Traffic density-aware multi-objective edge server deployment for vehicular edge computing

Vehicular edge computing (VEC) has been introduced as a promising paradigm to support latency-sensitive and computation-intensive vehicular applications. However, the dynamic mobility of vehicles and uneven traffic distribution bring great challenges in efficient edge server deployment. Inappropriate positioning of the edge servers may increase the communication latency, energy consumption, service interruptions, and handover frequency and thus degrade the overall network performance. To address these challenges, this paper proposes a traffic density-aware multi-objective edge server deployment framework for VEC environments based on the traffic density-aware non-dominated sorting genetic algorithm II (TDA-NSGA-II). The proposed framework exploits realistic simulation of urban mobility (SUMO)-generated vehicular mobility traces to identify high-density traffic regions and optimize edge server locations accordingly. The optimization process jointly minimizes communication latency and energy consumption while improving coverage and reducing handovers. In addition, the proposed scheme implements mobility-aware vehicle-to-server association to provide stable connectivity under dynamic vehicular movement. Experimental evaluations are conducted based on realistic traffic mobility scenarios and compared with conventional optimization approaches such as genetic algorithm (GA), particle swarm optimization (PSO) and ant colony optimization (ACO). Simulation results presented in this paper demonstrate that the proposed TDA-NSGA-II framework achieves substantially lower latency and energy consumption than GA, PSO and ACO, with coverage comparable to the best baseline, while handover counts remain comparable rather than consistently reduced. The results illustrate that traffic-density-aware edge server deployment can significantly improve the efficiency and reliability of VEC networks. The proposed framework provides an effective solution for adaptive edge server deployment in dynamic vehicular edge computing environments and can support latency-sensitive intelligent transportation applications through efficient resource utilization and improved communication reliability.