Robust optimization approach to two-echelon agricultural cold chain logistics considering carbon emission and stochastic demand

Cold chain logistics has become one of the main sources of carbon emissions. Meanwhile, the implementation of low-carbon economy has become an inevitable way to promote sustainable development. However, previous studies on the cold chain inventory routing problem (IRP) paid less attention to the cost of carbon emissions. In this paper, a linear programming (LP) model is established, which takes the costs of vehicle transportation, time window and carbon emission into consideration. Although the simple LP model is easy to be solved, it cannot handle the problems with uncertainty. Therefore, in order to overcome the influence of uncertainty, the proposed LP model is developed into three low-carbon robust optimization (RO) models. In addition, this paper takes a cold chain logistics enterprise in Yangtze River Delta as an example for empirical analysis. The results of the case study prove that the RO models can quickly solve the problems with uncertainty and still maintain robustness, while the LP model has failed. Specifically, the R-ellipsoid model produces the best result among the three RO models. It is suggested that when the carbon emission tax increases, the decision makers tend to choose a better path planning scheme, which will not only reduce the total cost, but also obtain environmental benefits. Finally, the findings of this paper generate some implications for the low-carbon transformation of cold chain logistics enterprises.