SUSTAINABLE MULTIMODAL TRANSPORTATION WITH A CONTEXT-AWARE ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM AND GENETIC ALGORITHMS

Multimodal transportation faces important challenges related to sustainability, efficiency, and equity, especially in rapidly growing urban economies such as China.To address these, this study introduces a friendly, context-aware optimization model.It combines the Adaptive Neuro-Fuzzy Inference System (ANFIS) for accurate predictions with a Genetic Algorithm (GA) that helps select the best routes and modes of transport.The model can estimate transportation costs and Carbon dioxide (CO2) emissions using real-time data such as distance, cargo weight, traffic, weather, and vehicle capacity.Meanwhile, GA works to find optimal routes while considering factors such as cost, traffic, emissions, fairness, and resilience, making the entire process more effective and sustainable.The model was validated using 2022-2023 data from China's Ministry of Transport, showing impressive predictive accuracy (R > 0.89) and surpassing traditional static models.The results are encouraging, a reduction in CO emissions is witnessed 17% below, with truck usage decreasing from 25% to below 10%, and a remarkable 150% increase in rail transport.A combination of railways and waterways proved to be the most sustainable and efficient transport option, highlighting the potential for greener and more effective freight movement.This study shows exciting possibilities for practical use in emission pricing, intermodal planning, and resilience strategies, opening doors for more effective and responsive policies.