A supply chain network optimization study can answer many questions that supply chain leaders wish to understand. Most commonly, how many manufacturing plants and distribution centers should we have in our network? Where should each of these facilities be located? What should the service territory of each facility be? These are some of the key questions companies typically ask when engaging in strategic network design.
When deciding on a network design, however, there are two levels of analysis that can help to answer the questions at hand, namely a SKU level or shipment level analysis. The primary difference between the two is the granularity with which data is collected, analyzed and eventually modeled. There are several other differences between these two types of analysis and other factors to consider when deciding on which level of analysis to select when evaluating and optimizing your network.
In a shipment level analysis, all inbound and outbound data is collected at the shipment level. There is no SKU detail or product master files collected. The amount of time required for data collection, analysis, and validation for this level of analysis is considerably less compared to a SKU level analysis. The network model is simpler and the effort required is less than for a SKU level analysis. However, not all questions can be answered at this level of analysis. For example, it is not possible to differentiate between full line and partial DCs. Also, models are not able to take into account direct plant shipment opportunities or any product specific scenarios. Shipment level analysis should be used in simple supply chain networks with sourcing that is generally all from a single location or region (i.e. Asia). Shipment level analysis will answer broad questions and give a directional answer in a relatively short amount of time.
In a SKU level analysis, all data is collected at the item or SKU level. Outbound transaction data is collected at the item and customer level. Inbound data is collected by vendor/plant and item. Therefore, significant time is required to collect, process, cleanse, and validate this type of data. Also, the modeling is more complex as SKUs need to be aggregated into product groups, product/customer specific flows are modeled, and sourcing rules need to be considered as vendors can produce all or a subset of SKUs. Finally, modeling growth is typically more involved as growth rates can be product or product line specific. However, due to the level of detail, the analysis and modeling can accommodate most questions asked of the supply chain practitioner.
When deciding on the type of analysis to perform, it is important to consider “what questions you want/need” or “what questions should be” answered by the analysis. These questions will drive the analysis and help determine the level of granularity required. This importance is confounded by the fact that if a shipment level analysis has begun, starting a SKU level analysis would require more effort than if the initial analysis had just been at the SKU level in the first place. Certain questions, like the following, can only be answered by a SKU level analysis.
Although a shipment level study would require less effort, it is evident that these and similar questions can only be answered by a SKU level analysis. What supply chain questions are driving YOUR network analysis?
—Brad Barry, St. Onge Company
We have been named to Inc. Magazine’s annual Best Workplaces list! Featured in the May/June 2023 issue, the list is the result of a comprehensive measurement of American companies that have excelled in creating exceptional workplaces and company culture, whether operating in a physical or a virtual facility.
From thousands of entries, we are one of only 591 companies honored.
Click here to see our listing!