The management of perishable product supply chains requires rapid distribution and flexible pricing to reduce waste and maintain profitability. Cross-docking networks with a zero-inventory approach provide an effective response to this need. The present study aims to design a multi-objective mixed-integer linear programming (MILP) model for a three-level perishable product supply chain network, considering cross-docking operations, freshness-dependent dynamic pricing, and a bulk-purchase discount policy. This research is based on mathematical modeling, and for validation, sample problems of different sizes were randomly generated. The model was solved and analyzed using the exact augmented epsilon-constraint (AUGMECON) method for small-scale problems and the Non-dominated Sorting Genetic Algorithm (NSGA-II) and Multi-Objective Particle Swarm Optimization (MOPSO) metaheuristic algorithms, with parameter tuning using the Taguchi method, for large-scale problems. Their performance was evaluated and ranked using the TOPSIS multi-criteria decision-making method. The findings indicate a significant trade-off between total profit maximization and operational risk minimization (including fleet breakdown and product spoilage risks), evaluated via measurable and interpretable metrics. While MOPSO exhibits superior performance regarding absolute objective values (higher profit, lower risk) and computational efficiency, NSGA-II provides better diversity and a wider Pareto front. Consequently, TOPSIS ranks NSGA-II higher due to its comprehensive solution-space coverage, highlighting the distinct suitability of each algorithm for operational versus strategic decision-making.