A Modern Tale of Long (Supply Chain) Tails -- Part III

Part II of this blog series explained ToolsGroup’s value proposition for achieving service level excellence in distribution environments. The point of the Service Optimizer 99+ (SO99+) suite's name is that a "99+ percentage" represents the gold standard in customer service levels, and it takes a product purposely built to achieve service level excellence and to support such a high standard.

ToolsGroup’s latest version of software continues to build on the functionality needed to reach this goal.

Back to Mitigating Long Tails

Having put the necessary pieces in place, over the past year ToolsGroup has turned a particularly keen eye toward how to succeed in environments with a “long tail” demand. The long tail theory originally held that in an environment such as the Internet-based retailers (so called “e-tailers”), which is less affected by physical manufacturing, product and distribution constraints than so-called “brick-and-mortar” retailers, demand will spread across a huge array of items (a.k.a., stock-keeping units or SKUs).

If demand for each SKU is illustrated graphically, starting with most popular items first, as demand wanes toward a zero point the graph depicts a “long tail.” The long tail theory was originally coined by Chris Anderson, editor of Wired magazine , who illustrated the concept with Internet-based selling companies such as Apple's iTunes or Amazon.com. His book entitled “The Long Tail: Why the Future of Business is Selling Less of More”  has also been a bestseller.

In his part of the Webcast mentioned in Part I, Joseph Shamir, ToolsGroup’s chief executive officer (CEO), ascertained that when the long tail is viewed from a supply chain perspective, the tail starts where demand becomes “intermittent.” Using this definition, ToolsGroup’s research shows that most brick-and-mortar companies (and not only e-tailers, as one might think) are also facing an increasingly long tail environment. This is because they are dealing with far more SKUs and slow-moving items with lumpy, unpredictable demand patterns.

This creates difficulties for these companies since, as AMR Research’s analyst Lora Cecere concludes in her report mentioned in Part I of this blog series, entitled “Of Long Tails and Supply Chains”:
“The deterministic, replenishment logic traditionally found in advanced planning and scheduling (APS) technologies are not a good fit for the long end of the tail… Bottom line, in this scenario, traditional inventory techniques—safety stock logic based on normal demand distribution — just don’t work.”

Analyzing the “long tail” from a supply chain perspective, the tail is typically longer and bigger than from the sales and marketing perspective. From a supply chain perspective, what is the problem of being in the tail? Well, for one, inventory mixes can be largely wrong: i.e., some products are unnecessarily over-served while others are underserved.

With lumpy demand, traditional forecasting methods of judging sales history are not good for predicting the future, since even forecasts with “high accuracy” are invariably wrong, due to high demand variability and highly skewed demand distribution.

Another unfortunate consequence of long tails is the so-called “bullwhip effect”, whereby on stock-outs, planners often over-react (as is human nature) and thus create overstocks and misaligned inventory. The inventory mix (and therefore the service levels) across the supply network will thus not reflect business objectives.

Although long tails and the bullwhip effect are broad independent concepts, they become related when traditional inventory management systems, which  do not correctly model and understand long tail demand,  order the wrong inventory to address that demand. To compensate for not having the right inventory, planners will expedite and over-order to plug the gaps and meet their customer’s needs. This overreaction is one of many causes of the bullwhip effect, and as a result, the working capital is shifted from the active to slow-moving stocks.

This negative use of capital is because companies compensate by manually adding lots of unneeded slow-moving inventory, which accumulates in the "tail", whereas the "head" (with fast-moving items) is underserved. And without good statistical demand and inventory models, companies will not have the right inventory mix for the long tail, especially if they still try to leverage the inept deterministic, replenishment logic traditionally found in APS and enterprise resource planning (ERP) systems.

The ironic and unfortunate business impact is that the tail consumes a lot of working capital, without delivering the desired service levels. The situation creates a significant opportunity to improve both the top and bottom line.

Namely, poor service levels, with up to twice as many stock-outs (mostly within fast-moving items) seriously affect revenues, while excessive expediting is quite expensive (besides lowering employee morale and raising their blood pressure). As for the balance sheet aspect, inventories can be bloated by as much as 50 percent and negatively affect cash reserves.

In a recent whitepaper entitled “Mastering the Long Tail of Demand” the ToolsGroup co-founders describe the growing long tail environment in brick-and-mortar companies and agree with Cecere that classic demand and inventory models do not perform well in this “tail” environment. They believe, however, that it is possible to make the most of and even succeed in this adverse environment.

ToolsGroup is trying to fill this gap where traditional inventory management approaches are not succeeding. The authors say that:
“By taking no shortcuts and mastering the full probability distributions of both demand and inventory across a wide range of possible behaviors, you can reach unprecedented efficiency and service level excellence in an increasingly challenging “long tail” world.”

How Does All This Come Together within ToolsGroup?

In a nutshell, the Problem would be Long Tails, a Solution could be the Inventory Mix Optimization capability, with the consequent Benefit of 99+ Percent Service Levels. Most companies have to deal with long tail demand. The discussion thus far shows that even for most brick-and-mortar companies, at least half of their SKUs are in the long tail, that is, where demand is intermittent and lumpy for those items.

Conventional ERP and supply chain systems were designed for the head of the demand curve, and they simply don’t work well in the tail of the demand curve. Conversely, inventory mix optimization and demand modeling are the astute technologies ToolsGroup offers to deal with long tail demand. Without them, most companies cannot achieve high service levels, because they cannot achieve an optimal mix of inventory.

In contrast to my coffee consumption-caused tale (that parents use as a deterrent to curious kids) at the beginning of this blog post series, the growing tail is a reality in supply chains. Fighting it becomes not an option, but rather a must for all businesses, even for fast-moving consumer products, and especially specialty retailers.

To win in the long tail situation, affected companies need more accurate demand and inventory models to support reliable service levels and inventory management. Also, they need highly disciplined processes to reduce bullwhip behavior.

Calling Spade a Spade

Now, my intent here was not to portray ToolsGroup as the best thing since sliced bread, since it is not a universally applicable inventory optimization (IO) solution. For instance, unlike ILOG's LogicTools [evaluate this product], SmartOps or Optiant, it does not offer full capabilities for complex Bills of Material (BOMs) (for instance, the items with multiple routings). Also, the solution is not suitable for pure “make-to-order (MTO)” manufacturing environments that do not carry finished goods. Finally, smaller finished goods distribution companies with less than US$100 million in revenues should also consider some simpler and cheaper products.

However, for any mid-size to large forecast-focused distribution company with a large number of SKUs, long tails and the need for service level excellence, ToolsGroup is an inventory optimization vendor that is differentiated in two ways. First, unlike some vendors who excel at providing a strategic planning tool, ToolsGroup offers a full-fledged operational solution that bolts onto an ERP or supply chain management (SCM) system, and automatically and self-adaptively generates optimized safety stocks and other inventory targets on an reoccurring (e.g., weekly or daily) basis.

Second, ToolsGroup offers a solution that is focused on the demand part of the supply chain, hence providing the right solution for companies with a significant amount of finished goods inventory and those trying to achieve very high (e.g. 99+ percent) customer service levels. The vendor offers forecast enhancement to precisely and statistically profile demand curves, even at the “tail” ends.

In addition, one cannot overemphasize the “inventory mix optimization” feature, where the system automatically adjusts service level targets for individual SKUs to obtain the optimal “mix” of inventory, and hence attain the target global service objective while deploying less inventory.

Competitors often put the ToolsGroup products down for looking like a traditional desktop application and for not having a snazzy and intuitive user interface (UI). The first issue might be moot by now given the recent rewrite on the Microsoft .NET Framework technology.

As for the UI metaphor, ToolsGroup claims to not compete in a beauty contest, but rather in saving its customers money in terms of inventory investment.  After all, the product is designed to run mostly automatically, requiring only occasional input from a user that is a seasoned planner with good understanding of both inventory management and what takes place “under the hood.”

In its “sweet spot” environments, ToolsGroup touts relatively low total cost of ownership (TCO) and fast time to benefit, since its solutions do not require customization and all transaction data is in ERP systems anyway. The product is also modular to enable faster implementation and quick results, while single-tenant on-demand hosting deployment could offer even faster time-to-benefit.

Generally speaking, all companies that are affected with long tails and are pondering selecting and deploying an IO solution should first find good answers to the following questions (issues):

  • How long is the tail for our business?

  • Are traditional inventory techniques working well for our supply chain?

  • Does our mix of inventory and service reflect our business objectives?

  • What kind of service levels, inventory turns and expediting is our supply chain generating?; and

  • What are the business impacts and compelling action items?

What are your thoughts in this regard, and your experiences with the solutions like ToolsGroups' DPM (Demand Planning Model) and SO99+?
comments powered by Disqus