High-Performance Computing for Distributed Route Computation in Traffic Flow Models

Abstract

In the dynamic landscape of smart cities and traffic management, it is necessary to further explore the synergistic potential of realtime traffic data and high-performance computing to optimise traffic flow through dynamic re-routing strategies. High-performance computing plays an essential role in achieving effective traffic flow optimisation. Our research builds upon existing studies highlighting the positive correlation between the integration of live traffic updates and route optimisation. The methodology involves simulations with our Ruth traffic simulator, where vehicles dynamically adjust routes based on up to date traffic information available to them at different levels. Scalability tests are conducted with varying numbers of CPUs and nodes to assess the simulator's capacity to scale. The results showcase the impact of live traffic data on both driving time and average speed, emphasising the adaptability of our approach for broader applications. In conclusion, our work not only advances the understanding of real-time traffic optimisation but also underscores the critical role of high-performance computing in achieving scalable solutions. The findings present practical implications for the implementation of dynamic re-routing strategies in transportation systems, paving the way for future research and real-world applications on smart cities.