Implementing a Model with a Rolling Horizon AIMMS Language Reference

In this section you will find two strategies for implementing a rollinghorizon. One is a simple strategy that will only work with certainrestrictions. It requires just a single aggregation step and a singledisaggregation step.

Model and metaheuristics for a scheduling problem integrating procurement, sale and distribution decisions

A STDSM approach strives to keep the promised delivery dates and to perform manufacturing at the lowest possible cost. Order repromising is required due to high uncertainty and the resulting changes in supply and available capacity. The STDSM function is similar to batch promising, however, all already promised orders compete for the supply and the capacity, while only the orders arriving within the batch interval are considered in batch order promising.

The decision rule approach to optimization under uncertainty: methodology and applications

The need for rolling horizon approaches for assessing demand fulfillment is conceptually discussed by Chen et al. (2008). A STDSM approach is proposed by Geier (2014) for a computer manufacturer. It is integrated with order promising in a rolling horizon setting, while feedback from the shop floor is considered. The STDSM approach proposed in the present paper is different since we compute the supply for BE facilities based on FE production planning. Moreover, we use an iterative approach that extends the delivery time windows of the orders. Seitz and Grunow (2017) propose an order promising approach that exploits product and process flexibility typical for semiconductor supply chains.

• This is indicated by the surrounding frame for the BE facilities (Step 3) in Fig.
• These objects are updated in an event-driven manner using notification functions of the commercial simulation software AutoSched AP 9.3.
• An allocation planning procedure for an assemble-to-order (ATO) supply chain is proposed by Chen and Dong (2014).
• With the facilitiesintroduced in the previous sections setting up such a model isrelatively easy.
• However, these improvements which are a result of the different cost settings are obtained at the expense of reduced stability.
• This means that production control is related to lots that are already released on the shop floor.

3.4 Impact of the cost setting for the STDSM scheme

Finally, the fabrication position (FPOS) aggregate is used to represent the FE level in the supply picture offered by master planning. In this paper, we differentiate between orders that are fulfilled by FP, DREP, and FPOS product aggregates. The structure of the considered semiconductor supply chains including the different product aggregates is shown in Fig.

3.2 Impact of demand settings

ATP reallocation approaches are responsible for releasing unused committed ATP quotas. All already promised but unfinished orders are considered within a STDSM approach (Fleischmann and Meyr 2004) taking into account the available supply and capacity. STDSM approaches are desirable in semiconductor supply chains due to the long cycle times and the process and demand uncertainty (Mönch et al. 2018b). A single simulation run leads to 728 planning epochs of the STDSM approach. The corresponding average computing time for a single simulation run of the SSC-S scenario is 3581 min whereas the corresponding time for the RBR heuristic is only 1388 min. Both the STDSM and the RBR approach require allocation planning, i.e. solving instances of the model (A1)–(A6).

The algorithm to implement the rolling horizon can be outlined asfollows. The BE facilities are much smaller with respect to the number of work centers rolling horizon approach and number of process steps in the routes (Mönch et al. 2013). Therefore, solving a simultaneous BE STDSM MILP instance for all BE facilities is possible. This is indicated by the surrounding frame for the BE facilities (Step 3) in Fig.

The probed wafers are stored in die banks (DBs) that serve as decoupling points between FE and BE. Distribution centers (DCs) are responsible for decoupling BE facilities and customers. Each FE and BE facility consists of machine groups which contain machines that provide the same functionality. We refer to machine groups as work centers in the rest of this paper. We start by describing different product aggregates, i.e. a grouping of products based on certain criteria, to characterize the supply.

• Two values are provided for each factor level and performance measure.
• Cloud manufacturing is a promising direction for semiconductor supply chains (Wu et al. 2014; Chen 2014; Yang et al. 2020; Herding and Mönch 2022).
• When the FE facility of a product is known, but the BE facility and the DC are not yet determined, the product is represented by a DB representative (DREP) in the supply picture provided by master planning.
• However, details are not provided for all these systems that provide demand fulfillment functionality.
• All already promised but unfinished orders are considered within a STDSM approach (Fleischmann and Meyr 2004) taking into account the available supply and capacity.
• The simulation continues after a plan is computed by the corresponding planning approach.
• The cost settings for the STDSM approach are summarized in Table 4.

The upper value is obtained by the RBR whereas the lower value is computed by the STDSM. Best values for each pair of performance measure values are marked bold. The FE and BE STDSM MILP instances can be solved individually for each single FE and BE facility since supply is provided by master planning for each single facility. This is indicated by individual boxes for the different FE facilities (indicated by FE1, …, FEm) in Step 1 and Step 4.

We refer to STDSM when an order-based matching takes place on a short-term level. In this paper, a STDSM approach for semiconductor supply chains was proposed. The approach is based on a decomposition that takes into account the structure of the semiconductor supply chain. The NP-hardness of the related planning problem was proven. The integration of the STDSM approach into a hierarchical planning approach that includes master planning, allocation planning, and production planning was discussed. Note that the STDSM approach is based on decomposition according to the physical structure of the underlying supply chain, i.e., optimization models are solved for the different nodes of the supply chain or groups of them.

Modeling and heuristics for production time crashing in supply chain network design

This small-sized semiconductor supply chain model is abbreviated by SSC-S. The infrastructure including the simulation model is shown in Fig. The first two planning epochs of the master planning and production planning function are indicated in the figure by vertical lines that are blue colored.