Your Privacy Matters

Parts of this website use cookies and content from third-party providers. More details can be read in our privacy policy, from where you can also amend your preferences at any time. Please select from the options below if you are OK with this.

DeclineAccept

Skip to main content

We ended last year’s benchmark report with the question, would 2023 be the year that the seemingly inexorable upward rise in big pharma inventory levels abates? The answer, emphatically, is no.

Once one-off effects (1) are taken out, in DIO terms median inventory grew a further 6.1% year on year, with mean inventory growing 5.1%. Total inventory grew from $160bn to $175bn, in a year where revenues and cost of sales were down 4%.

Overall inventory write offs, as far as can be estimated (2), held steady at about 4% of cost of sales, although there were significant write offs of Covid-19-related vaccines and treatments, most notably Pfizer’s eye-watering $6.2bn write off of Paxlovid and Comirnaty.

Days Inventory Outstanding (DIO) = inventory value/(cost of sales/365). See also the technical notes at the end of this article.

Overall, despite a good crop of new medicines coming to market, financially 2023 wasn’t a great year for the industry. There were notable blockbuster successes, such as MSD’s Keytruda, and growth in demand for weight-loss treatments carried on apace, although with greater competition as more medicines enter the market, but the overall lacklustre sales saw many companies forced to cut costs, with layoffs across the sector. The full impact of the Inflation Reduction Act on medicine pricing in the US still remains to be seen but is also driving an increased focus on cost.

We refresh the list of companies included in the benchmark from time to time, so you can’t compare the numbers from one year’s report to another, but in the seven years we have been producing this report, inventories expressed as DIO have increased every year but 2, with DIO remaining roughly flat one year and showing a small decrease the other. The other five years have shown increases. Looking at it another way, in 2017, median DIO was around 180 days, equivalent to approximately 6 months of demand. In 2023 median for those companies who were in the benchmark in 2017 was over 220 days, comfortably more than 7 months in demand.

In short, the industry’s inventory binge does not appear to be over. As we write every year, there are good reasons why pharmaceutical companies carry a lot of inventory (security of supply for products that are critical to patients, long lead times, regulatory restrictions) as well as bad ones (high gross margins, a lack of focus on inventory optimization).

One reason for the increase in inventories in 2023 was the playing out of a classic bullwhip, as concerns of shortages throughout the value chain in previous years had led many to increase inventories, leading to overstocking. Producers of API’s noted a slow-down in sales, as many pharmaceutical manufacturers had swollen their raw material supplies in recent years to ensure supply. For the companies in our benchmark, on average raw materials inventories are almost double what they were before the pandemic. Meanwhile the pharmaceutical manufacturers themselves reported overstocking at their customers as a reason for suppressed demand.

Of course, given the importance and utility of their products, no one would want the pharmaceutical manufacturers to run short of essential medicines, and this consideration features very highly in their approach to inventory. If the current levels of excess and obsolescence are simply a necessary evil to achieve such high availability, surely it is a price worth paying?

However, while a certain amount of waste is necessary to maintain very high service levels for products with very short shelf lives, most medicines have a shelf life sufficiently long for this not to be a significant factor on its own. When a product has a shelf life that can be measured in months, this is certainly the case.

Another way of looking at it is from a purely financial perspective. There are a variety of companies in the benchmark: generics manufacturers, research-driven pharmaceutical manufacturers, diversified groups, contract manufacturers, but the average gross margin in 2023 was 65%. The gross margin is so large because R&D accounts for such a large proportion of the true cost of doing business for a lot of the firms, but EBITDA is also very healthy in the sector. In this context, scrapping 4% of your inventory each year equates to little more than 1% of revenue. This is a very small consideration compared to the rewards for, say, successfully bringing a new blockbuster to market.

Inventory is not a cost to be minimized. It is a strategic asset to be optimized. Medicine shortages continue to be a global problem (3). As we highlighted in last year’s report, contrary to what intuition might tell you, shortages and overstocks are not inversely linked. Yes, if you only have one product then having too much of it means you can’t simultaneously not have enough. But once you consider the huge number of different medicines and dosage forms, it is obviously possible to have too much of some and not enough of others and this is the case in practice. Manufacturing capacity, raw materials and cash are all finite resources and overproduction in one area can and does cause shortages in others.

So when we hear people say “we don’t want to reduce inventories because we don’t want to risk shortages”, this is the wrong way of thinking about it. You should reduce unneeded inventories so that shortages can be better avoided. This is not to say that pharma supply chain teams are not already exerting themselves to better achieve this balance – of course they are, this is the very essence of supply chain. But as we look at another year’s inexorable rise in big pharma inventory levels, we have to conclude that there is still much room for improvement.

If you have any questions about this report, or would like help improving your organisation’s inventory levels (whether you feature in the benchmark or not), please contact us.

Technical Notes

1. “One-off effects” include the change of companies in the benchmark as well as major acquisitions or divestitures. When we compare DIO numbers from one year to the next, we exclude companies with this type of one-off effect so that they do not distort the underlying trend. Of course, a lot of these companies are making acquisitions or divestitures on a fairly regular basis, so we only exclude those where a significant impact on inventory levels is observable in either of the two years compared. The major one-off effects in 2023 are as follows:

• Bausch Health recorded a fair value step up on $162m of XIIDRA inventory acquired from Novartis. This should be amortized over 2 years, of which $20 was amortized through Cost of Sales in 2023.
• Biogen recorded a fair value step up on $1.3bn of inventory acquired through the purchase of Reata. This should be amortized over 3 years, with 31.5m amortized in 2023. Biogen have finished writing off Aduhelm.
• Pfizer recorded a fair value step up on $4.1bn of inventory acquired through the purchase of Seagen , $672m of inventory acquired through the purchase of GBT and $817m of inventory acquired through the purchase of Biohaven. Pfizer wrote off $6.2bn of unmarketable Paxlovid and Comirnaty (which also affected the DIO numbers by swelling Cost of Sales). In addition, Pfizer lost $209m of inventory from the tornado which hit its Rocky Mount facility.
• Amgen recorded a fair value step up on $5.025bn of inventory acquired through the purchase of Horizon Therapeutics.
• AstraZeneca recorded a $114m impact on Cost of Sales from the unwinding of fair value from the acquisition of Alexion.

At the time of publishing, Boehringer Ingelheim had not released their Finance Report for 2023, so we have replicated the 2022 figures as a placeholder and will update the benchmark once it becomes available. Boehringer Ingelheim do not report Cost of Sales so we have used an estimate as with previous years.

2. For estimates of inventory write offs, see our article “Inventory write offs in pharmaceutical manufacturing”.

3. See, for instance, the European Association of Hospital Pharmacists report, this summary from C&EN on shortages in the US, or this research briefing from the UK parliament. Of course, there are multiple reasons why medicines run short, but a lack of free capacity to react quickly to shortages is one.

Inventory optimization is a much-used phrase, but what exactly is it, how do you achieve it and do you in fact even need it? As our mission as a company is to help organisations get value from inventory optimization, this is a topic very close to our heart.

In this article we consider inventory optimization to be the application of mathematical techniques to define ideal inventory levels. Such theoretical models are not easy to apply in practice. The idea that a piece of software can allow you to optimize your inventories simply and automatically is attractive but at present far from possible.

We believe that the biggest value to be had from inventory optimization techniques is in showing you how much better your inventories could be, while the correct interpretation and application of that information is likely to require changes to how you manage your inventory.

The benefits of overcoming these challenges are substantial. Inventory optimization allows you to free up working capital, reduce shortages and obsolescence, reduce your environmental footprint and increase flow through your supply chain, usually by significant amounts.

You will often see any improvement to inventory management described as inventory optimization, but this is not strictly speaking accurate.

In mathematics, optimization means selecting the best option from a set of alternatives. When talking about inventory, this comes down to determining the best amount of inventory to hold to meet your objectives. Inventory optimization mathematically seeks to find the sweet spot between having too much and not enough.

This is a simple concept to grasp in principle, but not to apply in practise. In order to calculate an optimum level mathematically, you need to factor in a number of different considerations, some of which are themselves difficult to quantify.

Let’s say that you want to optimize on purely economic principles – what will make you the most profit? To do this you need to be able to quantify, on the one hand, the true holding cost of inventory and, on the other hand, the cost of shortages, which might include backorder costs, expediting costs, the cost of lost sales and lost market share, even the cost of damage to your brand. Noting that these costs are entwined with each other. It’s not that this is impossible to do, but it is laborious and somewhat imprecise.

In fact, it would be fair to say that few people, if any, actually try to calculate this. Instead, normal practice is to set a target service level, either implicitly or explicitly, and then try to minimize inventories while staying at or above that service level. This is a good, pragmatic move, that ties down one of the more difficult parameters to set. However, even with a pre-defined target service level, there is still a lot of work to do to calculate optimum inventory levels.

When it comes to improving inventories, we like to differentiate between optimizing, which involves working with the givens, and improvements based on changing the givens. What are the givens? They are the parameters you need as inputs to your inventory optimization calculations. We have already looked at target service level as a given, but this is just one of a range of such parameters.

When you work with the givens, you don’t challenge them, you accept them as fixed parameters and then try to optimize your inventory levels on that basis. Which is not to say that the givens can’t be improved, but you strictly speaking need to have such parameters defined to be able even to attempt an optimization calculation.

How can you optimize your inventories?

Inventory optimization comes down to defining two things for each item: the optimal order quantity/batch size (Q) and the optimal safety stock (SS), noting that the latter might be zero or even negative. But there is no one way to do this for all items you stock, since the method chosen will itself vary depending on a number of factors. Let us start by taking a couple of simple examples.

1. You have a product that is made to order. Whenever a customer orders one, you have 10 weeks to make it. It takes one week to make and 4 weeks to source the parts. Everything you need is at or above your minimum order quantities (MOQ’s) and you have no inbound supply variability or capacity constraints.
2. You have a product that is made to stock. Demand is relatively stable, with a limited amount of variability, as is supply. You check inventory levels of finished goods continuously; you schedule production to avoid stock outs and you try to maintain raw materials inventory to be able to produce when needed.

Note that if anything in bold changes, the change will need to be factored into your optimization calculation, as will many of the other factors shown in Figure 1 above.

Different types of inventory require not just different optimization calculations, but also different planning methodologies.

For the first example above, we can consider a deterministic planning model. This is appropriate when you know with a high degree of certainty what demand is going to be, or at least that it is going to be significantly different to the past. In this instance, your order and batch sizes can be set to exactly what is required and in principle you shouldn’t need safety stock, although you might need to have some safety stock to protect against variation in supply, such as late deliveries or quality problems in production.

For the second example, we can consider a replenishment model. This is appropriate for make-to-stock finished goods and buy-to-stock raw materials where demand is not too intermittent but displays variability. In this case, the EOQ model or one of its variants should be used to calculate optimal order/batch size. Safety stock can be calculated using the target service level, the standard deviations of demand and lead time, and the lead time itself.

However, this replenishment model, which is based on probably the most widely known text-book set of equations, has important sensitivities. The most commonly found version of the safety stock equation, for instance, assumes that demand is normally distributed, that average demand does not vary, and it also breaks down when the variability is very high. You will sometimes find people saying that the equations “don’t work”, but this is often just their understanding of the equations running into some of these barriers which they either don’t understand or at any rate don’t know how to work around.

Note that different models might work best at different points in your supply chain. For instance, you might, depending on your exact circumstances, choose a replenishment model for finished goods and a deterministic model for raw materials. To really optimize your inventories, you are likely to need to employ different strategies for different types of inventory.

And these two examples are far from representing the only types of stock you might carry. For instance, you may have make-to-order items where the customer promise time is shorter than the total supply lead time. You may have spare parts with highly sporadic demand but which need to be available on demand. And so on. Each different type of inventory needs an appropriate optimization approach.

In order to fully optimize your inventories, all of the factors above need to be taken into consideration for every single item that you stock. And so far, we’re just talking about mathematical optimization. In addition – and no less a challenge – you have to consider what processes, systems and people are in place and how they would need to change in order to deliver and maintain the optimization.

In our experience, the biggest gap is often in the people themselves and how they are used to working. The knowledge of optimization techniques and how to take advantage of them is a rare commodity amongst material planners, whereas data scientists do not necessarily have the requisite understanding of the practicalities of the business to enable practical application of their models. Technology is sometimes seen as a way around this – remove people from the equation as far as possible – but we are a long way from this being a satisfactory solution yet.

Do you need inventory optimization?

From the brief summary above, it should be evident that inventory optimization is not simple. In fact, it would be closer to the truth to call it impossible than to call it simple, but this is not an all or nothing situation. Given the complexities outlined, as well as the fact that your optimal inventory levels are constantly changing, it should be admitted from the start that fully optimized inventory is not an achievable goal. But the application of optimization techniques is incredibly valuable. You don’t have to win an Olympic gold medal to get value from learning how to swim. Moreover, many of the biggest benefits from inventory optimization come from the first steps.

In the absence of inventory optimization, inventory management is flying blind. You are searching for the sweet spot between too little and too much inventory without knowing where that sweet spot is. Think of it as trying to find a black mouse in a completely dark room. With optimization techniques, you are switching the light on. It may still be difficult to catch the mouse, but at least you can get closer to it because you can see where it is.

It is of course possible to manage your inventories without trying to optimize them, and especially in times of abundant cheap cash this is what many organisations choose to do. But without the analytics of inventory optimization you have no real way of knowing how well you’re doing. Or rather, your view becomes lopsided. Your service level performance will tell you if you have enough, but what is telling you if you have too much? Top-down aggregate measures like cover, inventory turns or DIO give you ratios that will tell you whether you’re getting leaner or not over time, and can be used for quick benchmarking exercises, but strictly speaking tell you nothing about how much you ideally should have.

Where organisations do not use optimization techniques, there is a strong tendency to apply “one size fits all” approaches to different types of inventory: for instance, setting target safety stock levels for all raw materials at 2 weeks’ average demand. This could only possibly be optimal if all of your raw materials had identical variability of demand and supply as well as identical lead times. This one size fits all approach means there is a strong correlation between shortages and inventory levels long before you reach an optimized level. Because inventory optimization by necessity works at an individual item level, it allows you in aggregate to reduce inventories and shortages simultaneously.

From our own empirical experience of working with multiple organisations, it is normally the case that inventory optimization alone (so taking the givens as given) allows net inventory reductions of 20-50% while maintaining or improving service levels. This improvement takes time and effort to achieve, but double-digit improvements are normally achievable in the first year. We can think of no other option that will deliver such a large improvement so quickly in such a risk-free fashion.

How do you take advantage of inventory optimization?

The good news is that not all of the technical problems need to be solved at once and that some of the biggest benefits come from the simplest steps.

To take advantage of inventory optimization you need two things. Firstly, the ability to run the optimization calculations. Secondly, and no less importantly, the ability to operationalize the findings of your analytics. You will only sustainably improve inventory levels if your organisation can use the insights generated by the analytics in its day to day work.

The biggest value in inventory optimization analytics comes from the visibility it gives you into where the biggest opportunities for improvement are. It highlights things that are not necessarily obvious or intuitive. In the first instance, a calculation using a number of approximations is normally good enough. Once this is well understood, you can go further and seek to remove or refine the input approximations one at a time, noting that you will never remove all of them entirely.

Because the analytical challenge is high, specialist skills are required to do it properly. But those skills are not necessarily the same ones required to make changes to how your business actually works. Too much emphasis on analytics, or a siloed mentality, can leave a large gap between theory and practice. The analytics itself is only one part of the picture. In addition, you need to build the people, processes and systems to take advantage of it.

Inventory optimization is not something you can implement and then forget about. Rather, it is an ongoing process which needs constant attention. All of the input parameters to your optimization calculations are in a state of almost constant change and the right process controls need to be in place to stay on top of your inventories, especially the closer you get to optimal levels.

If done well, inventory optimization can transform how well your supply chain functions, but this requires a clear understanding of what it can do and how to use it. Year one of a well-designed programme can easily deliver double digit improvements from some of the easiest steps, while those who want to get much closer to optimization can deliver that type of improvement for many years in succession.

What impact would a 25% reduction in inventory have on your business? Well:

• It would make a nice difference to your balance sheet. It would reduce working capital and overall capital employed. It would reduce the amount written off for obsolescence each year
• It would make a nice difference to your income statement. Typically inventory holding costs (including finance costs) can add up to 25% of the value of inventory held each year
• You would have some cash available for more interesting uses than sitting in a warehouse: capital investment, acquisitions or equity buy-back for instance
• And it would make a nice difference to the ratios your investors care about. Both Return on capital employed and return on equity will improve, as will Earnings per share (EPS), especially if you choose to use the liberated cash for a share buyback

Suggesting an improvement in working capital to most CFO’s is a case of preaching to the choir. To illustrate the impact of a 25% reduction in inventory on key financials we have included an example in an appendix to this paper. For a €2.5bn turnover company with €400m in inventory such as the one we take in our example it improves ROCE by 2%, EPS by 3% and ROE by over 4%. These percentages will heavily depend on your circumstances; the point is that the difference is enough to care about.

The case for holding inventory

There cannot be too many CFO’s who wouldn’t sign up for the scale of improvement above if they believed it was achievable. But is less inventory automatically a good thing? Well not if it is at the expense of customer service or sales. The whole purpose of inventory after all is to provide a buffer between demand and your ability to supply. If profit margins are high, and raw materials markets unpredictable, the temptation to hold plenty of inventory can be strong. But our contention is that most firms have significantly more than they need, whilst still managing to run short of some items.

A hundred years ago, inventory was considered a good thing. The attitude was “pile them high, sell them cheap”. Assets were a source of pride. But increasingly, leanness is considered a virtue. In times of seemingly constant innovation and change, the risk of obsolescence is high and almost all products can seem ephemeral. Increases in on-demand production and the ease of finding alternatives via online services is only intensifying this trend. Online retailers can actually achieve negative working capital. Whilst this isn’t the case in most other industries, we do see that the ability to be agile and to hold inventories much closer to ideal levels is increasingly going to be a source of competitive advantage.

To get a very approximate sense of how close to the mark you are, look at your Days Inventory Outstanding (DIO). This measure takes your total inventory value and divides it by cost of sales days (total cost of sales/365). The number you get will not tell you very much per se – different businesses will have significantly different inventory needs – but comparison with your peers should be informative, and the figure will give you the ability to sense check it yourself. If your DIO is equivalent to 80 days, for instance, as in the simplified example in the appendix below, that means on average you are holding over 2.6 months of stock. Does it take you that long to source, produce and ship your products, even allowing some spare for demand fluctuation? It might do, if you have long lead times, a lengthy production process and high demand variability, but these are the types of question to ask your Operations teams. Bear in mind as well that with an average of 2.6 months of stock, you will have significantly more of some items. Reducing DIO to 60 days, again following our example, still gives you 2 months’ inventory on average.

This is one of the headaches for a finance executive. Supply chains are different. Depending on your production strategy (make to order vs make to stock), the seasonality and variability of your demand, the lead times in your supply chains and the economics of your target service levels, the “right” amount of inventory for you can vary enormously. Appropriate policies will also vary depending on your industry and the nature of your supply chain.

Plus there are of course other levers to improve working capital, which may well seem much simpler to pull. Reducing accounts receivable is mostly about internal process discipline. Even increasing accounts payable, whilst often labour intensive, is usually just a case of lengthening payment terms with your suppliers. And increasingly, supply chain financing options exist to help improve this balance depending on your own particular working capital needs. By comparison, inventory can seem a very complex undertaking, and one which we know many firms have tried and failed to address in the past.

Between having an intuition that you might be holding excess inventory and actually being able to see exactly what you should have, let alone do something about it, there is quite a gap. Indeed, we find that between firms who have tried and failed to seriously get a handle on inventory and those who have consistently put off doing something about it, there are very few who have tight control of inventories and a strong understanding of the different levers which contribute to it. But why is this so?

In our experience, at the heart of the problem lies data. A reasonably large firm might easily have thousands of different items between finished goods, work in progress and raw materials. Each transaction should be captured in your ERP systems, and stocks should be reconciled with physical stocks at least once per year. With this data, you have everything you need to analyse your inventory and identify optimal levels BUT the data is hard to extract, manipulate and then calculate, whilst inventory optimization software, whether built into your ERP or integrated with it, is highly sensitive to relatively minor input adjustments which are frequently poorly understood. As a result, even mathematically sound planning tools fail to do the job of optimizing inventory and don’t really give you any insight into how you are performing overall and what can be done to improve the situation.

From a lack of clear and accessible data stems a number of other challenges: inability to correctly identify the root causes of sub-optimal stock levels, and a politicisation of inventory. In the absence of good data there are many competing voices with different agendas and no obvious objective mechanism to decide between them. Other KPI’s, which should be balanced through inventory management, are given priority: Supply Chain focus on On Time In Full (OTIF) deliveries, whilst Sales provide inflated forecasts because they are measured on total sales not forecast accuracy. Unions like inventory as it reduces pressure on workers. All of these factors will tend towards excessive stock being accumulated.

Cutting through the data complexity

How nice it would be to have the following picture available for every item you stock, as well as a roll-up of overall optimization potential:

A = actual historical stock on hand. B = average of A. C = target stock on hand. D = average of C. E = excessive stock, or difference between actuals and target.

This graph shows a familiar basic supply chain concept. The blue “sawtooth” line (C) represents what you should have had on stock. The green line floating above it (A) represents what you actually had. Fairly simple to plot a course of action from this picture.

Of course, the simplified model upon which this diagram is based needs to be enhanced with probabilistic considerations to account for all kinds of variability. You also need to understand demand patterns, which is a science in itself. This raises the bar somewhat on the mathematical front. It also demands segmentation along ABCXYZ lines to identify appropriate policies for different types of stock. But with this type of granularity you have a perfect springboard to drive root cause analysis item by item.

We meet many executives who understand the principles at play here, but precious few who have access to the ability to extract large volumes of data (many millions of transactions) and perform the appropriate mathematical analysis on them. We have personal experience of consultants being brought in to do the heavy lifting. A team can be buried away for 2 to 3 months crunching the numbers, which drives cost and is highly prone to human error. Or high-level assumptions are made to avoid going into this level of detail, but then the business case is less robust and the roadmap of what to attack is less clear and usually focuses on macro-processes such as planning and forecasting. Consultant-led projects normally do succeed in taking out inventory in the short term – if nothing else the executive focus their presence guarantees will deliver this – but it is less clear that this benefit is sustainable, and this engenders reluctance to try this route again.

We also find that inventory optimization software has a tendency not to deliver on its promise. Whilst the better of these tools are built on similar mathematical principles to the more basic of our own tools, they are mostly focused on deriving target inventory levels and they are highly sensitive to inaccuracies in input data such as lead times or demand forecasts. We find many instances where companies are simply overwriting the targets generated by the software since this is an easier fix than understanding and addressing the underlying issues. It is easy to spend a lot of money on inventory optimization software without getting much value from it.

What we have done at nVentic is to automate the data analytics required to truly understand what’s going wrong in your inventory. What others will take weeks if not months to do, we can do in a matter of hours or days. But we are not selling software. We find that a rather high level of expertise is needed to understand how inventory optimization software really works, and we find it more beneficial to our customers to provide them with the outputs of the analysis rather than the tools themselves. Plus our customers don’t need to train people or buy a licence for an analysis they only need to carry out once or twice a year.

In addition, we firmly believe that the only way firms will develop sustainably optimized inventories is by building capability in their internal teams. Our preferred way of working is therefore to partner with a client team to deliver change. nVentic provides the analytics and the expertise, and we help guide your team through a process of understanding the root causes and addressing them over a period of months and in some cases years. For if the truth be told, we know that many companies out there, particularly in the manufacturing sector, have the potential to reduce inventories by much more than 25% whilst actually improving service levels. Getting there is likely to take more than one year though – your teams need to learn to walk before they can run, and whilst you most likely already have a number of strong inventory people, they need to be able to bring the rest of the organization with them. 25% is typically a reasonable target for year 1 however.

Is it time for you to get your inventories in shape?

For more information about nVentic, please see our website www.nventic.com or contact us for an initial discussion: information@nventic.com

Appendix

Illustration of the impact of a reduction in inventory on key financials*

(All of the figures in this very simplified example are set out below and on the next page.)

Let’s say your business turns over €2.5bn each year, with cost of sales of €1.85bn and net income of €250m. On the balance sheet you have €400m inventory, €600m in receivables and €700m in cash. In addition, you have non-current assets of €1.2bn, giving total assets of €2.9bn. On the liabilities side you have €900m in current liabilities and €800m in non-current liabilities. With €1.7bn liabilities you therefore have €1.2bn in equity.

Your capital employed is €2bn. Using this highly simplified model, you therefore have ROCE of 12.5% and ROE of 20.8%.

Let’s say you reduce your inventory from €400m to €300m, a 25% reduction. Since studies have shown the holding costs of inventory to be approximately 25% of the value of inventory each year, you will at least in theory increase net income by €25m. (In fact, the P&L impact will almost certainly be less than this, since holding costs include both variable and fixed costs, and absorption can be a challenge, especially in the first year of a programme, but for simplicity here we will use the full 25%. See also our article on holding costs.) Your ROCE will thus improve to 13.8% and your ROE to 22.9%. If you just transform the reduced inventory into cash you will not have affected your net working capital overall, although your acid-test ratio will have improved. You will also most likely need to write down less inventory at the end of the year. Maybe an improvement from €20m to €15m written down. But with the example we have taken, liquidity ratios are not a concern and cash reserves are generous. More likely that with a return on equity of 20%+ there is a strong case for further investment or a share buyback to boost earnings per share (EPS). If price per share is €20 and there were 150m shares on average to begin with, 5m shares could be bought back and earnings per share increase from 1.833 to 1.896, an improvement of nearly 3.5%.

We can summarize the numbers as follows (assuming the share buyback):

* The “Before” figures for this illustration are all taken (approximately) from a real company, the one major change we made being to reduce their actual starting inventory from ~€700m to €400m.

Inventory science is a field well represented in the academic world. From at least Harris (1913) (1) onwards, work has been done to explore how to calculate optimum inventory levels: more than a hundred years, therefore, of research.

And yet if you look at what is going on in organizations today from a practical perspective, you will find surprisingly little of that research being applied. In this white paper we’re going to explore why this is, what impact the failure to apply inventory optimization principles has, and finally what we’re doing at nVentic to help bridge the gap. (2)

Mind the gap

So why is inventory science so popular with academics but under-developed in practice? In a word, complexity. Academics love a good challenge. And there is sufficient complexity in inventory to make it an attractive discipline.

Outside of academia, however, complexity is usually not so welcome. If something is difficult or onerous to do, and one can get by without doing it, there is a strong temptation to leave it alone. Let us have a look at why inventory science is so complex.

The image below is familiar to most students of supply chain:

On the basis of a few key parameters, we can determine how much should be ordered, and when, to maintain optimum inventory levels. We can also determine, based on a few other parameters, how much safety stock to keep.

This is a simple model, and not too difficult to follow, but in our experience, there are few organizations using it to any great extent, if at all. There are a number of barriers to using it in practice:

1. Some of the parameters are hard to quantify. To build the model above for any SKU we need to be able to quantify re-order costs, holding costs and shortage costs (the forth cost parameter, unit cost, is usually easier to quantify)
2. We need to layer variability into the model. Demand is not uniform or constant, and it may not even demonstrate normal distribution. The same may well be true of lead time. This demands some more advanced mathematics
3. Everything in this model needs to be done SKU by SKU. Organizations often have thousands of SKU’s, so the data burden alone is often a roadblock
4. The mathematical model is too precise. The underlying equations will deliver results in fractions, but in reality, organizations need to deal with minimum order quantities, stock controls that aren’t instant, and so on

We will come onto what nVentic does to counter these problems below, but first let’s consider what impact a failure to apply inventory optimization principles will have.

Pile them high, sell them cheap

When Harris was writing, in the early 20th century, inventory was considered a good thing. But over the last 50 years in particular, leanness has come to be considered a virtue. In a world where customers expect greater choice, and constant product innovation, and are only a click away from finding it online, holding too much inventory can be very risky. Holding inventory costs money and ties up working capital.

Of course, you can improve your inventory situation without applying optimization techniques. Measures such as shortening lead times and using consignment stock will make it easier to operate with less inventory on your books. But you still have to make decisions in terms of how much stock to hold, how much to order and how often. If organizations are not using optimization models, how do they do this?

If an optimization model is not used at all, the alternative is often a simplified concept of how many days’ or weeks’ stock you need to hold. Some ready reckoning will factor in average demand, demand peaks, perishability if appropriate, lead times and convenient (for logistics, production or purchasing) lot quantities. Although even doing this requires quite a lot of manual effort and data manipulation which isn’t always easy.

Accordingly, some organizations help themselves by using planning software which will leverage at least some of the mathematics we describe above. However, as we have already noted, these models are very sensitive to the input data, so using tools can be prone to error. We have seen many instances of software being overwritten or ignored because early experience with it led to undesirable outcomes.

Our view is that most organizations are not doing much in terms of inventory optimization. The 2017 REL Europe Working Capital Survey suggests €350bn is tied up in excess inventory in the biggest 1000 companies in Europe alone. And what is harder to quantify because not publicly reported, but no doubt an even bigger concern to many organizations, are the sales lost due to shortages.

That £350bn of tied up working capital is a wasted opportunity, since it could be put to more interesting uses, such as acquisitions, paying down debt, or equity buy-backs. Moreover, as the holding cost of inventory is typically in the region of 20%, that’s also €70bn of unnecessary cost. And that’s before factoring in the impact of lost sales. If companies haven’t addressed this sooner it is surely not only because of the difficulty, but also because most of their competitors haven’t either. Yet it represents a major, largely untapped, opportunity to gain competitive advantage.

Build a bridge and get over it

At nVentic, our mission is to close the gap between the academic theory and practical application of inventory science. So what do we do to enable this?

Firstly, we have built analytical tools which will do all of the calculations for you automatically based on historical data, so the opportunity for manual error is removed. Secondly, we work with clients to actually optimize their inventories. We believe that the tools in isolation have limited value but by working collaboratively with our clients we help them not just to realize the benefits identified, but also to build internal capability for sustainable improvement.

We come to inventory optimization with a strong command of the mathematics, which we have embedded in our own analytical tools, but moreover with many years of helping organizations make concrete improvements in inventory management.

In terms of dealing with the 4 barriers to applying the mathematical model in practice, we apply the following approach:

1. (Parameters which are hard to quantify.) Here the key is approximation. Sensitivity analysis shows that modest changes in these parameters will not make significant differences to the outcome. We can advise clients on what to use here based on experience
2. (Factoring in variability.) This is vital and we use advanced algorithms to deal with it. However, there are also practical ways to simplify it if necessary in the short term by looking at maximum values
3. (Dealing with large data sets.) We have developed tools which take raw data from your ERP system(s), and then automatically transform them and perform over 100 mathematical routines on them. The output is an overall quantification of potential, and the segmentation of your SKU’s by value/volume and variability (ABCXYZ analysis). In addition, our tools allow you to drill down by SKU: historical stock evolution, key parameters and deltas (how much stock did you have compared to how much you optimally should have had?) We can perform this analysis in a matter of days even on very large data sets
4. (The mathematical model is too precise.) We deal with this in two ways. Firstly, we automatically round values up depending on known constraints such as packing sizes or minimum order quantities. Secondly, we recommend an incremental approach to making improvements. Rather than jumping straight to the calculated target, aim to build in a little spare buffer in the first instance and work towards optimal levels incrementally, focusing on the most valuable SKU’s

We find that even taking a conservative approach, significant double-digit improvements are possible for most organizations in the first year, with plenty more to come in subsequent years. The transparency that our tools provide typically enables a certain number of quick wins – identifying excess stock holdings which can safely be drained. And then we work with clients to interpret the data and implement the underlying changes which are necessary to embed optimized inventory management. In this way they build internal capability to make year on year improvements with an increasing command of the subject.

Bridging the gap between theory and practice in inventory management is something we believe in passionately. To truly optimize inventory takes time and effort, but significant improvements are possible quickly and, thanks to the technology we have developed, easily. If you would like to build a bridge in your organization, reach out to us for a conversation today.

Notes

1. Ford W. Harris, “How Many Parts to Make at Once.” Factory, The Magazine of Management 10 (2), February 1913, 135-136, 152. And “How Much Stock to Keep on Hand.” Factory, The Magazine of Management 10 (3), March 1913, 240-241, 281-284.
2. Another topic that we’re not considering here is why Finance departments tend to abandon the topic of inventory management to Operations given how strategic working capital is. This in itself leads to inventory being neglected as a strategic asset, with more emphasis on the tactical topic of inventory control. See also Linda G. Sprague and Marc J. Sardy, “Inventory Management: Some Surprising news about classical views on inventory and some non-classical responses to traditional practice”, Inventory Management – non-classical views, ed. M.Y. Jaber, 2009, Chapter 2.

The opportunity to improve working capital is large across the economy. Macro studies from consultancies like The Hackett Group and PwC calculate the opportunity in the trillions of Euros. (1) There are many interesting points to note in these reports, of which we will highlight three:

1. There is a wide spread in performance between companies. Hackett finds that upper quartile companies convert cash 7 times faster than the median
2. Sustainability is a challenge, with very few companies demonstrating consistent improvements year on year. Both studies highlight this
3. The only working capital lever consistently applied over recent years is Days Payable Outstanding (DPO). PwC finds that companies have increased pressure on their suppliers rather than managing their own assets better. But new payment terms legislation in several countries is limiting how much more can be achieved through DPO

All of this resonates with what we at nVentic observe as inventory optimization specialists. The opportunity to improve working capital through inventory optimization is great, but few organizations have achieved sustainably high performance in this area.

Why should this be so? Hackett highlights what it believes to be a reason why inventory has been neglected as a lever:

“Inventories can often be a… complex area to drive working capital optimization due to competing cross-functional objectives (cash/cost/service) and the ability to easily identify core improvement drivers from within. […] Generally, no one function can drive change without the participation and collaboration of the others, and the process re-engineering and optimization requirements needed to release cash internally can sometimes be politically overwhelming if the exact causes of excess inventories have not been determined.” (Hackett, 2018 US Working Capital Survey)

Inventory can indeed be a complex area to improve, and not just due to the cross-functional challenges involved. Inventory optimization even as a paper exercise requires a strong understanding of the key levers, some of which are complex in themselves and most of which are in a state of flux: the data challenge alone can be daunting. And even once you’ve determined the best course of action you still have to take the rest of the organization with you.

Our mission at nVentic is to help companies cut a path through the complexity. We have developed advanced diagnostic tools which automatically identify optimum inventory levels based on real data. But equally importantly we know how to handle multiple variables simultaneously and produce a clear and pragmatic way forward. For the truth of the matter is that the vast majority of companies can deliver significant double-digit improvements in their inventory position even before they address cross-functional issues.

Let us dig into why you (probably) haven’t optimized your inventory:

1. Do you know what good looks like?

How many organizations really know much inventory they should have? While it is practically impossible, given all of the variables in play, to say how much inventory an organization should have precisely, and recognizing that this theoretical number changes all the time, there is still enormous benefit in calculating what you need as accurately as possible. In the absence of a statistically derived target, organizations may rely purely on top-down or operationally-derived targets. Each has its weaknesses.

Top-down, an organization may simply aim to increase its inventory turns each year. All very well, but how do you set a target? For instance, your inventory might currently be turning over 12 times each year. What is your target for next year going to be? 13 times? 14 times? The point is, without calculating an optimized inventory policy bottom-up, item by item, you can’t know what is possible or desirable. It could be that 12 is already optimal and that by targeting 13 or 14 you are simply going to increase shortages. Or, much more likely, it could be that optimal is more like 20, in which case a target of 13 or 14 is undershooting.

You could derive a target by benchmarking your peers, and this is generally not a bad starting place, but even here care must be taken since optimal inventory levels are very heavily influenced by variables such as lead times, capacity utilization and target service levels, which may well differ greatly even between peers. And how do you know if even your best performing peer is that close to optimal?

At an operational, planning level, organizations are most likely already deriving their own targets as they try to ensure they have enough to meet expected demand. Because this is done on an ongoing basis the implied targets may never be formalised as such. This type of target is the province of Operations and rarely gets visibility at a board level. Or put another way, the target is used for planning and not for performance management. It is used to drive immediate action (order more, produce less, etc.). It is not a goal to work systematically towards.

Related to this question of what good looks like is what you measure:

2. What KPI’s do you use to assess inventory performance?

KPI’s drive behaviours, so the KPI’s you use to measure inventory performance will have an important impact on your overall performance. We can divide inventory KPI’s into two broad groups:

1. Measures of whether you have enough inventory
2. Measures of whether you have too much inventory

You can judge whether you have enough inventory by measuring how successfully you meet customer demand. The vast majority of organizations use one or more measures of service level, whether it be fill rate, cycle service level, shortages, back orders, lost sales, ready rate, out of stock time, or similar.

Having enough inventory is usually a higher priority for most organizations than avoiding having too much. We find that most organizations lack KPI’s to avoid overstocking, unless it is a high level one like inventory turns. And this is already a driver of underperformance. If you don’t have some kind of balance in your KPI set – so that your teams are trying to keep inventories within a range, rather than just one side of a minimum – you will almost inevitably carry too much inventory across the board. The occasional, or not so occasional, shortage only heightens the perception that shortages are the real risk.

Knowing what good looks like is a prerequisite to having a good set of balanced inventory KPI’s.

Also important is how those KPI’s are used – who is measured by them, who sees them and how targets are set and balanced. If different parts of your organization (say Sales and Operations) are using KPI’s in direct conflict with each other without a robust Sales, Inventory & Operations Planning (SI&OP) process in place, it can quickly turn into a turf war rather than a collaborative effort to optimize overall business outcomes. Remember that KPI’s drive behaviours!

3. Who has inventory targets as part of their personal goals?

This flows from the KPI discussion. Mostly likely your inventory planners do. But how about your general managers? A planner can only do so much without the alignment of the rest of the organization. Of course, you don’t want everyone bogged down with the minutiae of inventory metrics, but how about charging 20% of the inventory value (a reasonable approximation of holding costs) back to the P&L of each general manager? See what difference that makes to inventory levels!

4. Do you consider inventory optimization to be risky?

If you hear people talk about reducing your inventory by 25% do you immediately think of shortages?

We have encountered few executives who wouldn’t welcome reduced working capital in principle. But if organizations consistently fail to take full advantage of this opportunity, one of the main reasons is surely the perceived risk. The whole purpose of inventory after all is to enable you to supply your customers with as much of your product as economically desirable. Especially when the cost of capital is low and profitability is strong, why risk running short when you can just carry a little more inventory? Yet inventory optimization and inventory reduction are not synonymous. Good inventory optimization initiatives simultaneously reduce inventory levels and shortages.

This counter-intuitive phenomenon is due to the individuality of every single item you hold in inventory. Each one has its own unique profile in terms of demand, demand variability, lead time variability, perishability, normality of distribution, and so on and so forth. What we invariably find when we analyse organizational data sets is that while some (and usually most) items are overstocked, a smaller number are understocked. Inventory optimization addresses both of these issues. This is why you can simultaneously reduce your overall inventory levels while increasing inventories of some items and therefore reducing shortages. This is one of the few occasions in life where you really can have your cake and eat it!

5. Are you spending money on inventory optimization tools but are still unsure what good looks like or what value you have really got from them?

We are not opposed to the many inventory optimization tools on the market. On the contrary, we believe there are several good ones out there. But here is the rub. Are they being used, and are they being used to their full potential? Plus do the people using them understand their limitations and sensitivities? We come across many organizations with licenses to advanced inventory optimization tools that they barely use, if at all. In many cases trials were made with the tools but problems were encountered so they are switched off and old and trusted manual methods are reinstated. The possible causes of confusion are numerous, but to list a few common examples:

• Do your tools require a single lead time, and do you know what to do if your lead time varies?
• Are your tools dependent on you feeding them a forecast and are you using unconstrainted sales forecasts which invariably fail to materialize?
• Are your tools reliant on data in your ERP system? Do you know which data and how accurate and up-to-date they are?
• Do your tools offer you alternative settings that you find hard to choose between because you don’t really understand what each means or what difference it will make?
• Do your tools produce values that you know from all practical experience to be incorrect, but you can’t tell why?
• Do your tools assume normal distribution of demand? If so, do you know how to test for normality and what to do with other types of statistical distribution?

In short, inventory optimization tools and statistical models more generally can be immensely valuable, but only if used correctly. Used incorrectly they can be positively detrimental. Because the expertise to get the most out of these tools is rare, they tend not to deliver on their promise. This is (often) not so much a fault in the tools themselves as in how they are used.

6. Do you manage to reduce inventories from time to time when you have a big push on working capital, but find the levels always seem to rise again?

It is true of most things in life that given attention they improve, but when neglected they soon revert to a lower level. Yet there seems to be something about inventory which makes it particularly prone to this phenomenon. Why does sustainability seem to be such a challenge? We believe there are two main reasons.

Firstly, inventory reduction can be quick and simple to deliver. You just need to buy and/or produce less! Thus, when companies have a sudden need to free up working capital a strong top-down mandate is usually sufficient. But done crudely, this creates problems, since shortages usually follow. After the initial squeeze, therefore, inventories return to their previous levels as operations revert to their normal way of functioning. Nothing really changed other than a temporary prioritisation.

Secondly, thorough approaches require a certain level of expertise and a concerted effort. When there is a big push on inventories, consultants may be brought in and additional internal efforts also allocated. But this increase in resource is temporary and this in itself creates problems with sustainability, while any deep statistical analysis conducted has the disadvantage of being static.

Rather like inventory optimization tools, one-off projects are not a silver bullet. To deliver sustainable improvements in inventory you have to invest in internal capabilities, frequently refresh analysis and maintain top-down pressure.

7. Do you find inventory a difficult topic? Would you rather focus on less complex things?

It makes sense to do easy things that will make a positive difference before turning to more difficult ones. The business case for optimizing inventories is normally extremely strong – a reduction in working capital and holding costs, combined with an improvement in service level, all delivered at a very strong return on investment. So the fact that most organizations have not done more inventory optimization comes down, above all, to this one consideration: that it is difficult.

While not underestimating the human or political difficulties involved, we believe that what differentiates inventory optimization from many other business change initiatives is the difficulty working out how much inventory you really need, which needs to be calculated on a regular basis. But without this essential first step, done really thoroughly, all of your other efforts will be sub-optimal. You won’t be able to set the right targets, using the right KPI’s, allocated to the right people. You will be at risk of running short and potentially spending money on initiatives without being able to tell how successful they really are.

Yet we believe this difficulty is also one of the reasons why inventory optimization is a great topic to address. Precisely because inventory optimization is a frequently neglected opportunity, significant double-digit improvements are possible in most cases.

To return to one of the Hackett metrics with which we opened: top quartile companies convert cash 7 times faster than median companies. On how many metrics do you have this kind of advantage over your competitors? Surely this is worth overcoming some difficulties to achieve?!

Conclusions

There are many reasons why organizations fail to maintain optimal inventories. We have explored 7 of the most common here, of which we consider the ability to know what good looks like to be the single most important. Based on this insight, nVentic has developed diagnostic tools which quickly and accurately show clients how much inventory they should hold. We then work with them to embed sustainable inventory capabilities in their teams.

Until the macro reports start showing a marked and sustained decrease in inventories, our work will not be complete.

If you would like to discuss how to define what good looks like for your organization and take your inventory optimization capabilities to the next level, contact us: information@nventic.com

Notes

1. For the most recent editions of these reports at the time of publication, see The Hackett Group, 2018 Europe Working Capital Survey and 2018 US Working Capital Survey, and PwC, Navigating Uncertainty, PwC’s annual global Working Capital Study 2018/19. Each of these documents looks at a different selection of companies and each finds a potential of at least 1 trillion Euros or Dollars for the selection they look at.

How much does inventory really cost to hold?

Inventory holding cost may not seem like a strategic lever. In fact, it is one of the key ways in which Finance departments can help improve working capital.

In 2019, a commonly encountered belief is that holding cost is low because interest rates are low. This is partly true, but misleading. In this white paper we will examine why, and why it matters.

One reason why organisations may hold too much inventory is that they underestimate the true cost of holding inventory. In this white paper we will also examine why this is an issue, what a realistic holding cost is and how it should be used.

What is inventory holding cost?

Holding cost is the cost of holding inventory and is expressed as a percentage of unit costs. If an item costs €100,000 to produce or buy, and it costs €20,000 a year to hold, its holding cost is said to be 20%.

Holding cost is made up of the cost of capital tied up in inventory, plus the operational cost of holding the inventory: warehouse space, material handling, obsolescence, insurance, and so on.

Importantly, holding cost is related to but not the same as the P&L (EBIT) impact of increasing or decreasing inventory. Holding cost is the total cost of holding inventory and is used in inventory management for calculating batch sizes and deciding on advance or discount purchases. It is greater than the P&L impact of changes in inventory for a number of reasons. Firstly, EBIT does not include the cost of capital, but secondly, holding cost includes a number of costs which cannot be switched on and off at will: warehouses, employment, loans, etc. Holding cost includes P&L items, but also balance sheet effects.

It is a mistake to ignore these “fixed” costs. If you believe inventory to be expensive to hold, you will try to hold very little. If you believe inventory to be cheap to hold, you will tend to hold a great deal. But smart organisations should avoid this. Why fill up a building with inventory when you might have better uses for the space, such as adding a production line, or subletting part of the building? And once the building is more or less full, a perception that all that inventory is needed can lead to sub-optimal longer-term decisions, such as adding a warehouse.

Remember there is no such thing as fixed costs, just variable costs with longer or shorter time horizons.

How much is holding cost?

Each organisation will have its own unique holding cost depending on its cost of capital and its operational cost structure. Holding cost can be difficult to estimate with any degree of accuracy, but it is worth putting some effort into generating an approximate figure for your own organisation.(1)

The operational cost can typically be broken down into the following categories:

1. Storage costs. Both external and internal warehouses, including, for instance
   1. (notional) rental costs
   2. Maintenance
   3. Temperature control
   4. Security
   5. Insurance
2. Obsolescence. This can be further broken down into sub-categories such as:
   1. Scrap (the total value of the inventory written off)
   2. Rework costs
   3. Discount costs (selling items at below cost)
3. Internal transfer costs. Moving inventory around between sites
4. Handling costs including, in particular, stock counting/stock control and especially those causing production stoppages
5. Management costs. The time spent by planning teams controlling existing inventory, dealing with obsolescence etc.
6. Overheads. All of the above requires IT, HR, Finance and other corporate time and expense

While building up an estimate, it is important to keep in mind the difference between total inventory holding cost and P&L impact. You may (rightly) feel that you need a storage warehouse, but this does not mean holding inventory in it has no cost. You will also need to make some estimates. If, for instance, you own a facility where half the space is filled with production lines and half the space is storage for inventory, then half the relevant costs for the facility should be included in inventory holding cost. Or if you have a team who are responsible for ordering, planning and controlling inventory, only their time spent controlling existing inventory should be included in holding cost. And so on.

Then consider the cost of capital. This should be the Weighted Average Cost of Capital (WACC). It is true that, writing in 2019, the global cost of debt is at historically low levels (2) and this in turn has led to relatively low WACC for asset-intensive companies, but this can be misleading. For a highly indebted company, the fact that WACC is low does not mean it can easily issue more debt. For such a company it would be undesirable for decisions to be made on the assumption of a low holding cost since it will lead to more capital being tied up in inventory. The opposite should be the case – these companies should do all they can to free cash from inventory and reduce their dependence on debt.

In fact, there is a compelling argument to ignore the calculated holding cost altogether and consider the opportunity cost instead. Look at the Internal Rate of Return (IRR) threshold for internal business cases to be approved. What is the highest IRR project that was rejected in the last 12-24 months? This can be considered the opportunity cost of the cash you have tied up in inventory and is a useful sense check that you are using a reasonable holding cost rate. The opportunity cost can easily be lower than your inventory holding cost, but if it is higher there is a strong strategic case to be made for using it as your holding cost.

Why does it matter – what is holding cost used for?

Holding cost is an important factor in the inventory levels organisations target and ultimately hold. This is because of the decisions which are made on the basis of assumed holding cost. To take the two most important instances:

• Optimal Procurement order sizes can be calculated by offsetting re-order costs and volume discounts against holding cost. This directly affects raw materials inventory levels
• Optimal Production batch sizes can be calculated by offsetting set up costs against holding cost. This directly affects finished goods inventory levels

Note in these two examples we say “can”. In many cases organizations make these decisions with no explicit holding cost. Instead, they might produce “lot for lot”. Or they might simply choose the batch/order sizes that minimize production/acquisition costs within the constraints of what is possible in terms of capacity and reasonable in terms of expected demand. If you find a year’s worth of a particular raw material in your warehouse because Procurement found a great deal on it, this effect is in evidence.

The importance of holding cost for working capital should not be underestimated. In inventory optimization initiatives a lot of focus is put on safety stock levels, but order/batch size is another key lever. Let me illustrate this with a simplified example:

Demand for a product is 100 units per month. It is produced in 400 unit batches every 4 months. This means on average (net of safety stock), 200 units are on stock over the course of a year.

Now halve the batch size to 200 units produced every 2 months. Average stock holding (again, net of safety stock) over the year is now only 100 units.

So what should you do?

Holding cost remains difficult to quantify precisely. Some approximations are necessary and diminishing returns are had from putting too much effort into it. But the final point to consider is the behavioural aspect. Operations departments have a constant focus on EBIT and only an occasional focus on cash. EBIT focus is a good thing, but can lead to an unconscious prioritisation of EBIT over cash. A key way to counter this is to ensure that a realistic measure of holding cost is used in calculations of batch sizes and factored into TCO calculations for procurement.(3)

We recommend you start by looking at how holding cost is used today and ensure that the holding cost used is not too low. 20% is normally a good starting point, or WACC plus 10% as an absolute minimum. We would then recommend trying to calculate your actual holding cost bottom up until you are confident you are in the range. Avoid having individual plants or business units do this themselves, since it leads to divergent results and much duplication of effort. Instead, Finance should establish a standard holding cost centrally and then ensure it is used globally. Multi-national corporations should find that holding cost varies from country to country in line with interest rates and there is a case to be made for high inflation countries to use a higher holding cost than the global standard.

From a behavioural perspective, if Operations has no working capital targets on their dashboard, there will be a strong tendency to hold too much inventory for the reasons we have set out in this paper. Fixing this is also important. But setting holding cost centrally is a strong lever that Finance can exert to free up working capital. Are you getting the most you could out of it?

If you would like to discuss how to take your inventory optimization capabilities to the next level, contact us: information@nventic.com

Notes

1. An internet search for inventory holding cost can quickly generate an approximate percentage for rule of thumb calculations. For instance, Investopedia suggests 20-30% although it does not cite a source for this statistic.

A textbook from 1993 reviews estimates of holding costs in academic studies between 1951 and 1990 (Lambert, D.M. and Stock, J.R., Strategic Logistics Management (third edition), Irwin (1993), page 366). All but one is within the range of 20-29% and while that one exception goes as low as 12%, that is only as part of a range from 12-34%.

Interest rates are indeed very low in 2019, even within the perspective of the last 100 years, although when Harris (1913) first derived the EOQ model he took holding costs to be 20% at a time when interest rates (in the USA) were at 4.5%, so only 2% more than 2019.

The CSCMP (Council of Supply Chain Management Professionals) commissions a yearly state of logistics report which includes a benchmark of inventory holding rates over the past 10 years in the US. The most recent figure, for 2018, was 18%.

2. The European Central Bank (ECB) at the time of writing in 2019 is actually offering negative interest rates on deposits (and 0% on loans). Of course, most companies cannot borrow at ECB base rates, but costs of borrowing are historically low, as is WACC, which factors in the cost of equity. Some very big companies, based on what they report in their annual reports, have WACC in the low single digits.

A benchmark of US firms from the start of 2019 found the average WACC for 6000 US non-financial firms was 8.22%. A KPMG study from 2018 focusing just on 276 firms in the DACH region, including 26 of the DAX30, is more or less identical (7% average WACC when including financial firms). Note that operational costs can easily add another 10-20% on top of WACC.

3. This often adds additional impetus to reducing EBIT effects too. An organisation that complacently uses large batch sizes with an assumption of low holding cost (or no sight of holding cost), when confronted with a requirement to factor in a holding cost of 20% or more, will often urgently have to address topics like SMED to avoid productivity measures being negatively impacted.

If you asked most supply chain professionals what super power they would most like, the ability to see the future would probably come fairly high up the list. Uncertainty – of demand, but also of lead times, failure rates, pricing, exchange rates, and many other variables – creates a high percentage of the headaches which supply chain professionals have to deal with on a daily basis.

But how much would 20-20 foresight really help? Ask Oedipus’s father! He got 100% forecast accuracy from the Oracle at Delphi, but little good it did him. (1) Amongst the many forecasting methodologies used by modern professionals, consulting high priestesses in Greece doesn’t normally feature too highly. But the Oedipus story illustrates an important limitation of forecasts: when you don’t have all the details, knowing what the future holds is not as useful as you think. This principle is very important in inventory management.

For the non-specialist, it can be tempting to think that forecasting is the challenge with inventory optimization. If only we knew what demand is going to be, goes the thinking, we could always have just enough inventory. There are two major dangers with this argument:

1. Forecasts are almost always wrong due to the inherent unpredictability of the future
2. Too much emphasis on forecasts can blind you to the importance of other factors. Even with a perfect forecast, you can easily get your inventory levels wrong

Let me illustrate this second point.

Product A Daily demand: 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2

Product B Daily demand: 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 60

In both instances the monthly demand is 60 units, with an average daily demand of 2 units. If your forecast was for monthly demand of 60 units, well done, your forecast accuracy is 100%!

Now think about how much stock you need to fulfil all orders in each case. Daily demand for product A is uniform – you need no safety stock. Each day you ship 2 units and produce 2 units for the next day. On average you hold 2 units. Product B is very different. In order to meet orders, assuming you don’t know which day the order(s) might fall, you would need to carry 60 units every day. That’s 30 times as much stock as for product A. Quite a difference, considering forecast accuracy was perfect and total demand was identical in both instances. (2)

It is said that forecasts are always wrong. The further into the future they look and the more detailed they are, the more wrong they are. Knowing this, forecasters normally refresh their forecasts regularly and avoid too much granularity – grouping demand into weekly or monthly “buckets”. But as our example has shown, even a perfect forecast of aggregate demand has limited usefulness without an equally good understanding of what variability to expect.

Rather than focussing too much on forecasting, inventory managers need to give adequate consideration to other factors, such as demand variability, lead time, target service levels, and so on. However, this is not to say that you don’t need to give proper attention to demand. It remains one of the most important inputs to inventory calculations and is probably the most likely to change. But to consider demand you don’t have to turn to forecasts.

Forecasting is a substantial discipline in its own right and one which we’re not going to explore in any detail here. At nVentic, our approach when calculating optimal inventory levels is to start by using actual daily demand instead of any forecast. There are a number of benefits to doing this:

1. You already know your actuals, so you don’t need to put effort into generating a forecast
2. Actuals are in their natural form and not aggregated into buckets that hide the actuals
3. You can use actuals to measure variability as well as average demand
4. Actuals are incontrovertible – you don’t need to spend time convincing other people how realistic they are
5. Actuals are immutable – as time progresses more actual data become available, but the actual data for any given time period will not change. Forecasts, however, are constantly overwritten and it is easy to lose track of what they were at the time plans were put together, especially since lead times vary by product

There are certainly cases where actuals are inadequate: for instance, with new product launches, where there are no actuals; or sporadic spare parts, where mean time between failures (MTBF) models may well be more useful. And there are events that planners need to be aware of and allow for that may not be visible in the actuals (sales promotions, patent cliffs, seasonality, etc.) as well as longer term product cycles which may be evident in the actuals (sales growth, decline, etc.).

We are not suggesting that using actuals is always better than using forecasts, or that you shouldn’t put efforts into improving forecast accuracy. Our approach is to start by using actual demand to calculate optimal inventory levels and then factor in the forecast if (and only if) you are confident it will be more accurate or more beneficial than the actuals. (3)

Conclusions

In inventory management, it is very important to understand and accept the limitations of any forecast and not to neglect the other key levers. Properly factoring demand variability into the approach is often a largely untapped area of potential. Shortening lead times takes some effort to achieve, but allows organisations to be more responsive and to carry less stock. And developing explicit service level targets and building them into your inventory approach is a very valuable way of balancing the opposing requirements of high service levels and low cost in an objective way.

If you follow this approach, we predict you will be pleasantly surprised by the results.

Notes

1. King Laius consulted the oracle at Delphi as to whether his wife, Queen Jocasta, would ever have children. He was told that any boy born from their union would end up killing his father. Therefore, when a boy, Oedipus, was born, they left him out on a hillside to die. Oedipus was spared but due to the sequence of events grew up ignorant of who his true parents were. In a classic example of a self-fulfilling prophecy, Oedipus ends up killing Laius precisely because he doesn’t know who he is – which one can assume would not have happened had Laius not heard the prophecy in the first place!
2. This example is deliberately simplistic. In reality a majority of items are likely to display more normally distributed demand, but these two extreme examples illustrate the point with a minimum of statistics.
3. How do you determine if the forecast you have been using is better or worse than actuals? In statistical terms, if the Mean Absolute Deviation (MAD) of your forecast vs your actuals is greater than the MAD of your actuals, you would be mad to use the forecast!

10 steps to supply chain heaven

Along with shortages, excess and obsolete (E&O) inventory is a good indicator of how well your supply chain is performing. Some waste is inevitable in most supply chains, since demand is unpredictable and shelf life is not infinite, but organisations can do a lot to minimise E&O, with both financial and environmental benefits.

Here we present a guide to the 10 steps organisations can take to reduce E&O, organised into 3 levels of increasing sophistication. As far as E&O is concerned, prevention is better than cure, but dealing with existing E&O should not be neglected, and is included here.

For those in a hurry, here are the 10 steps summarised:

1. Develop good basic inventory hygiene and visibility
2. Be sensitive to product lifecycles
3. Examine root causes and address them
4. Eliminate naïve desire for “100%” service level and apply better segmentation
5. Stop buying/making to a poor forecast
6. Shorten stock cycles
7. Reduce complexity
8. Reduce your need for safety stock
9. Use a richer blend of inventory management approaches to optimize by item
10. Leverage S&OP to drive inventory optimization

Level 1: The basics

Step 1 – Develop good basic inventory hygiene and visibility

There are a number of things which contribute to basic inventory “hygiene”. They are not necessarily pre-requisites to the more advanced steps below, but they are simpler to implement and will add significant accuracy and clarity to further steps.

First, ensure good booking processes. Inventory needs booking in and out accurately, and in a timely manner. Failure to do this can both distort demand and supply signals further down the line and give planners an inaccurate sense of how much inventory is held in the first place. It can also lead to inventory getting lost. Automated warehouses, barcodes and RFID can help minimise this type of error, although not eliminate it entirely. A good lagging indicator of booking issues is when systems show negative stock balances!

Then make sure accurate stock counts take place. However good you believe your booking processes to be, it is important to verify the actual figures regularly. Depending on the scale and complexity of your facilities, cycle stock counting might be a useful enhancement to, or substitute for, an annual stock check. Physical counts, even if randomly carried out, will show up both missing inventory and forgotten inventory.

And finally, in terms of hygiene, a sturdy broom is essential. There may be a natural aversion to loss, but holding on to obsolete stock in the hope that it eventually finds a buyer or another use is unwise. Accounting rules and financial considerations will determine what needs to be devalued when, but even fully devalued stock takes up space, effort and management time. Anecdotes of something generating value well after it should have been used are legion, but it is overwhelmingly better to clear such stock out. The first consideration should be the use of discounts to promote sales, followed by options to rework or recycle, before finally scrapping.

All of the hygiene factors above should help improve the quality of inventory data at your disposal. Beyond this, especially where you are coordinating inventories across multiple facilities, visibility is also critical. You need to know how much you have, where and when. “Blind spots” or inadequate coordination can lead to shortages or excesses when considering the network as a whole.

While good booking processes should deliver accurate data in ERP systems, those systems can be laborious and inflexible when it comes to visualising the data in user-friendly ways. Various ERP add-ons, as well as third party software solutions, exist to enhance inventory visibility. Increasingly, these visibility tools allow organisations to see inventories in their wider network, such as at suppliers or customers.

Step 2 – Be sensitive to product lifecycles

A common cause of E&O inventory relates to product lifecycles. Where historical data exists, supply chain teams can relatively easily ignore the more optimistic forecasts of their sales organisations, but with new product introductions a very optimistic forecast can often be king, leading to overstocks in the introductory phase. In extreme cases the initial supply can be larger than the all-time demand.

Similarly, as products reach the end of their lifecycle, E&O is in danger of being generated, precisely at the worst point in the cycle. This could be caused by a failure to recognize declining demand, a failure to communicate that sales focus is switching to new products, or simply a failure to plan for the switch.

While operational silos and the natural tension between sales and supply chain teams can seem one of the most intractable problems to solve – and we return to Sales and Operations Planning (S&OP) in step 10 below – sensitivity to product lifecycles and basic communication is sufficient to avoid the worst E&O effects in this area.

Level 2: Good practice

Step 3 – Examine root causes and address them

With the basics (steps 1 and 2) hopefully in place, the first step towards good practice in E&O reduction is creating a culture of continuous improvement relative to inventories. As with any business process, sustainable improvement comes from measuring outcomes, digging into root causes, addressing them and then measuring the improvement.

The key challenge with inventory is often quantifying the impact of different root causes, since inventory sits right at the heart of your supply chain and is influenced by many things simultaneously. In the absence of good data there is a danger of particular root causes being singled out and blinding the organisation to other contributory factors. This can be as much on the basis of individual perception as fact.

As you move through the more advanced steps below, you should along the way develop more sophisticated analytics to better understand the relative importance of different inventory levers. However, it is rarely a good idea to wait for perfect data before starting improvement efforts. A lot of your instincts as to what root causes are driving E&O will be correct. The important thing at this stage is to identify and take actions, and to measure the results.

A critical success factor here is the implementation of short interval controls and KPI’s to focus attention on E&O and inventory levels more generally. You can’t improve what you don’t control and you can’t control what you don’t measure.

Step 4 – Eliminate naïve desire for “100%” service level and apply better segmentation

Even before getting into the realms of inventory science (see below), you can reduce E&O by doing two simple things. The first is to make some conscious and differentiated decisions about the service levels to provide your customers – both internal and external. Consider if a lower service level would be more profitable. The second is to segment your inventory items in ways that allow you to prioritise what is most important and deprioritise the rest.

You may use different metrics to measure service level, such as OTIF, fill rate, out of stock time, back order quantities, waiting time, and so on. Whatever measure you use, in the face of uncertain demand and supply you can never entirely guarantee availability. The role of inventory is not to prevent all shortages but to keep shortages to an acceptable – in most cases the most profitable – minimum. This is an important principle to grasp. Depending on your market, you may be delivering an unnecessarily high service level.

Start by baselining your current service level performance (using whatever service level metric is already in use and familiar to you) and on that basis set some service level targets depending on how important something is for your business. There may be some items where you do need to have very high availability on demand, but this is not necessarily the case for everything. Moving to a more segmented model where you have different service levels for different items, rather than the same for all, will help reduce E&O.

In the absence of other criticality criteria (for instance medication that is essential to the preservation of life), use profitability or simply revenue to segment your products. An ABC analysis would be a good starting place. List your items by total value (value * annual volume) and then segment the top 80% (A), the next 15% (B) and the bottom 5% (C). An improvement would be to group by finished goods product, since a low value raw material might be essential to a very high value finished product.

ABC analysis is very widely used. Less so is XYZ analysis, where variability is a second segmentation dimension. XYZ analysis is more complex than ABC analysis to do, since it involves statistics, but it should be a bigger driver of inventory levels, since variability is precisely what inventory is buffering against. At this stage do not worry too much about the mathematics of XYZ but find a segmentation that works for you, is clear to all in your organisation, and separates your most variable demand products from your least variable.

Various other segmentations exist, such as VED (Vital, Essential, Desirable), or FSN (Fast, Slow, Non-moving), to name just two. All such segmentations can be useful methods to focus the prioritisation of management attention, depending on your individual circumstances.

At this stage don’t worry too much about the exact segmentation rules. Ensure that items you know to be the most critical to your organisation are in the top priority category.

Then set some target service levels, bearing in mind the nature of your business. If you stock items where it is critical to have very high availability on demand, for instance foodstuffs or critical medical supplies, then a target service level of 99% or higher might be appropriate. Whereas if your customers are used to waiting a few days or weeks then you can make do with a lower service level.

It is important to baseline your service levels and track their development. Inventory reduction programmes frequently meet resistance and then it is important to show that service levels have not suffered unless a conscious decision was made to lower them.

The most important principle here is differentiation – that not all products need the same levels of inventory. If you struggle to equate service level with required inventory levels then try to differentiate on the basis of cover, i.e. how many days’ or weeks’ average demand you aim to hold in inventory, although note that cover is far from optimal as a concept and will be addressed further in step 9 below.

By segmenting your products, by setting explicitly different target service levels and different target inventory levels by segment, however approximately, you are starting to familiarise your organisation with some of the most important principles in inventory optimization and already delivering benefits relative to a simplistic “one size fits all” approach.

Step 5 – Stop buying/making to a poor forecast

We frequently hear the complaint that poor forecasts are to blame for E&O inventory. Most of the time this is at least partly true, but not necessarily in the expected way. It is frequently the use of the poor forecast, not the poorness of the forecast, that is the biggest problem.

If you have a forecast demand for 100 units of a product and you produce 100 units, but then only have demand for 10, it is easy to blame the forecast. If you have a forecast demand of 100 units every month and produce 100 units every month, but only ever have demand for 10 units per month, the blame should be with you for using the forecast.

Of course, in most real-life situations it is less clear cut. Forecasts are not usually so consistently wrong, nor demand so uniform. In reality, there is a lot of “noise” to deal with – demand goes up and down, forecasts are frequently revised. It is not necessarily so obvious that forecasts should be ignored. But what do we mean by forecasts should be ignored? Surely some forecast is required for any inventory planning exercise.

A useful starting point is to consider actual demand for each product. There is incontrovertible evidence of what has already been sold. If you compare this with your forecast, you can calculate if the forecast is adding value or destroying value relative to just using actual sales as the forecast. While superior methods to measure forecast value add exist, here is a relatively quick method to check if your forecast is better or worse than using actuals as a forecast: consider whether the mean absolute deviation (MAD) of your forecasts vs your actuals is greater or less than the MAD of your actuals. To take a simple example:

Forecast demand for 10 periods: 10, 10, 10, 10, 10, 10, 10, 10, 10, 10

Actual demand for those periods: 7, 8, 8, 11, 6, 5, 7, 13, 6, 6

The MAD of the forecast vs actuals is (3 + 2 + 2 + 1 + 4 + 5 + 3 + 3 + 4 + 4) / 10 = 3.1

The MAD of the actuals (deviations from mean of 7.1) is (0.1 + 0.9 + 0.9 + 3.9 + 1.1 + 2.1 + 0.1 + 5.9 + 1.1 + 1.1) / 10 = 1.72

In this example, you would be better using your actuals as a forecast than the forecast. Or at least you probably would! A key thing to understand about any forecast is that it is almost certainly going to be wrong.

When doing this exercise, it is important to look at the forecast at the time of planning. For instance, if orders or production plans are confirmed 10 days in advance, the forecast on that day (i.e. actuals minus 10) is the one to compare with the actuals. Because forecasts are frequently revised, it is all too easy to lose track of what the forecast was at the critical planning horizon.

Before we go into more advanced approaches below, the important thing to retain here is that it is very valuable to understand where your forecast is destroying value and to decouple the forecast from the plan in those cases.

There is also a behavioural aspect to this step. We have worked with numerous organisations where the sales team were responsible for creating and maintaining forecasts in the system and where planning teams planned to those forecasts even though they knew them to be consistently wrong. There can be a culture of “we know the forecast is wrong, but if it is right and we haven’t got the stock then it’s our fault, whereas the forecast being wrong isn’t our fault”. This silo mentality needs breaking down.

It is also important to understand that forecasting is not the only game in town when it comes to reducing E&O. A lot of demand is inherently unforecastable and setting up inventory practices to mitigate this uncertainty is a more reliable approach than trying to second guess it.

Level 3: Leading practice

Step 6 – Shorten stock cycles

Large production batch and purchase lot sizes drive high cycle inventory. The more inventory you have, the more likely it is to be excessive and at risk of becoming obsolete. By making or purchasing fewer items at a time, more frequently, E&O can be reduced. The challenge of achieving this can be approached from two directions.

Firstly, by minimizing set up and reorder costs. Techniques like single minute exchange of die (SMED) drive down the cost of production setups. Procurement efficiencies, such as the use of automated purchasing tools, make it less costly to reorder.

Secondly, by ensuring that inventory holding costs are properly factored in. All other things being equal, production teams like large batch sizes because it reduces the number of changeovers necessary and procurement teams like large lot sizes if it allows them to take advantage of supplier volume discounts. But the “savings” of these high volumes need to be offset against the holding cost of the inventory they generate and this is where mistakes can be made. The cost of capital, especially as expressed by public interest rates, is much lower than inventory holding costs.

The true holding cost of inventory is made up of not just your organisation’s weighted average cost of capital (WACC) but also all of the operational costs involved in holding inventory, including not least the cost of obsolescence! By making your organisation aware of the true cost of holding inventory, in other words, you can help reduce E&O.

Where production and procurement are forced to use a higher holding cost, batch/lot sizes should decrease, assuming that they are using some form of total cost calculation to define their batch/lot sizes in the first place.

Step 7 – Reduce complexity

Complexity is the enemy of inventory efficiency for a variety of reasons.

First is the impact on forecast accuracy. Forecasts are more accurate at an aggregate level than at granular levels. It is much easier to accurately forecast demand for a whole class of product than for each individual variant of that product. Simply by having a high number of alternatives, you are increasing your chances of having E&O inventory. Of course, you still need to convince your marketing department of the benefits of standardisation!

The second is the sheer proliferation of items. We frequently work with clients with many tens of thousands of different items in inventory. The very volume of items makes it harder to stay on top of optimum inventory levels for each.

Thirdly, a lack of standardisation makes items much less flexible. By having a high degree of standardisation at a sub-component level you allow yourself to re-purpose or even re-use items in a variety of situations.

And fourthly, the number of stocking points you have drives how much inventory you need. The square root law is an approximation which states that the amount of inventory required by a system is proportionate to the square root of the number of locations in which it is stored. So, for example, if you move from 1 warehouse to 4 warehouses, your inventory requirements will double. Since demand volatility at each individual warehouse will be higher than the aggregate demand overall, risks of E&O increase. Of course, there can be good, even necessary reasons to increase your number of stocking locations – retail premises being a prime example – but if you are adding a new warehouse just due to capacity constraints at your existing one, the E&O effect is one to bear in mind.

Inventory is of course only one consideration in network optimization and technology exists to help optimize network design, but as a basic rule of thumb the square root law is a useful heuristic to facilitate E&O reduction.

Complexity quickly asserts itself in large organisations, and simplification is often a long-term lever to reduce E&O but it is a very significant root cause and should not be ignored.

Step 8 – reduce your need for safety stock

There are a number of important parameters which affect how much safety stock you need. Safety stock exists to ensure the ability to continue supply in the face of variability on both the demand and the supply side. Where demand is variable and lead times are long, you can feel squeezed in the middle. Both sides can be influenced to reduce E&O.

On the demand side, the most important step is trying to reduce variability. We have already (step 7) discussed the benefits of product standardisation to reduce demand variability, but price stability is also beneficial – discounts and other promotions introduce additional variability into the supply chain. Even when prices are being discounted to dispose of obsolete or close-to-obsolete stock (step 1) it should be remembered that this is liable to have a knock-on effect on other product lines.

On the supply side, there are benefits both to reducing the variability of lead times and to reducing the absolute lead times themselves. Reducing the variability of lead times sounds easier said than done, since, at least for raw materials, it usually comes down to external factors, but this is not to say that improvements cannot be made, especially within production.

A more straightforward lever may be reducing lead times. Within a factory, this comes down to reducing cycle times while maximising throughput. With third party suppliers, the key lever is finding more local sources of supply. Of course, there may be cost benefits to sourcing from further afield but this needs putting in the proper context of the holding cost of inventory (see step 6 above).

It should also be remembered that long lead times not only increase the safety stock you need, but also increase what we call pipeline stock: inventory that is not yet showing on your books but which is already in transit from your suppliers. Such inventory makes you less agile and less able to respond to changing demand signals. A similar phenomenon is caused by long production frozen periods. While frozen periods can be good for production efficiency, they reduce agility and will tend to drive higher E&O.

In summary, reducing lead times, delays, lags and variability throughout your supply chain will reduce your need for safety stock and so reduce E&O.

Step 9 – Use a richer blend of inventory management approaches to optimise by item

While standardisation helps reduce E&O if applied to product design (see step 7), differentiation is what holds the key to unlocking substantial benefits when it comes to the management of inventories. No two items that you stock are the same. Segmentation (considered in step 4) is a useful improvement on “one size fits all”, but flat policies across segments, such as setting cover targets by segment, ignore differing variability at an item level. In terms of the individual parameters that will allow you to reduce E&O while maintaining service levels, both safety stock and cycle stock need to be defined at an individual item level.

To calculate optimum stock levels by item, scientific approaches exist. Fortunately, given the data challenge involved, tools exist to calculate these parameters for you. Unfortunately, all such tools by necessity work on a number of assumptions, which means that their recommended parameters do not work for many items. Thus, while the optimum parameters can be calculated for each and every item, it is currently somewhat difficult and laborious to do so with a reasonable degree of accuracy for all items.

Safety stock and cycle stock are essential concepts when it comes to replenishment models. Here, by replenishment models, we mean an approach where target safety stocks and order quantities are set, along with a re-order level. When stock on hand drops to a given level, more is ordered or made. In replenishment models, target safety stock can be calculated to handle expected variability in demand up to a defined service level. In general, replenishment is an underutilised resource, since many people find variability a difficult concept to factor in well.

The main alternative to replenishment is deterministic planning. This seems to be more widely used since it requires less understanding of statistics. In deterministic models, stock is bought or produced in direct response to a demand signal, whether confirmed orders or a forecast.

Deterministic planning works well when demand is known (such as through confirmed customer orders), or at least is known to be different to the past (such as for a promotion, or the phasing out of a product). In other words, to go back to the theme of step 5, if you are certain that your forecast is better than actuals for predicting future demand, then deterministic planning should be superior.

While this is a simplification, and care needs to be taken with terminology, deterministic planning is often championed by proponents of Material Requirements Planning (MRP) systems, where “MRP” is selected as the production planning method. The principle here is that requirements for all components can be established through the explosion of the Bill of Materials (BOM). While finished goods might be managed using a replenishment approach, since end-customer demand is unknown, raw materials, WIP and semi-finished goods are often managed through a deterministic MRP model that reflects the production plan (which itself reflects the sales forecast).

The problem here is that if customer demand is unknown and MRP is being used to plan deterministically, you are just passing your variability, and stock imbalances, up the supply chain. Alternative approaches like Demand Driven Material Requirements Planning (DDMRP), CONWIP or Kanban, seek to counter this by introducing decoupled replenishment points in the supply chain.

The debate for and against MRP is sometimes presented as a binary choice. In reality, there are situations in which MRP works very well, and situations in which it doesn’t. This is where differentiation is essential. Enterprise Resource Planning (ERP) systems offer a choice of strategies, including not only MRP and replenishment as set out here, but multiple variants too.

The very number of options available, along with all of the essential parameters which go with them, can initially seem overwhelming, but restricting yourself to one or two means that an opportunity is being missed. Differentiating your approach can bring enormous benefits. We recommend an incremental approach. Do not go from 1 stock policy to 15 overnight. Do not trial a new stock policy with all of your items. Instead, identify classes of item suitable for a different approach and then test it with a restricted number of items before scaling up.

Similarly, where you are using a new tool or method to define target stock levels, do not implement them in full immediately, but work to them gradually. Let us say your new tool has suggested you can reduce your safety stock for an item by 50%. You may or may not understand exactly how it has made that calculation, and it may or may not be very precisely correct, depending on what assumptions the tool is making and how good they are.

The point is not that the tool is weak and should be discarded – it has suggested something that you had not thought of and it may well be right – the point is that if you drop safety stocks by 50% and that is too much, you will end up with shortages. This will damage the credibility of you, the tools, and any further optimization initiatives. It is much better in the example given to try reducing safety stocks by, say, 20% at first, watching what happens carefully, and then reducing them further. Don’t go for the full 50% until you are confident.

This step 9, which embraces differentiation in inventory management approaches, has enormous potential. nVentic’s extensive experience and diagnostic tools show that the application of this step 9 alone can typically deliver reductions in inventory levels of 20-50%, all but eliminating E&O, without sacrificing service levels. However, this approach also involves substantial complexities and numerous pitfalls. To be successful, we believe it is essential to develop capability in your supply chain teams so that you have a better understanding of the science rather than just trusting in “black box” solutions which you know from personal experience to have limitations. Here too, a phased approach with incremental improvements year over year is appropriate.

Scientific approaches to inventory optimization hold enormous potential, despite the complexities involved. Existing technology, whether driven by formulas, artificial intelligence or modelling, is currently unable to take account of all variables. The human mind still has a vital role to play. But advanced scientific methods and tools can help you to go a lot further in reducing E&O without service level suffering. Where you have a large number of items to manage then you will benefit greatly from using technology to optimize at an individual item level. The limitations of the technology do not mean you shouldn’t use it, they just mean you should seek to understand those limitations and work with them.

Step 10 – Leverage S&OP to drive inventory optimization

It may surprise you to find Sales and Operations Planning (S&OP) this far down the list. After all, S&OP is surely one of the most important planning processes at your disposal and you certainly wouldn’t want to exhaust all of the scientific approaches to inventory optimization (step 9) before turning to S&OP. Indeed, as we said at the start, these 10 steps are not to be thought of as a strict sequence, where each step needs to be completed before progressing to the next.

The reason we have placed S&OP at the end is due to its strategic nature. Done well, S&OP is the way your organisation makes its most important supply chain decisions – how much should be bought, manufactured and stored in order to maximise your corporate goals. Various strategic decisions can and should flow from S&OP. Achieving the right balance between conflicting management incentives. Deciding what should be made to order and made to stock. How to find the best balance between demand and supply lead times. (Where customer promise times are short and supply/manufacturing lead times are long then E&O will thrive.)

S&OP is also a vital counterweight to silo behaviour. In simplistic terms, sales organisations like high inventories, since they never have to worry about stock outs. Having everything available on demand makes their vital revenue-generating role easier. Production and supply chain organisations, on the contrary, look to control costs and maximise efficiency. This natural and healthy tension comes for resolution to the S&OP process.

However good your S&OP process, you will still generate some E&O inventory. Inventory is itself a strategic lever. You choose how much you will hold to buffer yourselves against variability. It is only one possible buffer, the other main ones being time and capacity: you can deliberately retain spare capacity to be able to react to changes in demand, or you can use time – asking customers to wait in the event of demand spikes.

We have placed S&OP as the last step of our 10 for two interrelated reasons. In the absence of robust analytics, strategic transparency is limited. Decisions are made on the basis of incomplete information and organisational politics can become the guiding force. Only once you have full transparency of the relative costs and benefits of alternative approaches can you be confident of making the best decisions. And at the same time, it is of limited value to have the best inventory analytics in the world if you cannot bring the rest of the organisation with you. S&OP needs inventory analytics, just as inventory analytics needs S&OP.

Reducing E&O inventory may be seen as a desirable tactic, but at a certain level it needs to be weighed in the strategic balance, since fully eliminating E&O is likely to be impossible without service levels suffering, although we have never seen any organisation reach that level in practice. Rather, the application of a full range of inventory optimization approaches will allow you to reduce E&O even while you maintain or improve service levels and this win-win scenario is vital to sustainability. You know that you have truly achieved mastery of E&O when all stakeholders see inventory optimization as something which helps them, rather than something which endangers the objectives of sales to further the objectives of operations.

Honourable mentions:

We are coming towards the end of our guide to reducing E&O inventory, and you may feel that some things have been missed from our ten steps. Two notable ones we chose to leave out are vendor managed inventory (VMI)/consignment stock and multi-echelon inventory optimization (MEIO). Let’s briefly consider them here:

VMI is where you outsource the management of some of your inventory to your supplier. Consignment stock is where inventory in your warehouse belongs to your supplier until you use it. You might use a combination of VMI and consignment stock or just have one or the other. Consignment stock is a tempting quick win when you want to reduce inventory levels, since you are getting inventory off your books without it leaving your warehouse. VMI might seem a good option if you believe your supplier to be better at managing inventory than you are.

There can be benefits to both VMI and consignment, especially in terms of vertical supply chain visibility – your supplier will have granular visibility of your actual consumption of their products, rather than relying on your forecasts. However, we chose to leave both out of our guide. VMI and/or consignment might be good for you, especially at lower ends of your own inventory management maturity curve, but it is essentially outsourcing a problem rather than solving it, and it has downsides as well as benefits: A lot of work needs to go into the service level agreement if it is to work well, the relationship needs managing closely, it can be difficult to have good visibility of what your supplier is doing or if you’re getting value for money. With consignment stock, you still have a number of the holding costs, such as storage and material handling, even if your cash is no longer tied up in the stock itself.

In short, while VMI/consignment stock doesn’t preclude good inventory management, neither does it automatically deliver it. Use it judiciously and, if you use it, put effort into doing it well.

MEIO we left out for a different reason. In theory, MEIO will be better than single echelon inventory optimization, but in practice it is fraught with difficulties. Here we need to be very clear with terminology. It is always good to work out the best place in your supply chain to hold inventory. This is often referred to as Strategic Inventory Positioning (SIP). It is also good to optimize all echelons in your supply chain. With single echelon optimization you decouple each site in your network from the others and treat demand signals between them as independent. So safety stock in your manufacturing facility might be calculated on the basis of the variability of the demand coming from your distribution centre. With MEIO you treat your whole network as one: you optimize what is held at each point on the basis of the whole.

This is good in principle, but there are currently a number of challenges with applying it in practice. Firstly, there is the computational challenge – effort is exponentially driven by permutations. Secondly there is an issue with fuzziness. MEIO is highly dependent on a number of parameters like waiting times, which are normally hard to quantify accurately. Finally, and perhaps most importantly, there are challenges with implementation. To realise the benefits of MEIO, you need a differentiated range of service levels. You might want a given location to deliver a service level of 50% for one item, 80% for another, 98% for a third. Where you have managers used to targeting, say, 95% for all items, on the understanding that 96% is even better, this is immensely complex.

In short, while MEIO has potential, especially for reducing E&O since you are lowering overall levels of inventory across the network, we wouldn’t recommend anyone to try it unless they are already very advanced in all of the other approaches. Implementing MEIO successfully demands high degrees of competency in our steps 9 and 10 as a prerequisite. And even then, it is probably only worth the effort beyond heuristics for a relatively small number of items. For the vast majority of organisations, it will be better to focus on our 10 steps.

Conclusions:

So there you have it. The ultimate guide to reducing excess and obsolete inventory. Did we miss anything out? If we did, or if you have any questions on the content of this guide, contact us and let us know.

This guide has deliberately been technical in nature and is intended as a checklist of things to try when aiming to reduce E&O. It does not pretend to be comprehensive on all topics. Rather, we hope it might act as a jumping off point for new approaches. Organisations that have very low E&O invest in their people, systems and processes to deliver sustainable benefits. Getting whole organisations to perform strongly on this measure takes time, effort and commitment.

And yet, as we head into the 2020’s, it is to be hoped that more organisations prioritise E&O reduction. E&O has a negative financial effect, but it is also bad for the environment. Every product that is surplus to requirements represents resources and energy wasted, from the extraction of raw materials, via multiple stages of processing, transportation, packaging and so on. Wherever you are in your E&O reduction journey, it is always worth striving for the next level.

When it comes to inventory optimization, for many people the first and sometimes only focus is safety stock. This is not unnatural, but actually one of the first things to understand about safety stock is that it is not the only show in town. Neglecting other types of stock, such as cycle stock, pipeline stock, anticipation stock, congestion stock, and so on, is a mistake.

But our topic for this article is safety stock. Let us start with a definition:

Safety stock is inventory kept to buffer against uncertainty on both the demand and supply side.

The value of safety stock is in allowing supply to continue uninterrupted most of the time. The downside of safety stock is that it can lead to excesses and eventually obsolescence.

So far so simple, but there are different ways of defining the amount of safety stock that you need. We will look at some of the most common ones in this article. We will also consider how often you should revise safety stock levels, where in your supply chain you should keep it and what alternatives to safety stock exist.

To start, though, we will differentiate safety stock from contingency stock and discuss how different types of inventory require different safety stock models.

Safety stock and contingency stock

First of all, let us differentiate between safety stock, which is designed to buffer against typical uncertainty, and what we shall call contingency stock, which is designed to buffer against unusual events. This is particularly relevant at the time of writing, since the impact of the Covid pandemic has been much more substantial than the regular uncertainty you get from the occasional supplier shipment being delayed or the usual fluctuation in customer demand.

Do you need contingency stock?

Contingency stock is appropriate in two scenarios.

The first is where your inventory is absolutely essential, i.e. related to the preservation or protection of life. Hospitals and the military are examples of organisations that routinely hold contingency stock: they need to be prepared to withstand substantial disruption on the supply side as well as major swings on the demand side.

The second is where you are trying to assure business continuity in the face of a specific supply-side risk. For instance, if there is only one possible source of supply for a key input material to your products and you have concerns about your supplier’s reliability, you might choose to increase your inventories of that material in case there is a disruption.

The right-sizing of contingency stock is a type of risk management and closely related to business continuity planning. You have to assess the level of protection you want to have, for instance, the ability to continue without any supply for 6 months, or the ability to handle a sudden doubling in demand (eg for flu vaccines), and so on.

You can never protect yourself against all possible scenarios (what happens if supply is disrupted for 12 months? For 24 months? What happens if demand triples or quadruples?) and you also have to weigh the risk of running short against the waste involved if you never need your contingency stock. This is why contingency stock is only appropriate in very specific circumstances, in a lot of which a public or governmental organisation will directly or indirectly be picking up the bill.

One size does not fit all

For the rest of this article, let us focus on safety stock itself. When working out how much safety stock to hold, there are 3 main considerations to factor in before you even start calculating. The first is the nature of your demand, the second is the service level you are targeting, and the third is the planning methodology you use.

1. The nature of demand. Demand can be uniform or variable, flat or nonstationary (i.e. growing or declining), regular or sporadic, predictable or unpredictable, and in terms of probability distributions it can demonstrate normal distribution or other types of distribution such as gamma, poisson, or negative binomial. Noting that whatever it is at the moment, it can change at any time. There is a lot of complexity here. The key point to understand in a first instance is that tools and models designed to define ideal safety stock levels are highly unlikely to factor all of these dimensions in and so will work better for some items than others
2. Service level. Service level is often poorly understood in this context, but is a very important input to a lot of safety stock models. It represents the likelihood that stock is available when required. Mathematically, only infinite stock will deliver a fill rate of 100%, but organisations can target a specific fill rate (frequently in the 95-99% range) and build safety stock to try to assure that level. Read our article on supply chain service levels for a deeper dive
3. Planning methodology. There is an important difference between deterministic and replenishment methods. In a replenishment model, you define the required safety stock to deliver a given service level for expected average demand and variability. With deterministic models you are planning on the basis of given demand. As such, the concept of safety stock is in theory redundant, although in practice deterministic methods (such as MRP) still have to deal with some level of variability. A common cause of overstocks is when organisations size safety stock as if they were using a replenishment model when in practice they are planning deterministically, often to a forecast

The three considerations above are important factors in segmenting your inventories. Before trying to optimize your safety stocks it is important to understand what different types of inventory you are dealing with. Trying to use the same model for calculating safety stock across all items is highly likely to lead to sub-optimal outcomes.

Different methods for calculating safety stock

To define what level of safety stocks you need, a number of methods are possible:

• Safety stock based on cover. This is where you define safety stock in terms of estimated average demand for a given period. For instance, your policy might be to try to maintain 2 weeks’ safety stock of all items. The only benefit of this method is its simplicity to understand and, no doubt for this reason, it is very commonly found. However, it is almost certain to lead to shortages and/or overstocks since predictability, demand variability and supply lead times are highly unlikely to be the same from item to item
• Safety stock based on maximum expected demand over the lead time. This is where you consider the maximum demand over the lead time and use it to define your safety stock. Typically, historical data is used to ascertain what the maximum demand ever experienced over the lead time is. Proponents of this approach like the fact that it is non-parametric (i.e. it does not seek to apply a specific statistical distribution), but this is also its weakness, since working out what proportion of the maximum to use requires deep statistical understanding, unless you just use the maximum itself (i.e. safety stock = Maximum lead time demand – Average lead time demand) but then you are in theory delivering 100% availability which will be excessive in most situations, especially if you have outliers in your historical data
• Safety stock based on service level. This is where you define the desired service level and then apply a statistical distribution to define the safety stock. A benefit of this method is that, unlike cover, various service level metrics, such as fill rate, can reasonably be set as standard across items. For instance, targeting 99% fill rate for all A items. The service level acts as a proxy for optimizing based on cost, but avoids a lot of the work calculating the actual costs, which can be very labour-intensive or imprecise to do. It also allows you to set higher service levels than a simple cost optimization might deliver, which is a strategic choice favoured in many industries. The main disadvantage of this approach is that many professionals are uncomfortable dealing with the statistics involved. However, most inventory optimization technology uses this approach, so in our view it is well worth making the effort to explore it further

This is a high-level overview of the main methods. We will now look a little deeper at the third method, defining safety stock based on service level. While not perfect, we find optimization based on service level most commonly to be the best option, especially when applied at scale. This is also no doubt the reason why it is the approach used by almost all inventory optimization software. We will also look at some of the major exceptions where it may not work well.

Defining safety stock using service level

To define optimal safety stock using service level, you can use one of a number of equations which use well-established mathematics. Let us break down one of the most commonly found equations to understand how it works and what some of its important sensitivities are.

Where:

SS = Safety stock
Z = A measure of how many standard deviations above or below the mean a value is
σ = (Sigma) a standard deviation from the mean
Vol = Average volume of demand in units
LT = Lead time (time between ordering and receiving an item)

What the equation does is define the amount of safety stock you need relative to three things:

1. The lead time. This is intuitive. If you can get something the next day you will need less safety stock than if it will take 3 months to arrive
2. The absolute variance in demand. So if demand fluctuates between 100 and 200 units you will need more safety stock than if it fluctuates between 100 and 120 units
3. The target service level you want to achieve. This is done by considering all of the likely values based on the observed distribution of demand and then defining what percentage of those you want to cover

The example given above only allows for variability on the demand side. If you also want to factor in supply-side variability, you can use a modified version of the equation:

This isn’t a maths lecture. The thing to retain here is that this version is also factoring in how much the lead time itself varies.

So in summary, if demand, variation in demand, lead time, variation in lead time or your desired service level increase, you will need more safety stock. If any of them decrease you will need less. So much you probably instinctively know. What the equation does is help you to quantify by how much. Other equations exist, this one was chosen as it is fairly commonly found and relatively easy to understand.

Exceptions and special cases

However, there are some important sensitivities to this equation. Firstly, it uses cycle service levels. (See our article on supply chain service levels on this point, especially the section at the end on calculating safety stocks using service level.) Secondly, it assumes normally distributed demand. This is frequently good enough, but especially where demand is very variable and/or sporadic it may not be robust. Thirdly, it assumes average demand to be flat (stationary), not increasing or decreasing. There are further sensitivities, but already you can see how using a well-known equation has a number of pitfalls which could lead to inaccuracy in your modelled safety stock levels, and that it is likely to be problematic for certain types of inventory, especially those with sporadic demand.

In short, the equation given works reasonably well in a replenishment model for items which have moderate variability. If you are managing items with very sporadic demand (eg spare parts that are only infrequently needed) then it is unlikely to work as well. More complex models exist for sporadics, but in a first instance a maximum lead time demand model might be a good safe place to start.

Similarly, care needs to be taken with demand which is clearly growing or declining. This too is a more complex area, but in short, two workarounds are either to factor in the smoothing constant from your forecast or to make a series of calculations at the projected future levels of demand.

Optimizing your safety stocks – a systematic approach

So what is the best approach to optimize your safety stocks? Well maybe a good place to start would be in measuring how much you have, compared to what you think you should have. Another way of defining safety stock is the average stock on hand when a new order arrives.

Safety stock is the average stock you have on hand when a new order arrives

Since actual demand rarely matches forecast demand, there will usually be a difference between your actual safety stock at the end of each cycle and your target safety stock. (The whole purpose of safety stock is, after all, to buffer against this variability! If in your mind safety stock should never be used then you are thinking about sediment, which is surplus, or the level of inventory never used.)

However, over time and on average the actual should be the same as target unless you have forecast bias. If you consistently have less safety stock than planned then there is a case to be made for increasing it, and if you consistently have more then you should consider decreasing it. (Or, if you never run out, a case can be made for reducing your planned safety stock!)

This doesn’t address how much safety stock is optimal, but it does at least allow you to cancel out any forecast bias, is simple to compute and may allow you to make some improvements quickly, by increasing or decreasing your target safety stocks where you observe systematic understocks or overstocks.

Another step to consider is what we call changing the givens. We saw above how lead time, demand variability and service level are all key inputs to any optimization calculation. Anything you can do to reduce lead times and demand variability, even to relax service levels in some situations, will reduce your need of safety stock. You still need to calculate how much you can reduce it, but all of these steps certainly help and in some instances are addressed in advance of any attempts to optimize safety stocks since they bring other benefits.

The next step is to segment your inventories. When implementing a new approach to safety stocks it is unadvisable to switch everything over in one go. Much better to trial with a smaller number of items and watch carefully what happens before rolling out further.

There are many ways to segment your inventory. A good starting place is an ABC analysis, whereby A items account for the top 80% of your inventory turnover (value * volume in a year), B items the next 15% and C items the remaining 5%. As a general rule, set C items aside. By definition they tend to be rarely used, but this doesn’t mean they can’t be critical when required. As such, it is often best to set a conservative safety stock for C items to avoid stock outs, which are likely to be more costly than the value of any benefit from optimizing them.

Amongst the A and B items you then want to identify items that have regular demand (i.e. demand in all or most time periods) that is not growing or declining markedly, and with moderate variability. (For the statisticians out there, variance coefficient is a good basis for carrying out an XYZ analysis on variability. Moderate variability would be X and Y items.)

A and B items with moderate variability are the best ones to start experimenting with. You can calculate optimal safety stock levels for these items and compare them to what you are using. What you are likely to find (based on our experience) is that you need to increase safety stock for a small number of items and can reduce it for a much larger number of items.

Before you rush in and change anything, however, there are a number of further steps to take. Maybe start with the items which show the biggest potential to change. Take the top 20 to 50 items by usage or potential, for instance. Then double check the input data to your calculations. Important parameters like lead time can frequently be wildly inaccurate (or even missing) in source data.

Once you’ve satisfied yourself that the data is as accurate as it can be, then try moving towards the suggested safety stocks for those items incrementally (i.e. bridge the gap between actual and target a little at a time, don’t jump straight to the calculated value). All models have a certain level of inherent inaccuracy, plus demand is never quite what you think it is going to be.

Once you have successfully tested this with your top items, gradually expand it to the rest of the stable ABXY items. Then you might want to consider more complex approaches for the rest of your inventories.

Doing this exercise may require you to try a replenishment model for the first time. Remember that deterministic planning in theory shouldn’t require safety stock. And in practise, if you’re using one of the deterministic MRP methods and planning to a forecast, then the safety stock calculations will not apply in the same way.

Safety stock and MRP

We sometimes see it said that replenishment isn’t even a planning method. Opinion can be very polarised on this topic. It really comes down to how predictable future demand is. With replenishment models, the way of thinking is “we don’t know what demand is going to be, but we expect it to carry on broadly speaking as it has been”. You then look at the variability in your demand and use that to set a re-order point which allows for enough safety stock to buffer against that variability. With deterministic models, the way of thinking is “we know that demand is going to be different from the past and we are going to plan for what we expect”.

What this means is that if you genuinely know future demand is going to be different to past demand (for instance, if you are planning for a promotion or a seasonal uplift, or if you are planning materials for a frozen production schedule) then deterministic planning should in theory be better than replenishment based on historical data. (Although of course, you can base your replenishment model on a forecast too, or temporarily add some anticipation stock to it.)

How do you choose between the two? Is your forecast consistently better than a naïve forecast? If not, you should see benefit from a replenishment model. What replenishment is good at is buffering the actual variability you have. However, because people are perhaps uncomfortable with probabilistic models, there is a strong perseverance of deterministic models, where you base your plan directly on what you expect. Variability can get lost in this model.

How often should you check and update safety stock levels?

As segmenting and calculating ideal safety stock levels is somewhat laborious to do well, the temptation is to do it as rarely as possible, and the good news is that this might not be a bad thing. One of the main issues with deterministic models that are not protected from variability is that they are “nervous”, i.e. they are constantly trying to readjust.

Part of the benefit of a replenishment model and the safety stocks they use is that variability is automatically absorbed by them. So does this mean you can set your re-order points and then head to the beach? Unfortunately not.

In reality, of course, change is a constant factor. The fact that variability was within a certain range in the data you used to define your safety stocks does not guarantee that it will remain in that range. While major events, like a delayed shipment, will no doubt come to your attention anyway, and be dealt with as part of the normal day-to-day inventory management work, a subtle but important change in demand patterns might escape you. Statistical process controls like the Western Electric rules are very useful where the volume of items managed is high.

And even setting sudden or substantial changes aside, most things are undergoing a gradual process of change. A lot will come down to your specific situation. If you have mostly mature products with limited change, once a year might be enough to check and update target safety stock levels. In a more dynamic environment, you will probably need to do it more often. If you find you need to constantly revise the levels, it is questionable whether you really need safety stock at all and some other approach, such as reserving spare capacity, might be preferable.

Where should you build safety stock?

The question also arises, where in your supply chain you should keep safety stock? This is partly a matter of business strategy (how quickly do you want to be able to fulfil customer demand) and constraints (how fast can you source materials and manufacture your products). As such, there is no “right” answer to this question. But as a generalisation for manufacturing industries, most keep some level of safety stock in finished goods since end demand is usually variable and make to order is not so prevalent. A replenishment model is frequently appropriate here.

At the other end of the supply chain, safety stock in raw materials can help protect you against delays or quality issues on the supply side. However, where there is considerable variability on the demand side and MRP has propagated it all the way through to raw materials orders, there is often a marked bullwhip effect and buying raw materials to a forecast that has negative forecast value added (FVA) leads to overstocks.

The most complex area for safety stock is WIP (understanding WIP here to include semi-finished goods) since it is heavily tied to capacity utilisation. In an ideal world, demand from the end customer would flow through and you would need no WIP except for the amount of time it takes to produce what is required. (And indeed, leading proponents of “Lean” manufacturing achieve impressively low WIP inventory levels.)

In reality, however, a lot of factories run at or close to capacity, which means there are frequent bottlenecks and waiting times for machinery to become available. Safety stock may usefully be built at relevant points in the manufacturing process such as at bottlenecks, at processes with very variable yields, or for semi-finished goods that are used for multiple end products. You will often find such safety stock in WIP referred to as decoupling stock (to distinguish it from the safety stock which is buffering against external uncertainties).

Such decoupling stock can be managed in a variety of ways, such as Kanban, CONWIP or DDMRP. It is a valid approach to define safety stocks at decoupling points based on the replenishment model, and the judicious positioning of replenishment points within an end-to-end production flow can mitigate some of the worst effects of MRP. A well-designed approach might build decoupling inventories with a replenishment model, optimizing safety stock for target service levels, and avoid safety stocks altogether for the parts of the process that are genuinely deterministic.

Are there any alternatives to safety stock?

While safety stock is perhaps the most common way to buffer variability, it is not the only alternative.

Another option is safety time. This is where you have quite a good idea of the size of demand, just not when exactly it will come. An example of this might be if you know that Production is planning to manufacture a fixed amount of a product, but the schedule hasn’t yet been fixed. Instead of getting additional raw materials, you just bring them in early.

Another option is reserving spare capacity and/or expediting products. Where you have a very limited ability to predict what demand is going to be, it might be preferable to be able to produce and despatch quickly in response to firm orders rather than building speculative safety stocks. An example for this might be for a new product launch where it is unclear how high demand will be.

A third option is waiting time. While in our modern world we have perhaps become used to having everything on demand and without delay, in certain circumstances it is possible just to make customers wait longer. This comes about through necessity in instances of major shortages and is also the option of choice in many instances of high value, low volume make-to-order products

Conclusions

Like many things in supply chain, a few simple safety stock principles can rapidly become extremely complex once you dig a little deeper and try to find optimum levels in the real world.

Supply chain professionals need to find solutions that are practical, proportionate and easy enough to understand to take people with them. However, safety stock is one of the most important levers you control and the benefits of getting it right, such as reducing production stoppages, late customer deliveries, or the need for expedited deliveries, are substantial.

There is perhaps a tendency to feel that having too much is better than having not enough, but having too much safety stock also has a real downside. Warehouses and production areas can get clogged up, production actually slows down because of the excess, working capital is stuck in unproductive surplus inventory and the risk of waste through obsolescence increases.

Many organisations are already taking advantage of various optimization models, either built in-house using spreadsheets or in tools from software vendors. This can deliver real benefits where it takes people away from cover targets. However, for some of the reasons explored in this article, a lot more is usually possible once you go beyond “one size fits all”.

Due to the complexity involved in optimizing safety stock, even companies already using advanced optimization technology usually have an opportunity to further improve their levels, reducing shortages and excesses simultaneously, often by 20% or more.

nVentic has automated the evaluation of inventory data to help clients identify and act on the improvement opportunity quickly and systematically. If you would like to discuss your safety stocks or other aspects of inventory optimization, please contact us.

Days Inventory Outstanding (DIO) is an interesting metric. At nVentic, we often use it as a conversation starter – a first outside-in look at how a company is doing in terms of inventory management. The fewer days’ inventory you have, the quicker your cash conversion cycle will be. As such, DIO is a useful component of a balanced supply chain scorecard. But what exactly is DIO and what does it really tell you about a company’s inventory levels?

What is DIO?

Days Inventory Outstanding is the value of inventory held divided by an average day’s cost of sales.

Let’s break that down.

Company A’s balance sheet shows an inventory value of €1 billion. The income statement shows cost of sales of €3.65 billion. This means that an average day’s cost of sales is €10 million (€3.65 billion / 365 days). Which in turn means that the €1 billion inventory is equivalent to 100 days’ cost of sales (€1 billion / €10 million).

If you like formulas, DIO = Inventory value / (Annual cost of sales / 365)

DIO of 100 implies that you have inventory to cover you for 100 days’ average activity, although this is misleading and we will return to what it is really telling you (and what it’s not) below.

The big advantage of DIO is that it is a very simple metric to calculate quickly based on publicly available information. Only privately held firms that do not report their figures, and a relatively small number of firms that do not break down their costs in a way which isolates cost of sales, prevent anyone from calculating DIO for any company. For more detail, and variants on the formula, see the technical notes at the end of this article.

What does DIO tell you?

DIO is a simple metric with which to compare companies, although this only really makes sense when comparing companies in the same or very similar industries. For instance, if you manufacture Scotch Whisky and you allow your product to mature for an average of 10 years before you sell it, there is little point in comparing your DIO with that of a business which buys and sells cut flowers.

If you compare the DIO of two companies in the same industry and with comparable products, then – all other things being equal – the one with a lower DIO is likely to be more efficient. That “all other things being equal” is an important point since differences in DIO can be driven by either structural or performance differences.

The structure of your supply chain is a major driver of DIO. If you have highly variable demand, make to stock and have very long supply lead times, you will need much more inventory than if you have stable demand, make to order and have short lead times. For instance, if your strategy is to source components from the other side of the world while your direct competitors source locally, you will need more inventory than them.

If there are no major structural differences at play, then comparing DIO between companies is indicative of efficiency, and having higher DIO than your main competitors is a strong hint that you have good potential to improve your inventory levels. Although it is also worth saying that, as very few companies are anywhere close to optimising their inventories, having a lower DIO than your competitors is not necessarily a reason for complacency.

What does DIO not tell you?

What DIO cannot do is tell you how much inventory you should have. Optimal inventory levels need to be calculated bottom up, based on the properties of each individual item that you stock. The optimal level will depend on the service level you are targeting, your lead times, the variability in your supply chain and a number of other factors.

If you can calculate your optimal inventory levels bottom up, then you could in principle turn that into a DIO target, although this is not so simple, since any change in inventory is also likely to involve a change in cost of sales. But if you are genuinely in a position where you regularly calculate how much inventory you should hold and compare it with actuals, then you have a much more operationally useful KPI than DIO in place anyway. At nVentic, we have automated the bottom-up calculation of optimum inventory levels.

DIO should not be seen as a proxy for service level. Intuition might seem to tell you that low DIO could indicate a higher risk of shortages, but there is limited data to support this view. Shortages are not caused by not having enough inventory overall, but by not having enough of the right inventory. By definition, if you have a positive DIO, you have not completely run out of inventory. Companies that manage their inventories well have lean inventories and high service levels simultaneously.

And finally, DIO does not actually tell you how much inventory cover you have. In our example, Company A had 100 days inventory outstanding. In a simplistic sense, this implies that they have enough inventory on hand to cover 100 days’ worth of sales. But, of course, they don’t, since DIO is just an average based on value. They actually have more or less than 100 days item by item – in a lot of cases much more or less.

This much probably seems obvious. There is no way you could actually hold exactly 100 days’ worth of sales in inventory unless you had completely uniform and predictable demand. But how much do you think the distribution of days’ inventory typically varies from the mean? Do you have something akin to a bell curve in mind? Let’s have a look at some actual examples of different types and sizes.

As you can see from the examples, which are typical for manufacturing industries, only a small percentage of items (less than a quarter) is within the two quartiles either side of the mean. The average DIO is therefore rather a poor indicator of how big an inventory buffer you actually have.

However, this pattern is not necessarily as bad as it might at first appear.

First of all, each item that you stock will have its own optimal level that can vary greatly in terms of the average number of days it should have on hand. For items that are quick to acquire or produce and have stable and predictable demand, you might only need to keep a day or two on hand at any one time. Whereas for items with high variability and long lead times, you might need to keep much more on stock. If you are trying to keep the same amount of stock on hand for all items – and this is what happens when you use “cover” targets – then you are actually working against inventory optimization.

The other thing you will notice from the examples above is how the graphs are skewed to the right. This is because the overall DIO figure is weighted by value, whereas the graphs are simply showing the number of items. Most organisations have a long tail of “C” articles, which are low volume and low value items. It is not uncommon to have well over a year’s stock for such C items, simply because minimum order quantities (MOQ’s) are greater than a year’s demand. This is hard to change and generally not worth the effort to try changing. If the aim is to reduce overall DIO, it is usually much better to focus on your high value/volume items.

Conclusions

DIO has a number of weaknesses. It is not a very precise metric, being open to a certain amount of manipulation (see the technical notes below). And it doesn’t, despite what its name might suggest, really indicate how well buffered you are against shortages, since the actual amount of inventory held by item is likely to vary so greatly from the mean.

And yet, DIO still has a useful role to play as part of a balanced supply chain scorecard. It is a valid indicator of how efficiently your supply chain uses cash. DIO may not represent how many days’ supply you have in inventory at all accurately, but it does represent how long your cash is tied up in inventory on average. And to this extent, it is a valid financial comparator when looking at competing organisations. It is a valuable lagging indicator of your supply chain efficiency. And should act as a spur to look at your inventories much more closely bottom up.

Technical notes on DIO:

You will sometimes see variations on the DIO formula we give above.

Some Finance departments calculate using 360 days in a year instead of 365.

The number of days in a year aside, the main valid alternative to the formula we use is to calculate inventory value by averaging the value of inventory on hand at the start and the end of the period in question. So, in our example above, we would sum the inventory value at the start and end of the year and divide the result by 2.

There are arguments for and against using this approach. If you have been growing very fast during the year, for instance, then your average cost of sales might be significantly below the cost of sales run rate at the end of the year and you could argue that averaging inventories is better, so that numerator and denominator are treated similarly. On the other hand, if your sales have been flat over the year but you have built up a significant amount of extra inventory, averaging inventory levels between the beginning and end of the year will understate the amount of inventory you’re holding at the end.

You will also sometimes see alternatives to the cost of goods sold being used. These alternatives typically fall into two camps:

1. Using sales instead of the cost of sales
2. Using a modification of cost of sales, for instance only using the cost of materials and overheads, or excluding inventory impairment charges from the cost of sales

The first of these has the one advantage that it is quick and simple – and not all organisations report cost of sales – but strictly speaking it is a dubious approach, since inventory on the balance sheet is valued at cost (or at resale value only if the value is less than cost – which you would hope to be the case only exceptionally!) By using cost of sales, you make sure you are aligning the denominator and the numerator in your DIO calculator.

The second alternative, using a modification of cost of sales, may be done for a variety of reasons that no doubt make sense to the organisations employing it. In any case, it is fair to say that cost of sales calculations leave a fair amount to any organisation’s discretion. Generally Accepted Accounting Principles (GAAP) have relatively little to say about cost of sales. The accounting methods used (FIFO, LIFO, average cost method) will often have a material impact on the cost of sales recorded. Plus, inventory valuation and the speed of writing off obsolete inventory can be managed to influence short term figures. Controlling for all such variables is not possible. We think it is best to accept that DIO is an approximate metric and work with the simplest metrics, requiring a minimum of analysis of financial statements.

A common alternative KPI to DIO is inventory turns, or inventory turnover, which is essentially measuring the same thing, just expressed differently.

Inventory turns = cost of sales / inventory value.

Going back to our example of Company A above, the cost of sales of €3.65 billion and inventory of €1 billion indicate an inventory turn of 3.65. So DIO of 100 is expressing the same thing as an inventory turn of 3.65. Noting that those who favour inventory turns as a measure tend to prefer using the average of start and end inventory since the focus is, as the name suggests, more on the movement of the inventory – the turns – rather than the amount of inventory – the days.