This paper presents a two-stage supply chain problem at the operational level involving a manufacturer and a third-party logistics provider (3PL provider). The integrated problem consists of two subproblems: An m-machine permutation flow shop with inventory considerations for the manufacturer and a routing problem for the 3PL provider. Both agents face tardiness penalty costs in case of late delivery to the customer locations. Assuming imperfect information sharing between the two agents and independence in their own decisions, we investigate the scenario in which the manufacturer dominates the negotiation.

In this scenario, the manufacturer imposes the number and composition of vehicles as well as the vehicle departure dates on the 3PL provider. However, lacking the information regarding the actual delivery times he is forced to estimate delivery times at the customer locations in the context of his planning. Following the planning of the manufacturer, the 3PL provider decides on the optimal routing of the vehicles. Both agents aim to minimise their total costs. After formulating the problems as a mixed integer linear program, two metaheuristic algorithms are proposed to solve the scenario. The performance of the heuristics is evaluated on a set of randomly generated instances. The results show that the genetic algorithm is the best performing method.

本文提出了一个涉及制造商和第三方物流提供商（3PL提供商）的两级供应链问题。集成问题包括两个子问题：一个m机排列流水车间，制造商要考虑库存问题，而3PL提供者要解决路由问题。两家代理商都面临着迟到罚款的费用，以防延迟交付给客户。假设两个代理商之间的信息共享不完善，并且他们自己的决策具有独立性，我们将研究制造商主导谈判的情形。

在这种情况下，制造商会将车辆的数量和组成以及车辆的出发日期强加给3PL提供者。但是，由于缺乏有关实际交货时间的信息，他被迫根据其计划来估计客户位置的交货时间。根据制造商的计划，3PL提供者决定车辆的最佳路线。两家代理商均致力于将总成本降至最低。将问题表述为混合整数线性程序后，提出了两种元启发式算法来解决该问题。在一组随机生成的实例上评估启发式算法的性能。结果表明，遗传算法是性能最好的方法。