<p>In modern air defense systems, the singular defensive capabilities of individual platforms fall short in meeting the demands of complex and ever-changing air defense operations. This paper delves into the investigation of the “weapon-target assignment” challenge for multi-platform missile interception in uncertain and rapidly evolving battlefield environments. Factors encompassing weapon task load balancing, damage assessment, interception cost, among others, are comprehensively considered. Consequently, novel models for task assignment and target threat estimation are introduced. To address this multifaceted problem, a layered approach to task allocation is proposed by integrating simulated annealing with Metropolis criterion, large perturbation, and an elite pool mechanism into a genetic algorithm framework. This approach effectively integrates an improved hybrid single-parent genetic algorithm with the multi-agent theory contract network algorithm. Through thorough simulation experiments, the convergence of this algorithm in the Weapon-Target Assignment (WTA) scenario is numerically verified. Comparative analysis with traditional genetic algorithms and simulated annealing methods underscores the efficacy of the proposed approach in facilitating holistic mission planning under emergent circumstances, as well as the dynamic adjustment of localized task assignments. Thus, this research presents a promising advancement in enhancing the overall effectiveness and adaptability of multi-platform air defense systems.</p>