This is the first article in a series on determining safety-stock for demand variation.
Analyzing Demand Variation for Kanban
The goal of demand variation analysis is to determine the target safety-stock for every kanban item necessary to prevent a demand-related stockout while limiting on-hand inventory. After you have quantified demand variation, you must determine whether to define safety-stock for each unique item, or for categories of items with similar demand variation patterns.
Setting safety-stock for each discrete item is the most precise approach. The process involves calculating actual demand variation for each inventory item and using each item’s calculation to determine its safety-stock.
Sometimes, operational needs, demand this level of precision. Unfortunately, maintaining lean, reliable safety-stock values for each individual item can be quite unwieldy, even when only managing a moderate list of inventory items.
Furthermore, the added effort and “precision” doesn’t always result in managing the safety-stock of unique items, differently. Thus, this is not the common or preferred method. However, in those rare cases where demand complexity dictates this approach, the option exists and can work.
The preferred approach of setting safety-stock is by group. Just as when setting safety-stock by item, the process starts by calculating actual demand variation for every inventory item. Next, items are segmented into groups with similar demand variation patterns. Finally, each item’s safety-stock is determined by its associated group, rather than by each item’s unique standard deviation. Many new managers intuitively find this approach wasteful for calculating each item’s demand variation, only to ignore it when setting each item’s safety-stock. In reality, this is the far more efficient and easier to maintain approach.
2 Safety-Stock Setting Prerequisites: Regardless of which approach you take, there are 2 vital perquisites that must be completed before setting safety-stock.
- Seasonality Assessment: Before attempting to set safety-stock for any item, you should always assess its seasonality, first. If an item experiences significant seasonally driven demand variation, then its safety-stock should be set by season. Otherwise, your operation will be at elevated risk of experiencing a demand-related stock-out or overstock situation.
- Demand Variation Quantification: After seasonality has been assessed, each item’s demand variation must be quantified in terms of the standard deviation of lead-time periods.
Only once both of these prerequisites have been completed, should you proceed with one of the safety-stock setting methods below.
Safety-Stock by Group Method
After you have quantified demand variation, you must experiment with your data to identify those items with comparable demand variation patterns for defining safety-stock groups. Once you’ve defined groups and safety-stock values, you will be able to make a table that your inventory kanban calculator can reference to automatically assign safety-stock values to individual inventory items.
Below are the steps for assigning items to safety-stock groups:
- Analyze Seasonality and Demand Variation
As mentioned, above, complete the 2 prerequisites; seasonality assessment and demand variation quantification. Be sure that demand variation is expressed as days of demand.
- Brainstorm Possible Primary Groups
Depending on your specific organization and business situation some categories may make more sense than others in influencing how safety-stock should be set. These groups should be relevant to the business and mutually exclusive. Thus, it should be impossible for an item to belong to more than one group. For example, a class-C item cannot also be a class-A item. A manufactured item can’t also be a purchased item.A few grouping possibilities are as follows:
- Lead time
- Product family
- Value Stream
- Manufactured or purchased
- ABC inventory class
- Brainstorm Sub-Groups
Next, you want to list possible sub-groups. To do this consider which of your primary groups may represent a large degree of demand variation within them. Ask yourself, which factors might influence this variation? For example, items grouped by product family may represent multiple demand-variation scenarios. A major influence on demand variation might be the specific market an item is sold to. Thus, it may make sense to define market as a sub-group.