Software for inventory optimization is most often used to crank out the analytical results you need to run your day-to-day business, such as Reorder Points (also known as Mins) and Order Quantities. This specialized software helps you find the sweet spot that balances inventory costs against item availability during routine operations.

Inventory optimization software can also be used to perform “what-if” analyses on scenarios that describe changes from your current operating environment. What-if analysis (also called “sensitivity analysis”) lets you elevate your thinking from the tactical to the strategic. It helps you imagine how you should change your operations to adapt to potential changes in your operating environment. These changes might be negative pressures imposed on you from the outside, or they might result from your own positive actions. In this blog, we provide an example of how to conduct “what-analysis” on lead times and order quantities.  Outputs from the analysis can be used by the business to assess the impact of these changes on inventory costs and service level performance.

How Suppliers Limit Your Freedom of Maneuver

Discussing with our customers the data inputs required by inventory optimization software, we noted that suppliers are a prominent influence on their operations. We leave aside for now such important topics as sharing demand forecasts with suppliers and working out responses to supply chain disruptions, such as Hurricane Matthew last year in the southeastern US. Instead, we focus on two more common ways that suppliers influence producers’ inventory costs: replenishment lead times and restrictions on order quantities.

Replenishment lead time is the number of days that elapse between inventory reaching or breaching a reorder point and the appearance of replenishment units in stock. Some portion of lead time is internal to the producer, perhaps due to slow reactions in a purchasing department. The rest of lead time is down to the supplier. In this discussion, we assume that suppliers’ contribution to lead times might be changed, for better or for worse. (But the same results could apply to changes in producers’ contributions to lead times.)

The restrictions on order quantities that we consider are order minima and order multiples. You might want to order 3 units of some item, but the supplier might impose a minimum order size of 6 units, so your 3 unit order would have to become a 6 unit order. Or you might want to order 21 units, handily exceeding the minimum order size of 6 units, but if the supplier also has an order multiple of 6, meaning every order must be a multiple of 6 units, then your 21 unit order would have to be increased to 24 units.

Scenario Analyses

To illustrate the use of inventory optimization software for what-if analysis, we examine two sets of scenarios. In the first set, lead times are varied from -20% to +20% of their values in a baseline scenario. In the second set, results are computed first with no supplier restrictions, then with order minima only, and finally with a combination of order minima and order multiples. We use Smart Inventory Optimization software for the calculations.

The baseline scenario uses real-world data on 2,852 spare parts managed by a progressive public transit agency. These parts have an extremely heterogeneous mix of attributes. Their per unit costs range from $1 to $23,105, and their lead times vary between 1 day and 300 days. Over 24 months, the mean demand ranged from less than 1 unit per month to 1,508 units per month, with coefficients of variation ranging from a manageable 10% to a scary 2,171%. Furthermore, the supplier picture is also very complex, involving 293 unique vendors, supplying an average of about 10 parts each. This heterogeneity implies that a real-world optimization would pick and choose among items and vendors. However, for simplicity of exposition and to develop basic insights, our what-if scenarios in this example treat every item and vendor equally. Similarly, we assumed in the baseline that holding costs equaled 20% of the dollar value of an item and that every replenishment order had a fixed cost of $40.

We conducted two what-if experiments. The first examined the effects of changing lead times. The second examined the effects of introducing restrictions on order quantities. In each experiment, we recorded the effects of the changes on two operational metrics: average number of units in stock and average number of orders per year. In turn, these influenced four financial metrics: average dollar value of inventory, average holding cost, average ordering cost, and the sum of the last two, which is total inventory operating cost.

In all scenarios, reorder points were calculated so as to achieve 95% probability of avoiding stockouts while waiting for replenishment. Order quantities, in the absence of supplier restrictions, were computed as what we call “feasible EOQ”. EOQ is the classic “economic order quantity” taught in Inventory 101; it is computed from average demand, holding cost and ordering cost. Feasible EOQ adds an additional consideration: inventory dynamics. If the reorder point is very low, it is possible for EOQ to be too small to sustain a stable, positive level of inventory. In these cases, feasible EOQ increases the order quantity above the EOQ to insure that average inventory does not go negative.

Effects of Changing Lead Times

Table 1 shows the results of changing the lead times. Working around the base case, we changed every item’s lead time by -20%, -10%, +10% and +20%.

It is no surprise that reducing lead times reduced the required level of inventory and increasing them did the opposite. Both the average number of units and the associated dollar value behaved as expected. What may be surprising is that the effects were somewhat muted, i.e., an X percent change in lead time produced a less-than-X percent response. For instance, a 20% reduction in lead time produced only a 7.9% reduction in on-hand inventory and only a 12.0% reduction in the dollar value of those units. Furthermore, the effects of reductions and increases are asymmetric: a 20% increase in lead time led to just a 7.3% increase in units (vs 7.9%) and only a 9.6% increase in inventory value (vs 12.0%).

Similar attenuated and asymmetric results held for operating costs. A 20% reduction in lead time decreased total operating costs by 7.0%, but a 20% increase in lead time caused only a 5.1% increase in operating costs.

Now consider the implications of these results for practice. In a competitive world, cost reductions on the order of 10% or even 5% are significant. This means that efforts to reduce lead times can have important payoffs. In turn, this means that efforts to streamline purchasing processes may be worth doing. Likewise, there is a case for engaging suppliers about reducing their part of lead time, possibly by sharing the savings to incentive them.

Table 1: Effects of changing lead times

Effect of Order Quantity Restrictions

Table 2 shows the effect of imposing supplier restrictions on order quantities. In the base case, there are no restrictions, i.e., the order minimum is 0 and the order multiple is 1, implying that any order quantity is acceptable to suppliers. Working away from the base case, we first looked at imposing an order minimum of 5 units on all items, then adding an order multiple of 5 for all items.

Forcing orders to be larger than they otherwise would be had the expected impact on the average number of units on hand, increasing it by 0.9% with only an order minimum and by 3.4% with both a minimum and a multiple. The corresponding changes in the dollar value of the inventory were more dramatic: 22.4% and 23.3%. This difference in the size of the percentage response probably traces back to the large number of low-volume/high-cost replacement parts managed by the public transit agency.

Another surprise was the net reduction in operating costs when supplier restrictions were imposed. While holding costs went up by 22.4% and 23.3% in the two what-if scenarios, the larger order quantities allowed for fewer orders per year, resulting in offsetting reductions in ordering costs of, respectively, -24.4% and -32.7%. The net impacts on operating costs were then reductions of 3.7% and 7.9%.

In general, placing restrictions on producer actions would be expected to reduce performance. So the results in these scenarios were counter-intuitive. However, the real message here is that using EOQ, or even enhanced EOQ, to set an order quantity does not give optimum results. Paradoxically, the order quantity restrictions we investigated seem to have forced order quantities closer to optimal levels.

Table 2: Effect of order quantity restrictions

Conclusions

The what-if analyses shown here do not lead to universal conclusions. For instance, changing the assumed cost per order from $40 to some smaller number could show that the supplier restrictions increased rather than decreased the producer’s inventory operating costs.

When doing what-if analysis in real-word situations, users would naturally craft scenarios at a lower level of detail. For instance, they might evaluate the effect of changes in supplier lead times on a supplier-by-supplier basis to find the ones that would have the highest potential payoffs. Or they might arrange for order minima, if they exist already for all items, to change by a specified percentage instead of a fixed amount, which might be somewhat more realistic.

The key takeaway is that inventory optimization software can be used in “what-if mode” to explore strategic issues, beyond its customary use to calculate reorder points, safety stocks, order quantities, and inventory transfers.

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.

To make the right decision here, you have to think systematically about the problem. That requires that you use probabilistic models of the inventory control process.

A Scenario

Let’s consider a specific, realistic scenario. Quite a number of factors have an influence on the results:

• The item: A specific low-volume spare part.
• Demand mean: Averaging 0.1 units per day (so, highly “intermittent”)
• Demand standard deviation: 0.35 units per day (so, highly variable or “overdispersed”).
• Supplier average lead time: 5 days.
• Unit cost: $100.
• Holding cost per year as % of unit cost: 10%.
• Ordering cost per PO cut: $25.
• Stockout consequences: Lost sales (so, a competitive market, no backorders).
• Shortage cost per lost sale: $100.
• Service level target: 85% (so, 15% chance of a stockout in any replenishment cycle).
• Inventory control policy: Periodic-review/Order-up-to (also called at (T,S) policy)

Inventory Control Policy

A word about the inventory control policy. The (T,S) policy is one of several that are common in practice. Though there are other more efficient policies (e.g., they don’t wait for T days to go by before making adjustment to stock), (T,S) is one of the simplest and so it is quite popular. It works this way: Every T days, you check how many units you have in stock, say X units. Then you order S-X units, which appear after the supplier lead time (in this case, 5 days). The T in (T,S) is the “order interval”, the number of days between orders; the S is the “order-up-to level”, the number of units you want to have on hand at the start of each replenishment cycle.

To get the most out of this policy, you must wisely pick values of T and S. Picking wisely means you cannot win by guessing or using simple rule-of-thumb guides like “Keep an average of 3 x average demand on hand.”  Poor choices of T and S hurt both your customers and your bottom line. And sticking too long with choices that were once good can result in poor performance should any of the factors above change significantly, so the values of T and S should be recalculated now and then.

The smart way to pick the right values of T and S is to use probabilistic models encoded in advanced software. Using software is essential when you have to scale up and pick values of T and S that are right for not one item but hundreds or thousands.

Analysis of Scenario

Let’s think about how to make money in this scenario. What’s the upside? If there were no expenses, this item could generate an average of $3,650 per year: 0.1 units/day x 365 days x $100/unit. Subtracted from that will be operating costs, comprised of holding, ordering and shortage costs. Each of those will depend on your choices of T and S.

The software provides specific numbers: Setting T = 321 days and S = 40 units will result in average annual operating costs of $604, giving an expected margin of $3,650 – $604 = $3,046. See Table 1, left column. This use of software is called “predictive analytics” because it translates system design inputs into estimates of a key performance indicator, margin.

Now think about whether you can do better. The service level target in this scenario is 85%, which is a somewhat relaxed standard that is not going to turn any heads. What if you could offer your customers a 99% service level? That sounds like a distinct competitive advantage, but would it reduce your margin? Not if you properly adjust the values of T and S.

Setting T = 216 days and S = 35 units will reduce average annual operating costs to $551 and increase expected margin to $3,650 – $551 = $3,099. See Table 1, right column. Here is the win-win we wanted: higher customer satisfaction and roughly 2% more revenue. This use of the software is called “sensitivity analysis” because it shows how sensitive the margin is to the choice of service level target.

Software can also help you visualize the complex, random dynamics of inventory movements. A by-product of the analysis that populated Table 1 are graphs showing the random paths taken by stock as it decreases over a replenishment cycle. Figure 1 shows a selection of 100 random scenarios for the scenario in which the service level target is 99%. In the figure, only 1 of the 100 scenarios resulted in a stockout, confirming the accuracy of the choice of order-up-to-level.

Summary

Management of spare parts inventories is often done haphazardly using gut instinct, habit, or obsolete rule-of-thumb. Winging it this way is not a reliable and reproducible path to higher margin or higher customer satisfaction. Probability theory, distilled into probability models then encoded in advanced software, is the basis for coherent, efficient guidance about how to manage spare parts based on facts: demand characteristics, lead times, service level targets, costs and the other factors. The scenarios analyzed here illustrate that it is possible to achieve both higher service levels and higher margin. A multitude of scenarios not shown here offer ways to achieve higher service levels but lose margin. Use the software.

Improve Forecast Accuracy, Eliminate Excess Inventory, & Maximize Service Levels

In this video, Dr. Thomas Willemain, co-Founder and SVP Research, talks about improving forecast accuracy by measuring forecast error. We begin by overviewing the various types of error metrics: scale-dependent error, percentage error, relative error, and scale-free error metrics. While some error is inevitable, there are ways to reduce it, and forecast metrics are necessary aids for monitoring and improving forecast accuracy. Then we will explain the special problem of intermittent demand and divide-by-zero problems. Tom concludes by explaining how to assess forecasts of multiple items and how it often makes sense to use weighted averages, weighting items differently by volume or revenue.

Many of our customers that saw demand dry up during the pandemic are now seeing demand return.  Some are seeing a significant demand surge. Other customers in critical industries like plastics, biotech, semiconductors and electronics saw demand surges starting as far back as last April. For suggestions about how to cope with these situations, please read on.

Surging demand usually creates two problems: inability to fill orders and inability to get replenishment due to supplier overload. This situation requires changes in the way you use your advanced planning software. Here are three tips to help you cope.

Tip #1: Narrow your temporal focus

In normal times (remember those?), more data implied better results. Nowadays, old data poison your calculations, since they represent conditions that no longer apply. You should base forecasts and other calculations on data from the current situation. Where to cut off past data may be obvious from a plot of the data, or you may decide to set a “reasonable” cutoff date based on a consensus of colleagues.  Smart Software has developed machine learning algorithms that automatically identify how much historical data should be optimally fed to the forecast model. Be on the lookout for these enhancements to the software that will be rolling out soon. In the meantime, conduct accuracy tests using held-out actuals using different historical start dates.  Smart’s forecast vs. actual feature will support this automatically.

Tip #2: Increase your planning tempo

When operations are stable, you can set your inventory policies and trust them to be appropriate for a long time. When times are turbulent, it is important to increase the frequency of your planning cycles to keep old policy settings from drifting too far away from optimality.  More frequent recalibration of your stocking policies and forecasts means that you’ll be quicker to catch trends that will surprise your competition and always keep you steps ahead.  With software capable of automatically selecting optimal values, all that work can be done in one shot by the software. You should review those changes and possibly tweak them, but it makes sense to let the software do the bulk of the work.