Search

Distributed AI-Enhanced Priority Queue for Real-Time Smart Transportation Systems: A Scalable Architecture for Adaptive Traffic Management

Author(s)

Chen Yang*

Affiliation(s)

College of Science and Engineering, University of Glasgow, Glasgow, United Kingdom *Corresponding Author

Abstract

We propose a distributed AI-enhanced priority queue (DAIPQ) architecture for real-time smart transportation systems (STS), addressing the critical challenge of scalable and adaptive traffic management in dynamic urban environments. The proposed method integrates a randomized approximate nearest neighbor (ANN) prioritization engine with locality-sensitive hashing (LSH) to efficiently rank traffic events based on spatial, temporal, and contextual features, enabling sublinear-time retrieval of high-urgency incidents. A hierarchical concurrent priority queue then enforces a relaxed fairness scheme across geographical partitions, balancing load distribution while preserving locality-aware scheduling. Furthermore, the system incorporates a multi-tier caching layer with learned indexing to accelerate event processing, dynamically adjusting priority weights based on cache availability. The architecture is designed to operate at the edge-cloud continuum, where lightweight feature extraction models preprocess raw sensor data from LiDAR, cameras, and V2X communications into compact representations. Experimental validation demonstrates that DAIPQ achieves significant improvements in latency-sensitive traffic control tasks, particularly in scenarios with high event throughput and heterogeneous urgency profiles. The modular design ensures compatibility with existing infrastructure, while differential privacy mechanisms safeguard user data in navigation recommendations. This work contributes a principled framework for real-time decision-making in intelligent transportation systems (ITS), offering practical solutions for scalability and adaptability in large-scale deployments.

Keywords

Distributed AI-enhanced Priority Queue (DAIPQ); Smart Transportation Systems; Real-Time Priority Scheduling; Edge-Cloud Continuum.

References

[1]T Thunig, R Scheffler, M Strehler & K Nagel (2019) Optimization and simulation of fixed-time traffic signal control in real-world applications. Procedia Computer Science. [2]G Dimitrakopoulos, et al. (2010) Intelligent transportation systems. In Ieee Vehicular Technology Conference. [3]DA Tedjopurnomo, Z Bao, B Zheng, et al. (2020) A survey on modern deep neural network for traffic prediction: Trends, methods and challenges. In IEEE International Conference on Knowledge and Systems Engineering. [4]O Lemeshko, O Yeremenko, L Titarenko & A Barkalov (2023) Hierarchical queue management priority and balancing based method under the interaction prediction principle. Electronics. [5]W Li, Y Zhang, Y Sun, W Wang, M Li, et al. (2019) Approximate nearest neighbor search on high dimensional data—experiments, analyses, and improvement. In IEEE International Conference on Knowledge and Information Fusion. [6]G Bovenzi, G Aceto, D Ciuonzo, et al. (2020) A big data-enabled hierarchical framework for traffic classification. In International Conference on Network Science. [7]AK Nori (2010) Distributed caching platforms. In Proceedings of the VLDB Endowment. [8]J Zhang (2020) An investigation of smart transportation system (STS) data integration within Chinese cities: A socio-technical system perspective. etheses.whiterose.ac.uk. [9]MU Tariq (2024) Smart transportation systems: Paving the way for sustainable urban mobility. Contemporary Solutions for Sustainable Transportation. [10]T Gong, L Zhu, FR Yu & T Tang (2023) Edge intelligence in intelligent transportation systems: A survey. IEEE Transactions on Intelligent Transportation Systems. [11]Z Xie, C Ji, L Xu, M Xia & H Cao (2023) Towards an optimized distributed message queue system for AIoT edge computing: a reinforcement learning approach. Sensors. [12]Q He, KL Head & J Ding (2011) Heuristic algorithm for priority traffic signal control. Transportation research record. [13]A Louati, S Elkosantini, S Darmoul & H Louati (2018) Multi-agent preemptive longest queue first system to manage the crossing of emergency vehicles at interrupted intersections. European Transport Research Review. [14]AYAB Ahmad, N Verma, NM Sarhan, EM Awwad, et al. (2024) An IoT and blockchain-based secure and transparent supply chain management framework in smart cities using optimal queue model. In IEEE 2024 16th International Conference on Ubiquitous and Future Networks (ICUFN). [15]I Sanchez, ZMM Aye, BIP Rubinstein, et al. (2016) Fast trajectory clustering using hashing methods. In IEEE International Conference on Big Data. [16]MF Iqbal, M Zahid, D Habib, et al. (2019) Efficient prediction of network traffic for real‐time applications. Journal of Computer Networks and Communications. [17]B Xu, Y Wang, Z Wang, H Jia & Z Lu (2021) Hierarchically and cooperatively learning traffic signal control. In AAAI Conference on Artificial Intelligence. [18]Z Chen, J Liang, Z Yu, H Cheng, et al. (2024) Resilient collaborative caching for multi-edge systems with robust federated deep learning. Ieee/Acm Transactions on Internet of Things. [19]J Vuillemin (1978) A data structure for manipulating priority queues. Communications of the ACM. [20]S Edelkamp, A Elmasry & J Katajainen (2017) Optimizing binary heaps. Theory of Computing Systems. [21]AJ Smith (1982) Cache memories. ACM Computing Surveys (CSUR). [22]N Megiddo & DS Modha (2004) Outperforming LRU with an adaptive replacement cache algorithm. Computer. [23]M Fomitchev & E Ruppert (2004) Lock-free linked lists and skip lists. In Proceedings of the Twenty - Third Annual ACM Symposium on Principles of Distributed Computing. [24]C Hedges & F Perry (2008) Overview and use of sae j2735 message sets for commercial vehicles. sae.org. [25]SK Swain & PK Nanda (2021) Adaptive queue management and traffic class priority based fairness rate control in wireless sensor networks. IEEE Access. [26]S Arya, T Malamatos & DM Mount (2009) Space-time tradeoffs for approximate nearest neighbor searching. Journal of the ACM (JACM).