Safety stock addresses the unpredictable timing of demand and lead-time random variations. We can’t forecast it, so we must continually carry it as safety stock.
The most obvious data sets we need for determining safety-stock levels – that consistently achieve our service-level targets with optimal inventory performance and minimal expediting – are:
1. Recent historical daily actual time-series demand
2. Recent historical daily actual time-series lead times (or an informed estimate of lead-time ranges)
This data is the source for defining and quantifying the structure and magnitude of random variations affecting safety stock and service level. However, we need more data than this. Some is intuitive, and some – not so much. Fortunately, you can obtain it with relative ease!
3. Target service level – Very intuitive. The percentage of demand quantity to be fulfilled on time. This is the point beyond which we cannot justify more inventory and carrying costs for a higher service level. Target service levels are usually readily known. Unfortunately, popular safety-stock calculations do not correctly represent the quantity-based fill rate we use to track and measure our actual service-level performance. See our blog The Six Factors Affecting Safety Stock – Part 4.
4. Replenishment interval (RI) – The elapsed time between reorders. It is different than lead time. MOQ and/or package size often determine RI, or a calendar-driven reorder-review interval. Intuition says that as we get more frequent replenishments, relative to lead time, we need less safety stock. Less intuitive, but true, is the fact that as RI gets larger than lead-time, we also need less safety stock! When RI gets large enough, we need no safety stock. RI is easily obtained from your MRP/ERP system. However, most safety-stock calculations do not take RI into consideration, while others do so in a simplistic and incorrect way. See our blog The Six Factors Affecting Safety Stock – Part 3.
5. Probability of unfulfilled demand becoming past-due backlog – What happens when you don’t fulfill a customer order on time? The order may become past-due backlog that must be fulfilled. If this is your situation, you know that past-due demand is disruptive, and often causes even more orders to become late. This is obvious, yet popular safety-stock calculations do not consider it. This probability is easily determined from your order-entry system. See our blog The Six Factors Affecting Safety Stock – Part 5.
6. Cost and ease of expediting – Perhaps you have some inventory items that can be expedited easily and inexpensively. More likely, expediting may be cost-prohibitive if not logistically impossible. For this, you need a high degree of confidence that the safety-stock level will consistently achieve the target service level without, or with minimal, expediting. For easily expedited items, though, you don’t need the additional inventory required for a high level of confidence. So far, this is intuitive. What’s not so intuitive is that confidence is not the same as service level or fill rate. Confidence is a separate, additional factor that’s not available in any popular safety-stock calculations. Likely, you don’t have established confidence levels, but you can easily estimate them. See our blog The Six Factors Affecting Safety Stock – Part 6.
What are your thoughts?
Can you gather these six types of data?
We believe that when you do this, you’re well on your way to correct, objective, optimal safety stock!
Let us know what you think.