Integrating Artificial Intelligence and Internet of Things for Next-Generation Intelligent Smart City Infrastructure: A Layered Review of Architectures, Analytics, Domain Readiness, and Governance Challenges
Abstract
ABSTRACT
Rapid urbanisation is placing unprecedented stress on transport, energy, water, environmental, and public-safety systems, and the Internet of Things (IoT) now allows these systems to be instrumented at fine spatial and temporal resolution. Sensing alone, however, does not make a city intelligent; value emerges only when heterogeneous sensor streams are converted into timely, trustworthy, and actionable decisions through Artificial Intelligence (AI). This review-based study examines how AI and IoT are being integrated to build next-generation smart city infrastructure. A structured thematic synthesis of 44 peer-reviewed articles, surveys, policy documents, and methodological sources was organised through a five-layer analytical framework spanning perception, connectivity, edge–fog, cloud data platform, and application layers, with a cross-cutting security and governance plane. The synthesis maps AI techniques to urban domains, appraises six domains against five readiness criteria on an ordinal 1–5 scale, and links recurring barriers to mitigation strategies. Traffic management and energy systems showed the highest deployment readiness, whereas public safety and healthcare services were constrained mainly by privacy manageability and data governance. Edge intelligence and federated learning emerged as the most consistent architectural enablers, reducing latency and raw-data exposure simultaneously. The paper closes with a practical integration roadmap and research directions relevant to Indian and other emerging-economy cities.
KEYWORDS: Artificial intelligence, Internet of Things, Smart city, Edge intelligence, Federated learning, Urban analytics, Smart infrastructure, Data privacy, Digital twin, Sustainable urban systems
Rapid urbanisation is placing unprecedented stress on transport, energy, water, environmental, and public-safety systems, and the Internet of Things (IoT) now allows these systems to be instrumented at fine spatial and temporal resolution. Sensing alone, however, does not make a city intelligent; value emerges only when heterogeneous sensor streams are converted into timely, trustworthy, and actionable decisions through Artificial Intelligence (AI). This review-based study examines how AI and IoT are being integrated to build next-generation smart city infrastructure. A structured thematic synthesis of 44 peer-reviewed articles, surveys, policy documents, and methodological sources was organised through a five-layer analytical framework spanning perception, connectivity, edge–fog, cloud data platform, and application layers, with a cross-cutting security and governance plane. The synthesis maps AI techniques to urban domains, appraises six domains against five readiness criteria on an ordinal 1–5 scale, and links recurring barriers to mitigation strategies. Traffic management and energy systems showed the highest deployment readiness, whereas public safety and healthcare services were constrained mainly by privacy manageability and data governance. Edge intelligence and federated learning emerged as the most consistent architectural enablers, reducing latency and raw-data exposure simultaneously. The paper closes with a practical integration roadmap and research directions relevant to Indian and other emerging-economy cities.
KEYWORDS: Artificial intelligence, Internet of Things, Smart city, Edge intelligence, Federated learning, Urban analytics, Smart infrastructure, Data privacy, Digital twin, Sustainable urban systems