Focusing on investment decision-making challenges within the textile industry, this study addresses the significant financial risks arising from high volatility in textile raw material prices and uncertain market demand for textile products. Financial management of textile design projects is enhanced by constructing a dual-objective optimization model to simultaneously maximize net present value (NPV) and minimize conditional value at risk (CVaR). The model utilizes a comprehensive risk index to characterize multisource risk factors specific to textile manufacturing. To solve this complex investment optimization problem in the textile sector, this paper proposes an improved multi-objective particle swarm optimization (R-MOPSO) algorithm with a dynamic risk perturbation and crowding distance guidance mechanism. Validation experiments based on operational data from a large-scale yarn textile enterprise show that the R-MOPSO algorithm achieves a superior Pareto solution set, where the NPV can reach 3.884 million CNY while controlling CVaR within 913,000 CNY. Sensitivity analysis further confirms the model's robustness, revealing that a 15% increase in textile raw material price leads to a 5.9% decrease in NPV and a 9.1% increase in CVaR. Results demonstrate that the proposed R-MOPSO is a robust and effective intelligent decision-making tool specifically tailored for capital investment projects in the highly volatile textile industry.

investment risk control, yarn production, multi-objective particle swarm optimization algorithm, textile design project evaluation, textile industry