Integrating Artificial Neural Networks and Six Sigma Metrics Operating Curves for Sustainable Quality Evaluation and Projection in Passenger and Freight Transportation Systems

This article proposes a new integrated methodological procedure to evaluate and project the quality and operational sustainability performance of an integrated passenger and freight transportation system. Grounded in a basic research approach, the core innovation lies in the mathematical formulation of Six Sigma metrics operating curves and their synergistic integration with artificial neural networks (ANN) to minimize operational variability, a key driver of resource waste, excess fuel consumption, and environmental emissions in transport networks. By transitioning quality management from a reactive diagnostic phase to an intelligent, predictive paradigm, the model stabilizes transport operations, directly contributing to economic and environmental sustainability. The results demonstrate that the optimal operational conditions are achieved at a global Sigma Z level of 4.45, reducing non-conformities by 97.7% and stabilizing the system with a 99.85% global yield. Furthermore, the proposed 3-3-9 multilayer perceptron successfully projects process quality with 95% reliability. This integrated framework offers a powerful, replicable tool for decision-makers to enhance service quality while driving sustainable, resource-efficient operations in logistics and transport sectors.