Enhanced Black-Winged Kite Algorithm for Agile Software Project Scheduling Optimization

To address the challenges of multi-objective optimization and dynamic load balancing in agile software project scheduling, this paper proposes an improved Black-winged Kite Optimization algorithm—the Lens-imaging Golden Sine Black-winged Kite Algorithm (LGBKA). This algorithm enhances population diversity through Chebyshev chaotic mapping, strengthens global exploration capabilities via a lens-imaging reverse learning strategy, and optimizes local exploitation efficiency by integrating a golden sine strategy, thereby balancing the multi-objective optimization problem of maximizing user story value while minimizing developer load disparity. Experimental results demonstrate that the proposed LGBKA exhibits outstanding convergence accuracy (reflected by average fitness values) and stability (measured through standard deviations) on the CEC2005 benchmark functions (F1, F10, F15), significantly outperforming the Black-winged Kite algorithm, Whale Optimization Algorithm, Grey Wolf Optimizer, and Particle Swarm Optimization. In agile scheduling case studies, the LGBKA achieved a 0.156 improvement in user story value points and a 0.234 reduction in developer workload differences, demonstrating its robustness and practicality in dynamic requirement scenarios.