<p>Transportation decisions in humanitarian relief must be made quickly, cost-effectively, and in an environmentally friendly manner under very high uncertainty. In this study, we consider a multi-level solid transportation problem under an uncertain environment (MLSTPUE) that jointly accounts for carbon emissions and route risk in the transportation network. The analysis includes uncertain emergency response times, route-failure uncertainty rates, (Liu-normal) uncertain carbon emissions, setups, and demands. A chance-constrained transhipment model is established to transform all uncertain parameters into their expected values, and a two-level structure is adopted to optimise three priority objectives in sequence iteratively: emergency response time, route-failure risk, and carbon emissions. The model is solved using a lexicographic optimisation algorithm, and its efficiency is illustrated through a realistic numerical example that includes three relief hubs, four affected areas, and two transportation modes. Results indicate that the model provides a practical, feasible transportation plan that can be executed under complex, uncertain conditions. The optimal solution achieves a minimum response time of 2332.37&#xa0;h, reduces the total risk value to 20.748, and generates a carbon footprint of 3259.08&#xa0;kg CO₂ during the evaluation trial. A sensitivity analysis of the confidence level α shows that a looser tolerance to uncertainty results in longer response times, greater risk exposure, and higher emissions, with trade-offs between robustness and operational performance that are important to quantify. The model presented here has the potential to be a valuable, practical decision-making tool for humanitarian logistics planners during disaster events, enabling more dependable, safer, and greener relief distribution.</p>