Optimizing Large-Scale Small UAV Procurement: A Multi-Strategy-Enhanced NSGA-III Approach

Large-scale procurement of small UAVs to perform specific missions has become an important pathway to enhancing operational capability. To address the challenges of multi-objective conflicts, complex constraints, and large-scale variable coupling in procurement decision-making, this paper constructs a multi-objective procurement decision model covering system-of-systems operational effectiveness, life cycle cost, operational readiness rate, and allocation cycle, systematically incorporating practical hard constraints such as authorized establishment and allocation, security clearance, and domestic content rate. To efficiently solve this model, a multi-strategy fusion-based many-objective optimization algorithm (MSF-NSGA-III) is proposed. The algorithm designs a three-segment hybrid encoding and intelligent repair mechanism to ensure the feasibility of the solution space; integrates particle swarm learning and a strengthened dominance relation to enhance selection pressure in the high-dimensional objective space; and introduces adaptive crossover and mutation operators to balance the exploration and exploitation capabilities of the population. Simulations on large-scale instances demonstrate that MSF-NSGA-III outperforms the other five mainstream benchmark algorithms in core metrics such as hypervolume (HV) and inverted generational distance (IGD), and ablation experiments confirm that the proposed improvement strategies effectively enhance the convergence accuracy and distribution uniformity of the solution set. Case analysis verifies that the model can generate a high-quality non-dominated solution set and provide multi-objective balanced decision support for small UAV procurement; sensitivity analysis indicates that, within the tested parameter intervals, MSF-NSGA-III has low sensitivity to key parameters and exhibits good parameter robustness.