A distribution model with two tobacco transfer stations is used to improve the tobacco distribution with the conventional cigarette transportation, and the maximal covering model is used to locate the transfer stations, then the transportation optimal model with a trapezoidal fuzzy parameter is proposed. Combined with the constraints of the suppliers' supply capacity and capability needs of demand points as fuzzy parameters, a new method of linear programming based on the concept of constraint satisfactory degree is proposed. In order to solve the problem of the transport planning model with fuzzy parameters, the interval-valued mean is used to transform the uncertain model into an interval number linear programming model, which is finally transformed into a deterministic linear programming model. With this method, the decision maker can obtain a corresponding satisfactory solution according to different conditions, so the decision process becomes more flexible and more adaptable to real situations. By adopting this method, the transportation costs of tobacco companies in this city are greatly reduced, saving 5.37% of the direct distribution model cost.