Cold Chain Network Design

Cold Chain Network Design addresses where to place infrastructure to minimize costs while maintaining the integrity of temperature-sensitive goods.

Facility Location with Cold Storage Costs

The core mathematical model is an extension of the Capacitated Facility Location Problem (CFLP). Cold storage facilities have significantly higher fixed setup costs ( $f_j$ ) and variable operating costs ( $v_j$ ) due to refrigeration.

\min \sum_{j \in J} f_j y_j + \sum_{i \in I} \sum_{j \in J} c_{ij} x_{ij} + \sum_{j \in J} v_j \sum_{i \in I} x_{ij}

Subject to capacity constraints and the requirement that all customer demand $i$ is met.

Network Topologies for Perishables

Direct-Ship: Best for highly perishable, high-volume goods (e.g., direct from farm to processing).
Hub-and-Spoke: Standard retail distribution, but adds transit time.
Cross-Docking: Goods are unloaded from inbound trucks and loaded directly onto outbound trucks with little to no storage. This is ideal for perishables as it minimizes holding time.

Temperature-Zone Consolidation

Designing a multi-temp Distribution Center (DC) requires partitioning the warehouse into specific zones (e.g., ambient, chilled, frozen). Optimization must balance the footprint of each zone with peak seasonal demands.

Stochastic Network Design

Because demand for fresh food is highly volatile (see FreshFoodDemandForecasting), network design must be robust under uncertainty. Two-stage stochastic programming is commonly used, where first-stage variables determine facility locations, and second-stage variables determine flows across scenarios.

Carbon Footprint vs. Freshness

Cold chains are energy-intensive. There is a fundamental trade-off between maximizing freshness (fast, frequent deliveries of small batches) and minimizing carbon footprint (large, consolidated, slower shipments). This creates a Pareto-optimal frontier where supply chain managers must choose a point that aligns with corporate sustainability goals. The emission function $E$ depends on distance, vehicle type, and refrigeration power.

Case Study: Regional Fresh Produce Distribution

Worked Example: A cooperative of 50 apple orchards needs to supply 200 regional supermarkets. Data:

Orchard supply: 500 tons/week.
Fixed cost of a cold storage hub: $2M/year.
Transit cost: $0.50/ton-mile. Using a GIS-based accessibility analysis and solving the CFLP, the optimization reveals that 3 cross-docking hubs placed centrally among the supermarkets minimizes total cost to $5.4M/year while ensuring apples spend no more than 48 hours in transit. Replacing one hub with direct shipping reduced transit time by 12 hours but increased transport emissions by $30\%$ , demonstrating the carbon-freshness trade-off. This structural decision heavily constrains the PerishableVehicleRouting problem.