Generative AI and Large Language Models in Industry 5.0: Shaping Smarter Sustainable Cities

Generative AI and Large Language Models in Industry 5.0: Shaping Smarter Sustainable Cities: Comparison

Please note this is a comparison between Version 2 by Dean Liu and Version 1 by Giulio Salierno.

This review paper examines how Generative AI (GAI) and Large Language Models (LLMs) can transform smart cities in the Industry 5.0 era. Through selected case studies and portions of the literature, we analyze these technologies’ impact on industrial processes and urban management. The paper targets GAI as an enabler for industrial optimization and predictive maintenance, underlining how domain experts can work with LLMs to improve municipal services and citizen communication, while addressing the practical and ethical challenges in deploying these technologies. We also highlight promising trends, as reflected in real-world case studies ranging from factories to city-wide test-beds and identify pitfalls to avoid. Widespread adoption of GAI still faces challenges that include infrastructure and lack of specialized knowledge as a limitation of proper implementation. While LLMs enable new services for citizens in smart cities, they also expose certain privacy issues, which we aim to investigate in this study. Finally, as a way forward, the paper suggests future research directions covering new ethical AI frameworks and long-term studies on societal impacts. Our paper is a starting point for industrial pioneers and urban developers to navigate the complexity of GAI and LLM integration, balancing the demands of technological innovation on one hand and ethical responsibility on the other.

generative artificial intelligence (GAI)
large language models (LLMs)
smart cities

The digital transformation of industry is entering a new phase, where the shift from Industry 4.0 to Industry 5.0 is not fast moving but rather a paradigm shift in the human–technology relationship. Industry 4.0 is characterized by automation, cyber-physical systems, and the Internet of Things (IoT), whereas Industry 5.0 emphasizes human centrality (as opposed to merely a human-in-the-loop), sustainability, and resilience as guiding principles for industrial innovation [1,2]^[1][2]. Recent studies indicate that this transition is reshaping not only manufacturing processes but also urban development patterns [3,4]^[3][4]. The Industry 5.0 approach focuses on human feedback in conjunction with AI and robotics to enable a more individualized, efficient, and sustainable production process ^[5]. Generative Artificial Intelligence (GAI), together with Large Language Models (LLMs), is one of the emerging technologies that is accelerating this trend.

GAI implies AI systems that can create original contents (that is, images, text, designs, etc.) from existing data ^[6]. LLMs are generally based on deep learning and transformer-based architectures. The GAI and LLM integration is commonly deployed in design automation to facilitate product development and improve innovation processes, thus helping the introduction of new products (NPI) and production processes. Moreover, it supports [2,7]^[2][7] the existence of creative industries and manufacturing systems. Specifically, the complex and dynamic output generation capability of GAI and LLMs in real time is transforming industries to applications such as predictive maintenance, operational efficiency, and smart urban infrastructure.

This transformation is particularly significant in urban environments, where cities consume approximately 75% of global resources and generate 80% of GDP ^[8]. In this paper, we review the utilization of Generative Artificial Intelligence (GAI) and Large Language Models (LLMs) in various fields ranging from industrial use cases to city services. Based on numerous case studies and quantitative data, we explore the ways in which these technologies are changing old industrial practices or public goods. We utilize not only quantitative metrics but also qualitative insights from stakeholders on performance, which allows for a holistic view of both technical and social performance. Through a comprehensive analysis of AI Watch observatory data, municipal documentation, and published implementation reports, we examine GAI and LLM deployments across three European cities during January–June 2024. This analysis of existing implementations provides valuable information on both the challenges faced and the successful integration strategies used.

Key areas of focus include the following:

Using GAI for predictive maintenance and process optimization in industries.
LLM applications in citizen services and urban planning.
Technical infrastructure requirements and integration frameworks for such migrations.
Privacy considerations and ethical implications.
Economic and operational impacts of implementation.
Future research directions and recommendations for sustainable development.

Through the lens of such considerations, we hope to open up practical pathways for industrial innovators and urban developers alike to mitigate this complexity as GAI and LLMs are integrated into our society along with technological advancement whilst respecting ethical considerations as well. These results indicate that the successful deployment of such a system may require not only strong technical underpinnings but also more subtle human considerations and adroit management of privacy and broader societal impacts in the long term. This paper makes several significant contributions to the understanding and implementation of GAI and LLMs in smart city contexts. First, we establish a novel theoretical framework that bridges Industry 5.0’s human-centric principles with smart city operations through GAI and LLM technologies. This integration is particularly evident in our analysis of how predictive maintenance systems and citizen communication platforms can coexist within a unified technological ecosystem. Second, we provide concrete evidence from European cities that shows how GAI and LLMs can improve both industrial efficiency and urban service delivery, as shown in our case studies of experiments in Barcelona, Copenhagen, and Vienna. Third, we develop a comprehensive technical framework for implementing these technologies in mid-sized cities (200,000–500,000 inhabitants), addressing practical challenges such as multilingual support, system integration, and scalability. Our framework specifically emphasizes the role of GAI in predictive maintenance and LLMs in citizen services, providing replicable implementation patterns for similar urban environments.

The paper is organized as follows. Section 2 describes the methodology of this work. Section 3 introduces case studies of applications of GAI and LLMs, measuring positive impacts as well as their limitations. Section 4 proposes a framework for integrating GAI and LLMs into smart cities. Section 5 discusses privacy and ethical implications of adoption GAI and LLM technologies. Section 6 sketches our conclusions and future directions.

References

Lee, J.; Su, H. A unified industrial large knowledge model framework in Industry 4.0 and smart manufacturing. Int. J. AI Mater. Des. 2024, 1, 41.
Kaigom, E.G. Potentials of the Metaverse for Robotized Applications in Industry 4.0 and Industry 5.0. Procedia Comput. Sci. 2024, 232, 1829–1838.
Sheikh, R.A.; Ahmed, I.; Faqihi, A.Y.A.; Shehawy, Y.M. Global Perspectives on Navigating Industry 5.0 Knowledge: Achieving Resilience, Sustainability, and Human-Centric Innovation in Manufacturing. J. Knowl. Econ. 2024.
Jiang, Y.; Chao, Q.; Chen, Y.; Li, X.; Liu, S.; Cong, G. UrbanLLM: Autonomous Urban Activity Planning and Management with Large Language Models. In Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, FL, USA, 12–16 November 2024; Al-Onaizan, Y., Bansal, M., Chen, Y.N., Eds.; Association for Computational Linguistics: Stroudsburg, PA, USA, 2024; pp. 1810–1825.
Tóth, A.; Nagy, L.; Kennedy, R.; Bohuš, B.; Abonyi, J.; Ruppert, T. The human-centric Industry 5.0 collaboration architecture. MethodsX 2023, 11, 102260.
Sengar, S.S.; Hasan, A.B.; Kumar, S.; Carroll, F. Generative artificial intelligence: A systematic review and applications. Multimed. Tools Appl. 2024.
Ullah, A.; Qi, G.; Hussain, S.; Ullah, I.; Ali, Z. The Role of LLMs in Sustainable Smart Cities: Applications, Challenges, and Future Directions. arXiv 2024, arXiv:2402.14596.
United Nations Human Settlements Programme. World Cities Report 2022: Envisaging the Future of Cities; Technical Report; UN-Habitat: Nairobi, Kenya, 2022.