In this paper, we address a real life the delivery of fresh agricultural products vehicle routing problem with a heterogeneous fleet of vehicles and a time window constraints (HVRPTW). Two objectives of minimization the total costs comprised of fixed costs, variable costs, waiting costs, and penalty costs, and maximization the customer services that can be rendered to customers in terms of acceptable delivery time (coverage) are considered simultaneously. By taking the demand on fresh agricultural products combined uncertainty of randomness and fuzziness as a random fuzzy variable, we propose a random fuzzy multi-objective dependent-chance programming (DCP) model for the problem. A hybrid intelligent algorithm integrating random fuzzy simulation and genetic algorithm (GA) is designed to solve the model. The results indicated that waiting cost and penalty cost can significantly influence total delivery costs and the schedule of customers service. It was found that our proposed models provide a effective tool for decision-maker to make effective delivery decisions under random fuzzy environment.