China is both a major producer and consumer of fresh agricultural products, making cold chain logistics essential for preserving quality and reducing post-harvest loss. However, insufficient pre-cooling capacity in production areas often leads to significant quality deterioration during the first-mile stage, which has not been fully addressed in existing cold chain network design studies. To bridge this gap, this study proposes an integrated optimization framework for designing a first-mile pre-cooling distribution center (DC) network. A multi-objective nonlinear mathematical model is developed to simultaneously minimize total logistics cost and maximize product freshness. To better characterize perishability, a stage-specific freshness decay function captures the nonlinear deterioration of products before and after pre-cooling. Transportation-related carbon emissions are also incorporated to enhance environmental relevance. Given the complexity of the location-routing problem, a genetic algorithm (GA) is used to obtain Pareto-optimal solutions. An empirical case study in Shandong Province, China, is conducted under three scenarios: (1) no pre-cooling, (2) decentralized pre-cooling at origins, and (3) centralized pre-cooling at regional DCs. Results show that the centralized strategy achieves superior performance, reducing total daily cost by 3.79% and producing the lowest freshness loss compared with the no-pre-cooling baseline. In contrast, decentralized origin-side pre-cooling improves freshness preservation but increases total cost by 5.27% due to higher equipment investment and weaker route efficiency. These findings demonstrate that an integrated location-routing perspective can provide more effective first-mile cold chain planning than treating pre-cooling as an isolated facility decision.
Chen et al. (2026) studied this question.