In this paper a model for the vehicle routing problem with stochastic components of demand, service time and delivery time for perishable products is presented. The purpose of the paper is to present an alternative to the problem of collecting the flower industry, using a model that reflects the stochastic behavior in collecting flowers, required by companies in Colombia. The model incorporates three stochastic components and a restriction not reported earlier by other authors. The solution method includes clusters for collection points, route design, allocation to truck routes, Monte Carlo simulation and a regression model to obtain the equation of the total system cost and optimal point of replenishment.