In the context of the strong digitalization of Industry 4.0, Machine Vision (MV) plays a key role in improving production efficiency and enterprise responsiveness. However, selecting appropriate MV projects is challenging due to incomplete data, volatile operating environments, and differences in expert assessments. This study proposes a new integrated decision-making model combining Picture Fuzzy Distance Measure (PF-DM), Picture Fuzzy Ranking Comparison (PF-RANCOM), and Picture Fuzzy Additive Ratio Assessment (PF-ARAS) to comprehensively assess and rank MV project risks, balancing both objective and subjective weights. The model is applied to 3 MV investment options in a Vietnamese industrial manufacturing enterprise, with 11 risk criteria. The results show that the P2 method achieves the highest utility score (0.9697), the equilibrium scenario with ζ = 0.5 was chosen as the baseline case because it simultaneously and harmoniously reflects both sources of information: subjective expert evaluation and objective data dispersion, thereby better representing the actual decision-making context in Machine Vision project selection, and remains stable in the sensitivity analysis, outperforming the PF-VIKOR method in terms of sharpness and ranking accuracy. The study provides an effective decision support tool, contributing to improving risk management and investment capacity in MV projects.