中文摘要：

以现金物流为研究背景，提出了一种基于在途风险的多类型现金的押运路线优化问题，以新币配送均衡、旧币回收和在途风险减少为优化目标，建立了相应的混合整数规划模型，并设计了一种混合禁忌搜索算法进行求解，其中禁忌搜索算法用以确定路线决策，嵌入的精确算法、贪心算法和混合贪心算法用以确定新币配送决策、旧币回收决策和风险决策。数值实验对问题特性和算法性能进行了分析，结果表明：①新币券别均衡优化和旧币回收导致在途风险增加；②混合禁忌搜索算法能求解更大规模的问题，并得到较好的解，嵌入算法很好地平衡了运行时间和求解质量。

英文摘要：

A multi-type cash routing optimization problem with transit risk was proposed in the context of cash logistics industry. The problem was modeled as a mixed integer programming problem to balance new currency denominations distribution, to pick up used currency and to decrease transit risk. A combined tabu search method was proposed to solve this problem. This method consists of a tabu search algorithm to determine routing decision and three embedded methods named exact method, greedy method and the mixed greedy method to determine the denomination balance decision, the used currency decision and the risk decision. The numerical studies were adopted to analyze problem characteristics and the method performance. The results show that new currency denomination balance optimization and used currency pickup link with an increased risk of transportation. The combined tabu search method can solve greater problems and produce high quality solutions. The three embedded methods can well balance the computing time and the solution quality.