Abstract: To optimize distribution plans of oil products for multiple distribution schemes with lateral transshipment, the models which can express multiple distribution schemes are developed with the objective of minimizing the total distribution cost. Besides, a heuristic algorithm is proposed for getting distribution plans of multiple distribution schemes, which can produce high quality solutions rapidly. The effectiveness of the proposed models and algorithm is verified by numerical results. The managerial insights of this paper are that co-distribution can never be the lowest-cost distribution scheme; when vehicles are not fully loaded, the optimization space of mixed distribution is larger; the cost advantage of mixed distribution is more significant when the transshipped depot is closer to the distribution district; and the time difference for getting distribution plans in different schemes by the proposed algorithm is small. When making distribution plans, the separate distribution plan and the mixed distribution plan can be generated at the same time, and then select the best one. This paper provides a scientific approach for choosing a better distribution scheme and optimizing distribution plans under the circumstance of lateral transshipment, which is favorable to reduce the distribution cost for distribution companies of oil products.

Key words: oil products distribution, co-distribution, separate distribution, mixed distribution, C-W savings algorithm

摘要： 针对多种横向转运配送方式下的成品油配送方案优化问题，以配送成本最小化为目标，构建了可同时表达多种配送方式的优化模型。设计了能够同时求解多种配送方式下的配送方案，且求解精度高、速度快的启发式算法。数据实验验证了模型、算法的有效性，并得出：合并配送方式不可能成为成本最低的配送方式；当车辆不满载时，混合配送方式的优化空间更大；随着转运油库与配送区域的趋近，混合配送方式的成本优势越来越明显；所提算法求解不同配送方式下的配送方案的时间差极小，在制定配送计划时，可同时生成分开配送方案和混合配送方案，择优选用。本研究为横向转运背景下成品油配送方式的选择和配送方案的优化提供了一种科学的决策方法，有利于降低企业的配送成本。

关键词: 成品油配送, 合并配送, 分开配送, 混合配送, C-W节约算法