Simulation and optimization of cold chain logistics system towards lower carbon emission: a state-of-the-art review

Abstract Cold chain logistics is an emerging carbon emission source. The transportation section is estimated to generate more than 80% of its total carbon emission. Under the global warming background, a number of simulation and optimization studies have been thus conducted. Lately, cold chain logistics has been greatly aided by the quick advancements in data science and computer science. Consequently, the state-of-the-art developments in cold chain logistics system simulation and optimization were critically assessed in this research. The crucial indicators including optimization models, simulation models, and optimization algorithms were also quantitatively analyzed. It was discovered that, in single-objective optimization, carbon emissions were frequently included in the total cost objective function; but, in multi-objective optimization, they were utilized as a separate objective function. There were many studies on path optimization models, but site selection optimization models started relatively late and had high economic sensitivity, which might be a future trend. The ant colony algorithm was widely used in solving path optimization models. Genetic algorithms played a leading role in optimization algorithms, and they had different improvements and applications under different carbon reduction methods. Under the influence of multiple objective functions and constraints, multi-objective optimization algorithms might be a promising solution. The difficulties in modeling and refining the cold chain logistics system are introduced by the model accuracy, energy supply system revolution, carbon emission calibration, optimization result realization, and application scenarios that still exist. The outlook for the future is presented in terms of both technology orientation and demand orientation.