Research on a Multi-Objective Optimization Model for Traffic Distribution on Short Video Platforms

With the rapid growth of short video platforms, traffic distribution has become a critical determinant of content exposure, user engagement, and ecosystem sustainability. Traditional single-objective optimization methods, which primarily focus on metrics such as click-through rate or viewing duration, are insufficient to balance competing goals including user experience, creator fairness, content diversity, and platform revenue. This study develops a multi-objective optimization model for traffic allocation on short video platforms. Based on clearly defined assumptions, the model incorporates objectives such as maximizing engagement and revenue while promoting fairness and diversity. The weighted sum method is applied to analyze outcomes under different strategic preferences, and the NSGA-II algorithm is introduced to obtain the Pareto optimal solution set and reveal trade-off boundaries among objectives. Comparative results demonstrate that the weighted method is computationally efficient and suitable for scenarios with explicit preferences, whereas NSGA-II better captures the interplay among multiple objectives and generates diversified strategic options. The study provides theoretical and practical insights for achieving balanced and sustainable platform governance, offering a systematic framework that platform operators and policymakers can adopt to design more equitable and efficient content recommendation mechanisms.