Last week we looked at cutting inventory through improving inventory quality (getting rid of obsolete or low value stock) and reclassifying mislabelled inventory. Once you’ve done those those 2 quick fixes, the next inventory cutting initiative is testing for forecast bias and reviewing products in transition. Like last week’s Quick Hits, these will yield fast results without much in the way of research, tools, or infrastructure changes. Removing forecast bias should yield measurable inventory reduction results in just two to three months.

Testing for Forecast Bias

Forecast bias is a problem in many organizations. The perpetual challenge in demand planning is to balance the aspirations of Sales and Marketing with reality as reflected in consumer sales and factory shipment trends. When aspirations consistently exceed the reality, bias is the result.

Supply chain planning typically begins with a forecast, and if it is overstated it creates excess inventory. Conversely, consistently understating a forecast (or “sandbagging”) causes a bias as well. This can lead to expediting, schedule cuts, and other detrimental supply chain behaviors. While targeting bias seems to be the realm of a longer-term forecast improvement project, measuring forecast bias often drives fast and significant change in both inventory reduction and forecast error levels.

Measuring bias requires only a modest effort, usually focused on report building, and can be implemented in a relatively short period of time. It’s a perfect quick-hit opportunity. For those not already familiar with measuring forecast bias, there are a number of different ways to calculate this metric; our experience suggests that simple rules are the best. One quick test is to track any item that has been over or under forecasted for three months in a row (in the same direction).

A directionally biased item becomes a ‘problem item’ if the average deviation or error over the same three months is greater than 25%.

Identifying Problem Items

If this is your first venture into measuring bias, you should expect 60%-70% of all items to have some level of directional bias. Of course, it may be impossible to track bias for a potentially large number of items, so you may want to manage the bias measure by exception, using a secondary filter to reduce the list of bias items to those where the directional bias is most significant. The typical secondary filter is aggregate error – the summing of forecasts and actuals to arrive at an error percentage over the course of three months. In one of the simplest implementations, a directionally biased item becomes a ‘problem item’ if the average deviation or error over the same three months is greater than 25%. Again, this is a simple rule of thumb that may not apply well to all businesses.

As you might expect, the number or percentage of biased items is a meaningful S&OP measure. Figure 1 shows the number of biased items (problem items) above a 25% error threshold on a two month lagged forecast with three months of consecutive directional error. Making biased items a key measurable in the forecasting process improves the quality of a forecast. In turn, a better quality forecast is more likely to help drive down inventory levels.

Once you work through some of these questions, you may present the bias information in your demand consensus process for root cause analysis and correction

Figure 2 shows how the total inventory level (and specifically the finished goods inventory) dropped significantly over the same time as the number of bias items was lowered. This works, and it works well. So, how do you get started? First, you need to create a bias measure that works for your company. Looking at historical data can help you determine the number of directional forecast misses over time. You will need to determine a relevant level at which to test for bias in an item, brand, product grouping, or category, and/or what percentage of a miss represents bias for your organization. Most business leaders start with 25% as a measure, and then shift the percentage up or down depending on what they think is best for their organization. Once you work through some of these questions, you may present the bias information in your demand consensus process for root cause analysis and correction. Implementing a simple test for forecast bias can pinpoint unmet aspirations within a forecast and help you quickly reduce inventory levels within 2 to 4 months. Again, this is one of those cases where measurement can help change business behaviors.

To actually reduce the inventory, the best strategy is to create a report of biased items, review them in a demand consensus meeting, and then report them in the Executive S&OP meeting for an appropriate action. Clearly explain that the data reveals bias and is causing excess inventory and a drag on profitability, using the hard data in the report to support your argument.

All too often, product portfolio decisions are made without facts or are based on simplistic rules or metrics such as annual revenues.

Keep an Eye on Products in Transition

As items progress through the normal product life cycle of introduction, growth, maturity, and decline, there is a need to constantly evaluate how these products are planned. Each phase brings its own expectations. A product at launch needs a great deal of inventory and flexible capacity because of the uncertainty of demand. Items at maturity or in decline need their inventory to be optimized as demand is much more certain and reserve stock is not needed. Inventory is at greatest risk during the early introduction phase and in the later phases of decline— often called the long tail of the product life cycle—where sales volumes have dropped so low as to be considered for discontinuation, either through a voluntary market withdrawal or by the loss of a key customer or two.

Product Rationalization To Cut Low Value Items

There are two challenges in managing inventory throughout the long tail. The first is correctly determining product rationalization, that is, deciding which products should stay in your portfolio and which should be discontinued. The second challenge is managing the exit of those products slated for removal. In some industries, products move through life cycle phases seemingly in the blink of an eye. Yet even in industries that specialize in introducing new and improved products each year, it is not uncommon to find poorly defined processes for anticipating/ determining and managing product discontinuation. Likewise, the processes for managing planning parameters that determine inventory utilization through the end of the product life cycle are often equally deficient.

Portfolio management meetings address items as they transition through all phases of the product life cycle—from the ideation to discontinuation

Portfolio Management Review Meeting Should Control Excess Inventory

To meet these challenges, some S&OP process models include a step called the portfolio management review meeting. Within these meetings, the product portfolio is discussed in detail, including discussion regarding current ideas for products as well as those that are in the launch phase. If your firm’s S&OP process doesn’t include such a meeting, you should consider instituting one. Portfolio management meetings address items as they transition through all phases of the product life cycle—from the ideation to discontinuation. A monthly dialogue on the product portfolio can be an enormous leap forward in providing a forum for tracking product strategy and execution. Within portfolio management review meetings, many firms are beginning to use detailed analytics to properly target items for deletion off the long tail. All too often, product portfolio decisions are made without facts or are based on simplistic rules or metrics such as annual revenues. Using a rigorous approach to identify products for discontinuation on an ongoing basis enables the company to shorten the supply while facilitating effective management of obsolete inventory. Once a product is identified for potential discontinuation, supply planning assumptions such as fill rates and batch sizes can be tuned downward to reduce raw and packaging inventory exposure and risk during the final run-out.

Reviewing Stock Levels For New Products Is Crucial

Managing the supply planning parameters during the course of a product life cycle is particularly important. As an example, at one consumer products firm it was commonplace for supply chain planning to carry an additional 35 days’ worth of forward coverage inventory for all new product introductions after the initial pipe fill. The company intentionally carried excess inventory at product launch because of the uncertainty about demand early on. The company was more concerned about having available inventory to support a successful launch rather than the holding cost.

As you might expect, there were multiple examples where this supplemental-coverage inventory was never adjusted downward during the post-launch because there was no process in place to routinely review planning parameters. Under normal circumstances inventory targets are reduced to standard, steady state coverage levels, but with personnel changes and focus shifting toward the next new product launch life cycle, these planning parameters are overlooked, resulting in excess inventory.

Keeping An Eye on Products in Transition requires at least a quarterly review of planning parameters for all products less than two years past the launch phase while also conducting a similar examination of those items nearing the end of their run. Special attention should be reserved for products on the long tail of their life cycles, since these items need focused corrections in their planning parameters as their demand recedes. The adjustments in these parameters should be made with a goal of limiting inventory exposure in the event the market collapses for a product or if key customers drop the product completely.

This is where a portfolio management review meeting can add tremendous value in anticipating and managing slow and no-turn inventories. Many companies review each and every planning parameter, at least once a year. This should be the minimum review. If your company does this, great! The quick hit opportunity in this case is a very detailed one time review of all inventory policy settings for both finished goods and components.

These two initiatives are quick and highly effective. Despite minimal effort, they can resolve systemic problems that place cost burdens on your company. Next week, we’ll be looking at Waste and Yield Percentages and QC Hold Time. Stay tuned!

—————————

This article has been adapted from the Journal of Business Forecasting and Planning, Fall 2011 issue. To receive the Journal of Business Forecasting and other benefits, become an IBF member today.

In the first of this mini-series, Pat Bower of Combe Inc. reveals the quickest and easiest ways to reduce inventory. Detailing the preferred methods used by leading S&OP professionals and consultants, he discusses how to cut poor quality or obsolete inventory for significant cost savings. Read on for quick improvements that can be leveraged easily in any supply chain focused organization. This week Pat discusses the two easiest inventory reduction initiatives: improving inventory quality and inventory classification.

Identifying Easy Targets For Inventory Reduction

As a consultant hired looking to improve supply chain efficiency in many different businesses, I would look for inventory reduction opportunities that could be implemented without much in the way of research, tool, or infrastructure changes, with the intent of yielding measurable inventory reduction results within a two-to three-month time frame. I call these Inventory Quick Hits. But where do these opportunities lie? In most cases, Quick Hits are a subset of the below inventory reduction solutions, but are narrower in scope and focus:

1. Improve inventory quality by reclassifying R&D inventory and removing junk inventory.
2. Testing and get control over forecast bias.
3. Keep an eye on products in transition so you can manage supply chain parameters reflective of their phase in the product life cycle.
4. Examine waste and yield percentages as well as QC (Quality Control) hold time.
5. Examine “actual” buy-side tolerances.
6. Investigate size, component, and formula consolidation opportunities. Look for easy-to-implement product and component.

Quick Hit 1: Improve Inventory Quality

Inventory has many classifications: active, slow-moving, obsolete, and inactive or run-out mode. Inventory that turns quickly and is actively sold is thought to be of high quality, while inventory that is slow moving, has no turn over, is obsolete, or residual is considered low quality. This “junk” or low-quality inventory is a favorite target of inventory consultants trying to reduce inventory. Slow-moving inventory has low volume sales or consumption relative to on-hand inventory. It simply does not turn over that fast or frequently. There are countless potential reasons why inventory is slow moving.

It’s possible that inventory turns slowly because of batch sizing, where each production run makes a large portion of an annual forecast, or because the product’s demand is slowly eroding, or for other reasons. Reducing slow-moving inventory requires a reassessment of supply chain parameters such as batch sizes, EOQs (Economic Order Quantity), and the like.

By contrast, obsolete inventory has no demand, and is often a combination of residual finished goods and leftover components. Obsolete inventory has lost the primary customer base, and the only sales opportunities it has are typically in distress or closeout channels. Run-out inventory is typically a combination of finished goods and components that are still being sold, but clearly are at the end of the product life cycle. Inventory is often placed in a run-out mode to help reduce the final inventory liability and may be kept as an active finished good, even with a greatly reduced customer base as a placeholder for a new product introduction. Run-out inventory usually requires a greater understanding of what the total inventory liabilities are as well as the possible disposal options. Inactive inventory generally has no raw material or packaging consumption for some period of time, typically six months to a year. It is not uncommon for ongoing finished good sales to be maintained for a variety of reasons, including trying to exhaust raw and packaging liabilities, or to cater to a niche market, or a special customer.

Analyse Inventory Holding Costs Vs. Potential Liability Savings

Eventually, when the finished goods or components are exhausted, the product ceases to exist, and the raw and pack components will be inactivated. In one company, for example, we observed there was a considerable inactive inventory for components sold into foreign affiliate markets. Despite the fact that components were ordered in the smallest possible increment, a single production run yielded enough finished goods inventory to cover demand for a couple of years. These unused components sat idle for a number of years waiting for the next production run. An analysis of inventory holding costs vs. potential liability savings are typically a first step in determining whether the inventory should be held. Most planning systems (ERP or SCM) have some variation of reporting tools for slow or not-moving, obsolete and excess inventory (often referred to as SLOB) that planners can run to examine any inventory that has not been used in the last 12 months or more, or that has exceeded coverage tolerances. Such reporting is very helpful in targeting inventory that falls into the junk classification, yet in our experience it’s often under-utilized. The quick-hit opportunity for inventory reduction is really focused on becoming very familiar with this reporting.

Understand Why Inventory Is Not Turning

Certainly, finished goods inventories should be examined, but the real opportunity is usually with the raw, pack, and component inventory left unused for 12 months or with excess coverage. There are many reasons why this excess inventory exists, including discontinuations, poor EOQs, packaging changes, and formula changes. The real magic in this suggestion is not to simply toss out the old inventory, but to get to the root cause of why inventory is not turning, and see if there are ways to correct the problem from a planning perspective. To get a better grasp of this “junk” and to create a unified metric for inventory quality, you may want to use tools that provide an Inventory Quality Ratio, (IQR) in which active inventory (net of excess, slow-moving, inactive, or obsolete) is divided by the total inventory.

In a newly developed S&OP process, the inventory quality ratio (IQR) can be an important metric used to track overall inventory values. Some companies have a very high percentage of slow-moving, inactive, or obsolete inventory. When I was consulting, it was not uncommon to see clients with low-quality inventory in the range of 25% to 50%. Periodic reviews and measurement of inventory quality, followed by purging the warehouses and plants of useless products, is among the simplest ways to reduce inventory.

Quick Hit 2: Reclassify R&D Inventory

This is a bit of a variation on the “cleanup of junk inventory” theme. Inventory should be properly classified from an accounting perspective and yet it is not uncommon for companies to accidently place inventory used for research and development purposes into an active (everyday use ) status. For example, a company that makes mixes, compounds, or blends chemistry may have a slew of random R&D chemicals, materials, or components that have been mistakenly classified as active production inventory. The list of companies with misclassified inventory includes more than just big industrial/commodity chemical companies— OTC pharma, cosmetics firms, health, beauty, and even food manufacturers blend chemistry. How does inventory get misclassified?

Cut 2% Of Misclassified Inventory Right Away

Sometimes inventory is ordered using the wrong accounting code, and sometimes-unused production inventory (mostly chemical compounds) of obsolete products are saved for a future R&D use. Either way, this inventory is often sitting on the books in an active status yet has no consumption or use against it, historically or planned. A full review of inventory may reveal a number of these odd ducks; one such review at a client revealed that over half of the inventory saved for future R&D use was expired and unusable, even for research purposes. While the percentage of this type of misclassified inventory is relatively small (usually within 1%- 2%), experience has taught us that most of it serves little useful purpose and occupies valuable storage space, making it a relatively easy target for reducing your total inventory tally. A full review of all active items with little to no consumption over the last year or more is likely to turn up in some of these stray inventory exceptions.

These two quick hits are easy targets for inventory reduction, and should be your first port of call when attempting to streamline inventory. They often provide significant cost savings in a relatively short amount of time, and with relatively little effort. Next weeks Quick Hits are Testing for Forecast Bias and Keeping An Eye On Product Transition. Stay tuned!

This post has been adapted from an article that appeared in The Journal of Business Forecasting, Fall 2011 issue. To receive the Journal of Business Forecasting, and wide range of other benefits, become an IBF member today.

Do you want to write for www.demand-planning.com? Then submit your article ideas and the Editor will get back to you.