In this post, Bradford continues our discussion of Multi-Echelon Inventory.
As we discussed earlier, you likely don’t need the “optimization” in Multi-Echelon Inventory Optimization (MEIO). But, you do need to consider multiple echelons and you need good inventory science. We call this Multi-Echelon Inventory Calculation (MEIC).
Watson Says... |
 |
| While having a Multi-Echelon Inventory Calculation process is very important getting it implemented in a complicated business is not trivial. |
 |
What do you say? |
 |
| Click here to send us your comments |
|
|
|
What is MEIC? MEIC allows you to simultaneously set the right safety stock levels across each echelon in the supply chain. It considers the interactions across the echelons—it propagates demand and forecast error throughout your supply chain, it properly accounts for risk pooling, and it handles your finished goods and raw material inventory at the same time.
The following sections explain how you should use MEIC at your organization to get the most out of it. We explain this through an analogy to flying. The destination represents the desirable inventory and service levels. And, the pilot is the inventory planner.
Use MEIC to stay oriented:
Many of us have felt the disorienting effects of flying through the clouds, where it can actually become very difficult to tell exactly which way is up. In this situation a pilot needs to avoid the natural instinct to try to point the plane in the direction they think is up (which may actually be down), and to instead trust their instruments and training to navigate the plane to its destination. Similarly, senses can be deceiving when setting safety stocks, because it can be very complex. In order to handle this complexity, we need an instrument that calculates a scientific safety stock for us.
What are “scientific” safety stock calculations? We often use this term when describing MEIC because MEIC takes the guess work out of setting safety stocks and replaces it with a calculation that is based on actual measurements from the supply chain. Careful measurements are taken of things like replenishment lead times and their variability, demand levels, forecast accuracy, minimum and incremental order quantities, review and reorder frequencies, and the reliability of production yields, just to name a few. Each of these factors has important implications for how much safety stock a business should carry in order to meet service targets at a minimum cost.
This is in contrast to a large variety of simpler safety stock calculations, such as specifying a certain number of days of supply to hold. Such approaches don’t account for most of the factors listed above, but they are attractive because they are often simple and intuitive. However, as intuition is often wrong when flying in clouds, it’s also often wrong when it comes to safety stock. If demand increases, for example, it seems intuitive to increase safety stock, but in many cases an increase in demand actually warrants a decrease in safety stock. And the effect that other factors have on safety stocks can be even more unintuitive. MEIC gives us a scientific instrument to guide our clients to the right safety stock for each product.
Don’t Panic- Use Your Instrument Panel:
Despite having instruments in the plane, many pilots fail to use them when they are most needed. This is usually because of panic. Panic makes us take irrational action in order to avoid some horrible consequence. For planners, pressure to avoid stock-outs can also make it difficult to plan rationally. In order to avoid being blamed for poor service, a natural thing to do is “pull up!” and order more inventory. On a larger scale, this a common reason that companies struggle with excessive inventory levels. The best way to prevent panic is through proper training, and one huge benefit of implementing MEIC is a more informed supply planning team. The training is realized in a couple of ways.
First, it’s informative to actually measure supply chain attributes that were thought to be known. As an example, we start every inventory project by asking clients for certain data, including historical purchase order records. We do this so that we can measure how much time it takes from the time an order is placed until an order is received from a supplier. This often appears to be a waste of time, since there is usually some table that says what this value is supposed to be. However, when we look at the actual order times they rarely match whatever times are being used for planning (Figure 1).
Second, seeing how changes in the supply chain typically relate to changes in inventory levels allows planners to further develop an understanding of their supply chain. When inventory starts to go up and we need to know why, we need only to look at what supply chain metrics are driving that change.
Use MEIC as an auto-pilot:
When using MEIC, it should be as part of an ongoing process. This is the plane’s auto-pilot, keeping the plane pointed in the right direction and at the right altitude. Every week, month, or quarter, the input factors should be re-measured to account for changes in a supply chain, whether seasonal or non-seasonal, and new safety stock numbers should be published, reviewed by planners. By establishing this process we create a supply chain that is responsive to change. As a business grows or suppliers change, the safety stocks recommended by MEIC will adjust as well.
This is an important point. If you do not have a good MEIC process in place, improvements in your demand forecasts or reductions in lead-times may not show up in terms of reduced inventory.
In addition, turning on the auto-pilot allows the planner to put less time and attention on the hundreds or thousands of products that are not problematic and to put more time into figuring out how to make meaningful improvements to their portion of the supply chain.
Use MEIC For What-Ifs Before You Take Action:
The other big benefit of having a MEIC platform in place is that is allows you to test different ideas before you implement. You can ask questions about the benefits of reducing supplier lead-times, ordering or making product more frequently, or improving forecast error. You can then use this to help prioritize where you put your energy.
Final Thoughts
While having a Multi-Echelon Inventory Calculation process is very important getting it implemented in a complicated business is not trivial. In the next post, we’ll discuss how we have found success implementing this process at numerous firms.
Bradford is a data scientist. He uses his diverse background in optimization and statistical modeling to bring creative solutions to difficult problems. In addition to bachelor’s degrees in mathematics and economics from Indiana University, he completed a Master’s Degree in industrial and systems engineering at the University of Wisconsin in Madison.
Any reaction to this Expert Insight column? Send below.
Your Comments/Feedback
|
