The Smart Forecaster
Pursuing best practices in demand planning,
forecasting and inventory optimization
An inventory professional who is responsible for 10,000 items has 10,000 things to stress over every day. Double that for someone responsible for 20,000 items.
In the crush of business, routine decisions often take second place to fire-fighting: dealing with supplier hiccups, straightening out paperwork mistakes, recovering from that collision between a truck and the loading dock.
In the meantime, however, your company’s accumulated inventory control policies keep on doing what they do, even if they are leaking money. A good manager will make time to listen to the “background noise” even when he or she hears loud crashing in the warehouse.
Consider the current settings for your inventory control parameters (e.g., reorder points and order quantities). It’s easy to think of these as “fire and forget” decisions. But these settings usually accumulate over time and end up comprising a mish-mash of forgotten judgement calls that may be misaligned with your current operating environment. Many factors can drift away from their previous levels, such as supplier lead times, ordering costs, or average item demand. These changes can force invisible tradeoffs that are not to your best advantage.
It’s wise to revisit these control settings now and then to see if it’s possible to align your day-to-day operations with current realities. Of course, it would be infeasible for a busy manager to manually calculate the effects of changing the control settings on, say, 10,000 items. But that’s what modern inventory optimization and demand planning software is for: making large scale analytical tasks feasible. Such software will allow you to automatically process new information and compute adjustments at scale. The result will be easy wins – many of which would otherwise go unrealized. And continuously saving a little here and there adds up to significant dollars when you are managing thousands of items.
Consider this example. Company A uses a periodic review inventory system. Every 30 days, they check on-hand inventory for all their items and decide how much replenishment stock to order. Each of their 10,000 items has a specified Order-Up-To Level that determines the size of their replenishment orders.
For instance, suppose Item 1234 has an Order-Up-To Level of 74, determined by factoring in the average item demand of 1.0 units per day, an average replenishment lead time of 8 days, and a target fill rate of 90% for this item. The choice of 74 as the Order-Up-To Level lets Company A meet its 90% fill rate target for Item 1234, but it also results in an average on hand inventory level of 40 units. At $1,500 per unit, this item alone represents $45,000 of inventory investment.
Now supposed that average item demand were to drift up from 1.0 to 1.2 units/day. Without anyone noticing, the fill rate for Item 1234 would drop to 82%!
Now suppose demand were to shift in the other direction and drift down to 0.8 units/day. As with the increase in average demand from 1.0 to 1.2 units/day, kind of change is difficult to see when looking at a plot (see Figure 1) but can have a significant operational impact. In this case, the fill rate would zoom to a generous 96% but on hand inventory would also zoom: from 40 units to 46. Those six extra units would represent $9,000 in excess inventory.
Figure 1: Samples of daily demand with two different average values. The difference in demand is unnoticeable to the naked eye but if not accounted for will have a large operational impact on inventory spend and service levels
Now imagine similar small shifts happening unnoticed across a full fleet of 10,000 inventory items. The total financial impact of all such shifts would be sufficient to get onto the radar of any CFO. Trying to keep on top of this turbulence would be impossible if done manually but modern inventory optimization software could calculate the proper adjustments automatically as frequently as your company can handle, even daily helping you realize substantial improvements in service levels, inventory efficiency, while lowering stockout and holding costs!
Just-In-Time (JIT) ensures that a manufacturer produces only the necessary amount, and many companies ignore the risks inherent in reducing inventories. Combined with increased globalization and new risks of supply interruption, stock-outs have abounded. So how can you execute a real-world plan for JIT inventory amidst all this risk and uncertainty? The foundation of your response is your corporate data. Uncertainty has two sources: supply and demand. You need the facts for both.
Let’s start by recognizing that increased revenue is a good thing for you, and that increasing the availability of the spare parts you provide is a good thing for your customers. But let’s also recognize that increasing item availability will not necessarily lead to increased revenue. If you plan incorrectly and end up carrying excess inventory, the net effect may be good for your customers but will definitely be bad for you. There must be some right way to make this a win-win, if only it can be recognized.
If you both make and sell things, you own two inventory problems. Companies that sell things must focus relentlessly on having enough product inventory to meet customer demand. Manufacturers and asset intensive industries such as power generation, public transportation, mining, and refining, have an additional inventory concern: having enough spare parts to keep their machines running.
This technical brief reviews the basics of two probabilistic models of machine breakdown. It also relates machine uptime to the adequacy of spare parts inventory.