing sustainability, resilience, and co-creation. The findings underscore the significance of human–machine collaboration, process personalization, AI education, and ethical governance as foundational pillars of this new industrial era. This review highlights the emerging role of enabling technologies that reconcile technical performance with social and environmental values, promoting a more inclusive and sustainable model for industrial development.

Integrating Generative Artificial Intelligence and Human Design: The Impact of Automation Level on Human Creative Experience and Efficiency

Generative artificial intelligence (GAI) has emerged as an indispensable tool across various design disciplines, increasingly integrated into different stages of the design process. However, research on the effectiveness of human-AI collaboration in design remains limited, and the role of AI in the creative process is still underexplored. This study presents a single-factor between-subjects experiment involving 79 participants to evaluate the impact of varying degrees of automation in generative design tools on user experience and design efficiency. The independent variable is the degree of automation of the GAI tool, categorized into three levels: low, moderate, and high. The dependent variables include time, creative experience, and creative quality. Creative experience encompasses task load, perception, and creativity support, while creative quality pertains to aesthetic and usability outcomes. The results indicate that increasing the level of automation significantly reduces design time. Moderate levels of automation are most effective in lowering task load and balancing human-machine collaboration, whereas extreme levels of automation may be counterproductive. Although highly automated GAI can rapidly enhance visual aesthetics, its impact on stimulating designers’ creativity and critical thinking is limited. This research contributes to evaluating the effectiveness of human interaction with AI in the design process and offers insights for optimizing the application of AI in design.

Research on the effective application of Artificial Intelligence in automation control in electronic engineering

In recent years, Artificial Intelligence (AI) has made rapid advancements, with its applications extending across numerous fields, including electronic engineering. This paper explores the various applications of AI in electronic engineering automation control, analyzes its current status, potential benefits, and future trends. AI is transforming various aspects of electronic systems such as intelligent prediction, precision control, smart grids, and factory automation. The integration of AI into these systems promises improved operational efficiency, cost savings, and sustainability. AI is currently widely used in areas like automated design, fault detection, and predictive maintenance, showcasing its value in improving system reliability and reducing downtime. This paper also outlines how AI is expected to evolve in conjunction with advancements in hardware and algorithms, paving the way for future innovations in smart manufacturing, energy management, and autonomous systems. AI is revolutionizing electronic engineering with applications in intelligent prediction, precision control, and automation, enhancing efficiency and reliability. As AI advances, it is expected to further improve human-machine collaboration and drive innovations in smart manufacturing and energy management. This paper examines the current and future impact of AI in electronic engineering, highlighting its potential for cost savings and sustainability.

Bridging Human and Machine Intelligence: The Role of AI in Next-Generation Collaborative Robotics

The integration of artificial intelligence (AI) into collaborative robotic systems marks a pivotal advancement in smart manufacturing, facilitating dynamic interactions between human workers and machines. This study examines the role of AI in enhancing the adaptability, perception, and decision-making capabilities of collaborative robots (cobots) within complex industrial environments. By synthesizing recent developments in machine learning, commonsense reasoning, and human-robot interaction, the paper explores how AI enables robots to perceive contextual cues, interpret human intent, and perform shared tasks with greater autonomy and safety.A mixed-methods approach is employed, combining a systematic review of peer-reviewed literature with simulation-based evaluations of AI-enabled cobot frameworks. Specific focus is placed on reinforcement learning, speech recognition, and semantic planning mechanisms that support robust human-machine collaboration. Key findings highlight the emergence of AI architectures that allow real-time task adaptation, reduction in programming overhead, and improved ergonomic outcomes in human-robot teams.The implications of these findings suggest a shift toward more human-centric, resilient manufacturing systems that leverage both human expertise and robotic precision. The study concludes by identifying technical challenges—such as trust, explainability, and standardization—and proposes directions for future interdisciplinary research in the context of Industry 5.0. Keywords collaborative robotics, artificial intelligence, smart manufacturing, human-robot interaction, reinforcement learning, commonsense reasoning, Industry 5.0, cognitive robotics, intelligent automation, human-centric systems

Research on Employee Innovation Ability in Human–Machine Collaborative Work Scenarios—Based on the Grounded Theory Construct of Chinese Innovative Enterprises

Against the backdrop of the booming digital economy, innovation has emerged as the core driving force for enterprise development, with employees’ innovative capabilities serving as a key competitive advantage for innovative enterprises. Adopting grounded theory as the methodological framework, we obtain multi-source data to investigate the factors influencing employees’ innovative capabilities and their underlying mechanisms. Furthermore, we develop a theoretical model elucidating the formation mechanism of employees’ innovative capabilities in human–machine collaboration contexts, identifying four core dimensions—innovation drivers, human–AI collaboration patterns, knowledge conversion pathways, and technological breakthroughs—that dominantly shape these capabilities. Thus, we reveal that the formation of innovative capabilities constitutes a dynamic interplay of technology empowerment, cognitive restructuring, and collaborative reinforcement and demonstrate its spiral progression characterized by “triggering, collaboration, and iteration”. This research not only contributes to academic discourse but also offers actionable theoretical and practical insights for innovative enterprises to enhance employees’ innovative capabilities, thereby fostering sustainable development in global competition.

The Effect of Machine Feedback on Foreign Language Writing

The advancement of artificial intelligence technology has made machine feedback an important tool for teaching foreign language writing. This article systematically explores the technical principles, historical development, and dual impact on writing quality through a literature review. Machine feedback is based on natural language processing (NLP), evolving from early spell checking to intelligent systems that support grammar, vocabulary, and logic optimization, significantly improving learners' vocabulary diversity (such as semantic word recommendation) and text accuracy (reducing grammar error rates by 20% -30%), and reinforcing learning motivation through instant feedback. However, research also reveals its limitations: excessive dependence leads to a decline in students' ability to write independently, algorithmic standardization ignores cultural context and creative expression, and even triggers academic ethical risks. Through analyzing international empirical research in the past decade, it has been found that machine feedback needs to be coordinated with teacher feedback to balance efficiency and critical thinking development. In the future, we should focus on the design of human-machine collaboration models and pay attention to the differentiated needs of learners with different language proficiency levels, promoting innovation and balanced development of foreign language writing education.

A Knowledge Push Method of Complex Product Assembly Process Design Based on Distillation Model-Based Dynamically Enhanced Graph and Bayesian Network

Under the paradigm of Industry 5.0, intelligent manufacturing transcends mere efficiency enhancement by emphasizing human-machine collaboration, where human expertise plays a central role in assembly processes. Despite advancements in intelligent and digital technologies, assembly process design still heavily relies on manual knowledge reuse, and inefficiencies and inconsistent quality in process documentation are caused. To address the aforementioned issues, this paper proposes a knowledge push method of complex product assembly process design based on distillation model-based dynamically enhanced graph and Bayesian network. First, an initial knowledge graph is constructed using a BERT-BiLSTM-CRF model trained with integrated human expertise and a fine-tuned large language model. Then, a confidence-based dynamic weighted fusion strategy is employed to achieve dynamic incremental construction of the knowledge graph with low resource consumption. Subsequently, a Bayesian network model is constructed based on the relationships between assembly components, assembly features, and operations. Bayesian network reasoning is used to push assembly process knowledge under different design requirements. Finally, the feasibility of the Bayesian network construction method and the effectiveness of Bayesian network reasoning are verified through a specific example, significantly improving the utilization of assembly process knowledge and the efficiency of assembly process design.

Theoretical Mechanism Research on the Shaping of Digital Economy in Digital Gig Work Participation

The digital economy, relying on the deep integration of platformization and digital technologies, has restructured the resource allocation logic of the labor market and accelerated the development of gig work participation. Technologies such as algorithms and big data significantly reduce the asymmetry of supply and demand information and transaction costs, enabling dynamic matching and efficient coordination of labor resources. This enhances flexibility in terms of time, space, and skills, providing workers with more autonomy. However, the labor market exhibits contradictory characteristics of "decentralized collaboration" and "platform-centered control." On the one hand, task decomposition and crowdsourcing models reconstruct complex jobs through modularization, activating global labor resources, lowering skill barriers, and improving human capital allocation efficiency. On the other hand, digital gig workers face multiple challenges, such as ambiguous labor relationships, income volatility, invisible algorithmic control, limited skill accumulation, and lack of social security. Platforms shift operational risks through algorithms, and workers, lacking legal protection and collective bargaining power, are forced to accept unequal terms, exacerbating the vulnerability of their rights and interests. To address these issues, this paper proposes the following optimization pathways: First, leverage artificial intelligence and blockchain technologies to achieve intelligent human-machine collaboration and transparency of labor rights; second, build a lifelong learning ecosystem to support skill upgrading and sustainable career development for gig workers; third, promote adaptive institutional design to clarify the legal status of "third-party workers" and establish a flexible social security mechanism. Through the dual drive of technological progress and institutional innovation, the digital economy is expected to improve resource allocation efficiency while promoting fairness and innovative development in the participation model of digital gig work.

Rethinking industrial modernity: Institutional heterogeneity and the limits of technological convergence

Industrial modernization defies the myth of technological convergence, instead fracturing into distinct pathways shaped by local institutional DNA. Our investigation reveals how governance architectures dynamically rewrite the rules of technological efficacy through four revolutionary shifts. Where traditional models predicted standardization, we observe German manufacturers transforming initial 14% blockchain adoption delays into 29% fewer contractual disputes within five years—a testament to institutional learning in action. Toyota’s breakthrough 19% reliability gain through Kaizen AI demonstrates institutions as living systems: by weaving hourly worker feedback into autonomous processes, they unlocked recombinant innovation that redefines human-machine collaboration. Regulatory landscapes expose irreducible trilemmas—China’s $2.3 million robotics recall costs versus Europe’s 59% small-business adoption gaps prove one-size-fits-all solutions are obsolete. From Sweden’s precision-crafted quality premiums to Foxconn’s hyper-optimized throughput, each successful model blooms from unique institutional soil. Crucially, Germany’s Autonomik initiative shows worker feedback efficacy varies across cultural contexts, while falling blockchain disputes reveal measurable adaptation curves. Volkswagen’s factories and Chinese supply chains aren’t converging; they’re diverging toward equally valid futures. This evidence dismantles the century-old pursuit of a universal “best way,” revealing institutional heterogeneity as the unexpected engine of 21st-century progress—where plural modernities thrive through continuous reinvention.

MetalMind: A knowledge graph-driven human-centric knowledge system for metal additive manufacturing

In the Industry 5.0 era, increasing manufacturing complexity and fragmented knowledge pose challenges for decision-making and workforce development. To tackle this, we present a human-centric knowledge system that integrates explicit knowledge from formal sources and implicit knowledge from expert insights. The system features three core innovations: (1) an automated KG construction pipeline leveraging large language models (LLMs) with collaborative verification to enhance knowledge extraction accuracy and minimize hallucinations; (2) a hybrid retrieval framework that combines vector-based, graph-based, and hybrid retrieval strategies for comprehensive knowledge access, achieving a 336.61% improvement over vector-based retrieval and a 68.04% improvement over graph-based retrieval in global understanding; and (3) an MR-enhanced interface that supports immersive, real-time interaction and continuous knowledge capture. Demonstrated through a metal additive manufacturing (AM) case study, this approach enriches domain expertise, improves knowledge representation and retrieval, and fosters enhanced human-machine collaboration, ultimately supporting adaptive upskilling in smart manufacturing.

Modes and Challenges of AIGC-Empowered Animation Short Film Creation A Case Study of Goodbye Rabbit

With the rapid development of Artificial Intelligence-Generated Content (AIGC) technology, the creation of film and animation is undergoing a paradigm shift. Taking the AIGC short film Goodbye Rabbit as a case study, this article systematically examines its application in pre-production planning, visual design, production processes, and post-production optimization, exploring the models and advantages of AIGC-empowered animation short film creation. Through case analysis, the article highlights AIGCs significant advantages in lowering creative barriers, enhancing production efficiency, and enriching audiovisual imagination. However, challenges such as ambiguous authorship, aesthetic homogenization, and unresolved copyright ownership are also identified. The study argues that AIGC should serve as an assistive creator to support human artistic expression. Future efforts should focus on optimizing human-machine collaboration mechanisms, technological ethics, and aesthetic standards to provide innovative momentum and value guidance for the animation industry. This research aims to offer theoretical and practical insights for AIGC-driven film creation and academic discourse.

A Study on the Tripartite Logic of AI Empowerment in Cultural Industry Development Theory Practice and Field

This paper grounded in the Marxist theory of the dialectical relationship between productive forces, production relations, and superstructure, systematically constructs a three-dimensional logical framework for the development of the cultural industry driven by artificial intelligence. The theoretical dimension elucidates the causal chain where intelligent technology restructures the cultural productive forces, innovates production relations, and fosters new paradigms in the superstructure. The practical dimension analyzes the "technology-culture-business" integration of new business models formed through the synergy of technological logic, data intelligence, and intelligent ecosystems. The field dimension focuses on three key areas: innovation and creation, heritage preservation, and international dissemination, proposing mechanisms for collaboration among government, enterprises, and society, as well as pathways for optimizing the value ecosystem. The study finds that artificial intelligence is driving the cultural industry to transition from an efficiency tool to a paradigm revolution, but it warns against the risks of algorithmic dominance and cultural alienation. It recommends building a "self-evolving" ecosystem of human-machine collaboration, achieving sustainable development through the dialectical unity of technological empowerment and cultural innovation.