Multi-Objective Chance-Constrained Programming of Railway Cold Chain Logistics

Abstract:

Aimed at the lowest total cost， carbon emission， and freight loss rate of transportation network， an optimization model of railway cold chain logistics network with stochastic demand and transportation capacity constraints was established. The model is transformed deterministically using the chance-constrained programming theory， and an adaptive genetic-simulated annealing algorithm （A-SAGA） is designed. The optimal cold chain logistics transportation scheme set is obtained by MATLAB simulation calculation. The sensitivity analysis of freight subsidy， train travel speed and freight category are conducted. The simulation results show that under the multiple conflicting objectives of considering transportation costs， carbon emissions， and cargo loss rates， rail cold chain market share can be improved significantly when 5% of the freight subsidy is applied， or only 70 km?h^-1 of train travel speed is increased. These two methods can be used flexibly under different infrastructure conditions. Cargo with varying degrees of sensitivity to transit time directly affects the change in the rate of cargo loss and the initial value of rail market share.