为了使即时配送企业能以较低的成本提高配送准时性，从而维护并发展高价值客户，首先，针对即时配送客户的特点改进RFM模型，基于已有数据使用DBSCAN算法进行客户聚类，根据聚类结果使用GBDT算法构建客户分级预测模型对即时配送客户进行分级预测。在此基础上，以即时配送的固定成本、变动成本及客户超时点种类、数量为优化目标，构建基于客户分级优先的即时配送路径优化模型，再设计遗传算法对该模型进行求解。最后，以沈阳市某一站式冷链即时配送企业为对象进行实例分析。结果显示，相比该企业原配送方案，应用客户分级优先的即时配送路径规划方法规划后的方案在配送总成本仅提高4.8%的情况下，高价值、潜在高价值客户超时点数量由6减少为2，且超时点均为边缘客户，同时配送总时间减少了7.3%，验证了该方法的有效性。采用该配送路径规划方法，企业的配送成本虽然会小幅增加，但因配送准时性提升，可以更好地维护高价值客户，同时发展潜在高价值客户向高价值客户转变，进而保持或提高长期收益。

In order to enable real-time delivery enterprises to improve delivery punctuality at lower costs, meanwhile maintain and develop high-value customers, firstly, the RFM model was improved based on the characteristics of real-time delivery customers, and customer clustering based on existing data was realized using the DBSCAN algorithm. Using the GBDT algorithm, a customer grading prediction model based on the clustering results was constructed to predict the grading of customers. On this basis, with the fixed and variable costs of instant delivery, as well as the types and quantities of customer timeout points as optimization objectives, a real-time delivery path optimization model based on customer classification was constructed, and a genetic algorithm was designed to solve the model. Finally, a case study was conducted on a one-stop cold chain real-time delivery enterprise in Shenyang. The results showed that compared to the original delivery plan of the enterprise, the plan planned using real-time delivery path planning method with the customer classification priority only increased the total delivery cost by 4.8%. Meanwhile the number of high-value and potential high-value customer timeout points decreased from 6 to 2, and the timeout points were all edge customers. At the same time, the total delivery time decreased by 7.3%, verifying the effectiveness of the planning method. By adopting this delivery path planning method, although the delivery cost of the enterprise may slightly increase, the improvement of delivery punctuality can better maintain high-value customers, while developing potential high-value customers and transforming them into high-value customers, thereby maintaining or improving long-term profits.