Methodical decision support for mid-term order planning in global production networks for multi-variant series production under uncertainty of customer order configurations

在全球生产网络中利用多变量系列化生产的生产能力是一项挑战，但这是有竞争力的生产和具有成本效益的定制产品供应的先决条件。此外，要求较短交货时间的客户在收到订单之前，在客户需求与最终装配能力和供应商能力之间的平衡方面造成不确定因素。因此，必须明确考虑订单配置的不确定性，以便在中期范围内进行稳健的规划。现有的计划方法要么完全忽略了中期订单计划，要么忽略了订单配置的不确定性，针对这些不足，提出了一种在客户订单配置不确定的情况下支持全球生产网络中期订单计划决策的方法。因此，必须根据客户订单的中期可获得性来区分几个行业的不确定性类型-要么客户订单可用，但其配置不确定，要么不可用。如果客户订单在没有特定订单配置的情况下可用，则在生产网络中分配订单时必须考虑这种不确定性。客户订单的可能配置必须通过场景进行预测和表示。因此，可以根据基于情景的优化模型的结果将客户订单分配到生产设施和生产周期。在客户订单在中期内不可用的情况下，可以通过生成计划订单并应用确定性优化模型将其分配到生产网络来控制不确定性。稍后，可以将传入的客户订单分配给已在生产网络中分配的计划订单。因此，优化模型可以支持关于配置和交货时间的最优计划订单的选择。为了启用分配，所选订单将根据客户订单进行调整。总而言之，中期订单计划允许中期和短期计划展望期保持一致。如果只有短期计划是基于订单的，这一点就无法得到保证。此外，一致的信息基础使公司的销售、生产和采购等职能单位能够考虑到各自的需求而进行综合规划。

The utilization of production capacities in multi-variant series production in global production networks is a challenge, but is a prerequisite for the competitive production and cost-efficient supply of customized products. Additionally, customers requiring short delivery times cause uncertainties prior to the receipt of orders regarding the balancing of customer demand with capacities of final assembly and capacities of suppliers. Thus, the uncertainty of order configurations has to be explicitly considered for robust planning in the mid-term horizon. Current planning approaches neglect either mid-term order planning at all or the consideration of uncertainties of order configurations.In order to address these shortcomings, the research proposal at hand aims at the development of a methodology to support decisions regarding mid-term order planning in global production networks for multi-variant series production under uncertainty of customer order configurations. Therefore, the type of uncertainty has to be differentiated for several industries depending on the mid-term availability of customer orders - either they are available, but their configurations are uncertain, or they are unavailable.In the case customer orders are available without certain order configurations, this uncertainty has to be considered when assigning orders in the production network. Possible configurations of customer orders have to anticipated and represented by scenarios. Thus, the customer orders can be assigned to production facilities and periods based on the results of a scenario-based optimization model.In the case of customer orders being unavailable in the mid-term horizon, uncertainty can be controlled by generating planned orders and assigning them in the production network by applying a deterministic optimization model. Later on, incoming customer orders can be assigned to the planned orders already assigned in the production network. Therefore, an optimization model can support the selection of the optimal planned order regarding configuration and delivery time. In order to enable an assignment, the selected order is to be adapted to the customer order.All in all, mid-term order planning allows for consistency between mid-term and short-term planning horizons. This cannot be guaranteed if only short-term planning is order-based. Moreover, a consistent basis of information enables integrated planning of functional units of a company such as sales, production and procurement considering their respective requirements.