Probabilistic Forecasting for Intermittent Demand

The Smart Forecaster

Pursuing best practices in demand planning, forecasting and inventory optimization

Probabilistic

Forecasting

Technology

Intermittent, lumpy or uneven demand —particularly for low-demand items like service and spare parts — is especially difficult to predict with any accuracy. Smart Software’s proprietary probabilistic forecasting dramatically improves service level accuracy.  If any of these scenarios apply to your company then probabilistic forecasting will help improve your bottom line. 

  • Do you have intermittent or lumpy demand with large, infrequent spikes that are many times the average demand?
  • Is it hard to obtain business information about when demand is likely to spike again?  
  • Do you miss out on business opportunities because you can’t accurately forecast demand and estimate inventory requirements for certain unpredictable products?
  • Are you required to hold inventory on many items even if they are infrequently demanded in order to differentiate vs. the competition by providing high service levels? 
  • Do you have to make unnecessarily large investments in inventory to cover unexpected orders and materials requirements?
  • Do you have to deliver to customers right away despite long supplier lead times?  

If you’ve answered yes to some or all of the questions above, you aren’t alone. Intermittent demand —also known as irregular, sporadic, lumpy, or slow-moving demand — affects industries of all types and sizes: capital goods and equipment sectors, automotive, aviation, public transit, industrial tools, specialty chemicals, utilities and high tech, to name just a few. And it makes demand forecasting and planning extremely difficult. It can be much more than a headache; it can be a multi-million-dollar problem, especially for MRO businesses and others who manage and distribute spare and service parts.

Identifying intermittent demand data isn’t hard. It typically contains a large percentage of zero Save & Exit values, with non-zero values mixed in randomly. But few forecasting solutions have yielded satisfactory results even in this era of Big Data Analysis, Predictive Analytics, Machine Learning, and Artificial Intelligence.

 

DOWNLOAD THE ARTICLE

 

Traditional Approaches and their Reliance on an Assumed Demand Distribution

Traditional statistical forecasting methods, like exponential smoothing and moving averages, work well when product demand data is normal, or smooth, but it doesn’t give accurate results with intermittent data. Many automated forecasting tools fail because they work by identifying patterns in demand history data, such as trend and seasonality. But with intermittent demand data, patterns are especially difficult to recognize. These methods also tend to ignore the special role of zero values in analyzing and forecasting demand.Even so, some conventional statistical forecasting methods can produce credible forecasts of the average demand per period.  However, when demand is intermittent, a forecast of the average demand is not nearly sufficient for inventory planning.  Accurate estimates of the entire distribution (i.e., complete set) of all possible lead-time demand values is needed. Without this, these methods produce misleading inputs to inventory control models — with costly consequences.  

Collague with gears ans statistical forecast modeling

 

To produce reorder points, order-up-to levels, and safety stocks for inventory planning, many forecasting approaches rely on assumptions about the demand and lead time distribution.  Some assume that the probability distribution of total demand for a particular product item over a lead time (lead-time demand) will resemble a normal, classic bell-shaped curve. Other approaches might rely on a Poisson distribution or some other textbook distribution.  With intermittent demand, a one-sized fits all approach is problematic because the actual distribution will often not match the assumed distribution.  When this occurs, estimates of the buffer stock will be wrong.  This is especially the case when managing spare parts (Table 1).  

For each intermittently demanded item, the importance of having an accurate forecast of the entire distribution of all possible lead time demand values — not just one number representing the average or most likely demand per period — cannot be overstated. These forecasts are key inputs to the inventory control models that recommend correct procedures for the timing and size of replenishment orders (reorder points and order quantities). They are particularly essential in spare parts environments, where they are needed to accurately estimate customer service level inventory requirements (e.g., a 95 or 99 percent likelihood of not stocking out of an item) for satisfying total demand over a lead time.  Inventory planning departments must be confident that when they target a desired service level that they will achieve that target.  If the forecasting model consistently yields a different service level than targeted, inventory will be mismanaged and confidence in the system will erode.

Faced with this challenge, many organizations rely on applying rule of thumb based approaches to determine stocking levels or will apply judgmental adjustments to their statistical forecasts, which they hope will more accurately predict future activity based on past business experience. But there are several problems with these approaches, as well.

Rule of thumb approaches ignore variability in demand and lead time. They also do not update for changes in demand patterns and don’t provide critical trade-off information about the relationship between service levels and inventory costs.

Judgmental forecasting is not feasible when dealing with large numbers (thousands and tens of thousands) of items. Furthermore, most judgmental forecasts provide a single-number estimate instead of a forecast of the full distribution of lead-time demand values. Finally, it is easy to inadvertently but incorrectly predict a downward (or upward) trend in demand, based on expectations, resulting in understocking (or over-stocking) inventory.

 

How does Probabilistic Demand Forecasting Work in Practice?

Although the full architecture of this technology includes additional proprietary features, a simple example of the approach demonstrates the usefulness of the technique. See Table 1.

intermittently demanded product items spreedsheet

Table 1. Monthly demand values for a service part item.

The 24 monthly demand values for a service part itemare typical of intermittent demand. Let’s say you need forecasts of total demand for this item over the next three months because your parts supplier needs three months to fill an order to replenish inventory. The probabilistic approach is to sample from the 24 monthly values, with replacement, three times, creating a scenario of total demand over the three-month lead time.

How does the new method of forecasting intermittent demand work

Figure 1. The results of 25,000 scenarios.

 

You might randomly select months 6, 12 and 4, which gives you demand values of 0, 6 and 3, respectively, for a total lead-time demand (in units) of 0 + 6 + 3 = 9. You then repeat this process, perhaps randomly selecting months 19, 8 and 14, which gives a lead-time demand of 0 + 32 + 0 = 32 units. Continuing this process, you can build a statistically rigorous picture of the entire distribution of possible lead-time demand values for this item. Figure 1 shows the results of 25,000 such scenarios, indicating (in this example) that the most likely value for lead-time demand is zero but that lead-time demand could be as great as 70 or more units. It also reflects the real-life possibility that nonzero demand values for the part item occurring in the future could differ from those that have occurred in the past.

With the high-speed computational resources available in the cloud today, probabilistic forecasting methods can provide fast and realistic forecasts of total lead-time demand for thousands or tens of thousands of intermittently demanded product items. These forecasts can then be entered directly into inventory control models to insure that enough inventory is available to satisfy customer demand. This also ensures that no more inventory than necessary is maintained, minimizing costs.

 

A Field Proven Method That Works

Customers that have implemented the technology have found that it increases customer service level accuracy and significantly reduces inventory costs.

Warehouse or storage getting inventory optimization

A nationwide hardware retailer’s warehousing operation forecasted inventory requirements for 12,000 intermittently demanded SKUs at 95 and 99 percent service levels. The forecast results were almost 100 percent accurate. At the 95 percent service level, 95.23 percent of the items did not stock out (95 percent would have been perfect). At the 99 percent service level, 98.66 percent of the items did not stock out (99 percent would have been perfect).

The aircraft maintenance operation of a global company got similar service level forecasting results with 6,000 SKUs. Potential annual savings in inventory carrying costs were estimated at $3 million. The aftermarket business unit of an automotive industry supplier, two-thirds of whose 7,000 SKUs demonstrate highly intermittent demand, also projected $3 million in annual cost savings.

That the challenge of forecasting intermittent product demand has indeed been met is good news for manufacturers, distributors, and spare parts/MRO businesses.  With cloud computing, Smart Software’s field-proven probabilistic method is now accessible to the non-statistician and can be applied at scale to tens of thousands of parts.  Demand data that was once un-forecastable no longer poses an obstacle to achieving the highest customer service levels with the lowest possible investment in inventory.

 

Hand placing pieces to build an arrow

DOWNLOAD THE ARTICLE

 

Leave a Comment

Related Posts

Backing into Safety Stock is the Safe Play

Backing into Safety Stock is the Safe Play

Safety stock is a critical component in any system of inventory management. Indeed, some inventory software treats safety stock as the key decision variable in the quest to balance inventory cost against item availability. Unfortunately, that approach is not the best way to strike the balance.

Probabilistic Forecasting for Intermittent Demand

Probabilistic Forecasting for Intermittent Demand

The New Forecasting Technology derives from Probabilistic Forecasting, a statistical method that accurately forecasts both average product demand per period and customer service level inventory requirements.

How to Choose a Target Service Level to Optimize Inventory

How to Choose a Target Service Level to Optimize Inventory

When setting a target service level, make sure to take into account factors like current service levels, replenishment lead times, cost constraints, the pain inflicted by shortages on you and your customers, and your competitive position.

Recent Posts

  • Manager In Warehouse With Clipboard and Safety stock controlBacking into Safety Stock is the Safe Play
    Safety stock is a critical component in any system of inventory management. Indeed, some inventory software treats safety stock as the key decision variable in the quest to balance inventory cost against item availability. Unfortunately, that approach is not the best way to strike the balance. […]
  • Hand placing pieces to build an arrowProbabilistic Forecasting for Intermittent Demand
    The New Forecasting Technology derives from Probabilistic Forecasting, a statistical method that accurately forecasts both average product demand per period and customer service level inventory requirements. […]
6 Essential Steps to Better Recovery Planning

The Smart Forecaster

Pursuing best practices in demand planning, forecasting and inventory optimization

As we approach the midpoint in 2013, there is still a lot of economic uncertainty complicating your supply chain planning processes. Some look at this shaky economy and postpone needed investments that can position their organizations for a strong future.

However, this is not the time to retreat from your supply chain improvement initiatives. Rather, it’s a time to double-down on your efforts to prepare for the inevitable business opportunities that lie ahead.

Economic recovery is a time of sales opportunities. You want to make sure that you’re prepared to take advantage of them. Good demand and inventory planning can help.With the right software and planning processes, you can achieve a sound statistical basis for decision-making going forward while making informed adjustments as circumstances dictate. You can improve your ability to read demand signals, spot trends, model future events, and bring your inventory into balance with demand.

Here are six areas of demand and inventory planning where changes you make now can lead to big payoffs when new opportunities arise:

1. Optimize your inventories

When the customer calls, you want to be able to ship. At the same time, you want to control your costs. The surest way to meet that goal is to find the inventory “sweet spot.” That’s where you have the minimum amount of inventory required to satisfy product demand over a specified lead time and at a desired service level.

The ability to accurately set safety stock and inventory levels can set you apart from the competition, and make a difference in your bottom line. However, getting to that point requires a shift in your planning focus from just forecasting future demand to optimizing stocking levels to fill future orders.

If you’d like to know more about achieving the “sweet spot,” you can find a good article published in APICS Magazine here.

2. Implement intermittent demand forecasting solutions

Companies in the service parts, auto aftermarket, and capital goods industries commonly experience intermittent, “slow moving” demand for a large percentage of their inventory items. Accurately forecasting demand and estimating safety stock levels for these types of items is probably the toughest challenge demand planners face. If you can accurately forecast your intermittently demanded parts and products, and have the correct amount of inventory and safety stock on the shelf, you’ve got most of the competition beat!

The reason for this is that items that have intermittent demand do not have normal demand patterns or distributions, making them difficult to forecast using traditional forecasting methods (see the diagram below).

Bar chart illustrating intermittent demand

So, if you have an accurate means of forecasting intermittent demand and estimating safety stock requirements, you’ll be ahead of your competitors that don’t.

If you’d like to know more about forecasting and planning items with intermittent demand, you can find an informative white paper here.

3. Improve lead times

The economic downturn has forced companies to rethink their sourcing strategies because of uncertain demand back home, long lead times to obtain their goods, rising labor costs abroad, and increasing transportation costs. Shortening replenishment lead times can reduce the time required to get the products you need and helps make your supply chain more efficient. It also makes it easier to react to changes in demand when recovery comes.

4. Prioritize service levels

Prioritizing service levels for your products can help insure that the items important to your sales are given the attention they need. For items that are highly demanded, consider setting service levels higher than for those with less demand. Also try doing a revenue-based ABC analysis of your company’s stock-keeping units (SKUs) and set service levels accordingly in your software planning solution.

For example, you might set the service levels for your “bread and butter” items at 95-99% or higher, while setting service levels much lower (at 70-80% or even less) for other items. In this way, you may find that you need much less stock for some of your SKUs and more stock for others to effectively achieve your overall service level goals.

5. Use more recent demand history in creating your forecasts

Because the economy has been changing so fast, it may be time to shorten the demand history used in generating your forecasts so more emphasis is placed on recent trends and demand patterns—reflecting the “new normal”—rather than those contained in outdated history from 3 or 4 years ago. This, of course, should be done in consultation with your management team and preferably as part of an organized S&OP process that thoroughly evaluates both the risks and benefits of adopting this strategy.

6. Invest in technologies and resources that help you capitalize on opportunities

Investing in the right tools and processes increases your competitive advantage. If you aren’t doing so already, here are some valuable things to consider:

• Start an S&OP process, or fine tune your current process, to include key stakeholders in the supply chain and also ensure that demand forecasting and inventory planning provide key inputs in that planning process.

• If your forecasting software is not good at picking up trends, or cannot handle the portion of your inventory with intermittent demand, find software that’s up to the task.

• Find software that will take your forecast results and generate accurate inventory stocking levels to satisfy demand for your products, components or raw materials over specified lead times and at service levels you desire.

• Look for software solutions that are scalable, yet have a relatively low total cost of ownership, fast payback and high ROI.

• Finally, don’t scrimp on training; get all the training and consulting you need to get the “biggest bang” from your software investments.

Do you have anything to add? What are you doing to prepare for the economic recovery? Please leave a comment.

Charles Smart is the founding President of Smart Software. He currently serves as Vice Chairman, on Smart Software’s Board of Directors, as a company spokesman and in development of strategic business relationships. Prior to founding Smart Software, he was a management consultant at the Stanford Research Institute (SRI International) and Policy Analysis, Inc., and served as a Lieutenant in the U.S. Navy.

Leave a Comment

Related Posts

Probabilistic Forecasting for Intermittent Demand

Probabilistic Forecasting for Intermittent Demand

The New Forecasting Technology derives from Probabilistic Forecasting, a statistical method that accurately forecasts both average product demand per period and customer service level inventory requirements.

How to Choose a Target Service Level to Optimize Inventory

How to Choose a Target Service Level to Optimize Inventory

When setting a target service level, make sure to take into account factors like current service levels, replenishment lead times, cost constraints, the pain inflicted by shortages on you and your customers, and your competitive position.

Ten Tips that Avoid Data Problems in Software Implementation

Ten Tips that Avoid Data Problems in Software Implementation

Once a customer is ready to implement software for demand planning and/or inventory optimization, they need to connect the analytics software to their corporate data stream.This provides information on item demand and supplier lead times, among other things. We extract the rest of the data from the ERP system itself, which provides metadata such as each item’s location, unit cost, and product group.

Recent Posts

  • Manager In Warehouse With Clipboard and Safety stock controlBacking into Safety Stock is the Safe Play
    Safety stock is a critical component in any system of inventory management. Indeed, some inventory software treats safety stock as the key decision variable in the quest to balance inventory cost against item availability. Unfortunately, that approach is not the best way to strike the balance. […]
  • Hand placing pieces to build an arrowProbabilistic Forecasting for Intermittent Demand
    The New Forecasting Technology derives from Probabilistic Forecasting, a statistical method that accurately forecasts both average product demand per period and customer service level inventory requirements. […]