We often come into contact with potential customers who claim that they cannot use a forecasting system since they are a “build-to-order” manufacturing operation. I find this a puzzling perspective, because whatever these organizations build requires lower level raw materials or intermediate goods. If those lower level inputs are not available when an order for the finished good is received, the order cannot be built. Consequently, the order could be canceled and the associated revenue lost.

I agree that in such an environment, forecasting the finished good is not always possible or particularly helpful. Sometimes it’s helpful, but not sufficient. In any case, it is critical to make sure that the underlying raw materials and intermediate goods that go into the finished good are available. Demand for these can certainly be forecasted.

The organization’s goal would be to maintain service level inventories for these intermediate goods that are high but not unaffordable. Planners will need to set optimal stocking levels for these materials, balancing service level requirements against available budget. Since a given intermediate good could serve as an input to more than one finished good, the volatility of the demand for the intermediate good would be less than the volatility of the demand for a specific finished good. Hence, the safety stocks necessary to keep high service level inventories of the intermediate goods would be relatively lean.

Three companies, all users of SmartForecasts, serve as interesting examples. The first is a chemical company, Bedoukian Research, which manufactures custom chemicals for various clients. Each of these “finished goods” is a unique combination of intermediate chemical compounds. Bedoukian begins its demand planning with a finished goods forecast, which drives the production schedule and allocation of essential production resources. This requires exercising considerable judgment, as finished goods demand changes dynamically.

Once these finished good forecasts are created, raw material requirements can be estimated via a bill of material disaggregation. Bedoukian combines these results with safety stock estimates, based on actual utilization rates and service level objectives to be achieved, to generate the complete, service level-driven forecast for raw materials. This has allowed Bedoukian meet its production requirements with significantly less inventory.

The second company manufactures the internal components for mobile phones, where finished goods are specialized combinations of these components. For example, an order may call for a certain number of phones with unique labels on the case. This is the finished good for this order. Everything that goes into that order, except for the label, is built out of standard components. Again, SmartForecasts will be used to keep lean, high service level inventories of the components. This company thought that the only way to manage component inventories was via bill of material aggregations. They are now looking at the actual utilization rate for the components and setting much leaner inventories while maintaining high component availability.

A third company, NKK Switches, which explored this topic in their recent webinar (see CFO Bud Schultz’ guest blog post), considered their products to be “unforecastable”. You can read more about it below, but overall NKK Switches was able to forecast components and meaningful aggregations of product families. By tracking forecast vs. actuals over several months, NKK was able to demonstrate the accuracy of its forecasts to its Asian factory suppliers, and convince them to shift from a “build-to-order” model to “build-to-forecast.” This change has resulted in dramatic reductions in lead times, in many cases cutting them in half, increasing customer satisfaction and the overall sales close rate.

The bottom line here is that there is a perfectly viable—I would say essential—method of demand forecasting for build-to-order businesses, setting high service levels for pivotal input resources. If you would like to know more, please drop me a note, at nelsonh at smartcorp dot com.

Nelson Hartunian, PhD, co-founded Smart Software, formerly served as President, and currently oversees it as Chairman of the Board. He has, at various times, headed software development, sales and customer service.

There’s a stale old joke: “There are two types of people – those who believe there are two types of people, and those who don’t.” We can modify that joke: “There are two types of people – those who know there are three types of supply chain analytics, and those who haven’t yet read this blog.”

The three types of supply chain analytics are “descriptive”, “predictive”, and “prescriptive.” Each plays a different role in helping you manage your inventory. Modern supply chain software lets you exploit all three.

Descriptive Analytics

Descriptive Analytics are the stuff of dashboards. They tell you “what’s happenin’ now.” Included in this category are such summary numbers as dollars currently invested in inventory, current customer service level and fill rate, and average supplier lead times. These statistics are useful for keeping track of your operations, especially when you track changes in them from month to month. You will rely on them every day. They require accurate corporate databases, processed statistically.

Predictive Analytics

Predictive Analytics most commonly manifest as forecasts of demand, often broken down by product and location and sometimes also by customer. These statistics provide early warning so you can gear up production, staffing and raw material procurement to satisfy demand. They also provide predictions of the effect of changes in operating policies, e.g., what happens if we increase our order quantity for Product X from 20 to 25 units? You might rely on Predictive Analytics periodically, perhaps weekly or monthly, when you look up from what’s happening now to see what will happen next. Predictive Analytics uses Descriptive Analytics as a foundation but adds more capability. Predictive Analytics for demand forecasting requires advanced statistical processing to detect and estimate such features of product demand as trend, seasonality and regime change.  Predictive Analytics for inventory management uses forecasts of demand as inputs into models of the operation of inventory policies, which in turn provide estimates of key performance metrics such as service levels, fill rates, and operating costs.

Prescriptive Analytics

Prescriptive Analytics are not about what is happening now, or what will happen next, but about what you should do next, i.e., they recommend decisions aimed at maximizing inventory system performance. You might rely on Prescriptive Analytics to best posture your entire inventory policy. Prescriptive Analytics uses Predictive Analytics as a foundation then adds optimization capability. For instance, Prescriptive Analytics software can automatically work out the best choices for future values of Min’s and Max’s for thousands of inventory items. Here, “best” might mean the values of Min and Max for each item that minimize operating cost (the sum of holding, ordering, and shortage costs) while maintaining a 90% floor on item fill rate.

Example

The figure below shows how supply chain analytics can help the inventory manager. The columns show three predicted Key Performance Indicators (KPI’s): service level, inventory investment, and operating costs (holding costs + ordering costs + shortage costs).

 Figure 1: The three types of analytics used to evaluate planning scenarios

The rows show four alternative inventory policies, expressed as scenarios. The “Live” scenario reports on the values of the KPI’s on July 1, 2018. The “99% All” scenario changes the current policy by raising the service level of all items to 99%. The “75 floor/99 ceiling” scenario raises service levels that are too low up to 75% and lowers very high (i.e., expensive) service levels down to 95%. The “Optimization” scenario prescribes item specific service levels that minimizes total operating costs.

The “Live 07-01-2018” scenario is an example of Descriptive Analytics. It shows the current baseline performance. The software then allows the user to try out changes in inventory policy by creating new “What If” scenarios that might then be converted to named scenarios for further consideration. The next two scenarios are examples of Predictive Analytics. They both assess the consequences of their recommended inventory control policies, i.e., recommended values of Min and Max for all items. The “Optimization” scenario is an example of Prescriptive Analytics because it recommends a best compromise policy.

Consider how the three alternative scenarios compare to the baseline “Live” scenario. The “99% All” scenario raises the item availability metrics, increasing service level from 88% to 99%. However, doing so increases the total inventory investment from $3 million to about $4 million. In contrast, the “75 floor/99 ceiling” scenario increases both service level and reduces the cash tied up in inventory by about $300,000. Finally, the “Optimization” scenario achieves an 80% service level, a reduction from the current 88%, but it cuts more than $2 million from the inventory value and reduces operating costs by more than $400,000 annually. From here, managers could try further options, such as giving back some of the $2 million savings to achieve a higher average service level.

Summary

Modern software packages for inventory planning and inventory optimization should offer three kinds of supply chain analytics: Descriptive, Predictive, and Prescriptive. Their combination lets inventory managers track their operations (Descriptive), forecast where their operations will be in the future (Predictive), and optimize their inventory policies in response in anticipation of future conditions (Prescriptive).

Do you know which items have too much or too little inventory? What if you knew? How would you go about cutting overstocks while still ensuring a competitive service level? Would you be able to reduce stockouts without incurring a prohibitively expensive inventory increase? How would these changes impact service levels, costs and turns—for individual items, groups of items and overall?

Most companies know they have too much or too little inventory but lack a key ingredient for optimizing inventory: Service Level-Driven Demand Planning. To take action, you must know how much inventory is needed to satisfy the service level you require. More fundamentally, you need to know the specific service level that will result from your current inventory policies, the gap to be addressed and its financial implications.

Many organizations, especially those with intermittent demand, find this to be an exceptionally challenging trial and error process.

Moving to a service level-driven approach will overcome this challenge and ensure that rebalancing inventory improves service level performance at a lower cost. Start with the most accurate demand forecast possible, calibrate for forecast risk and then determine your optimal inventory position. In a recent webinar, I demonstrated Service Level-Driven Demand Planning and how SmartForecasts can be used to drive this process:

1. Measure the service levels that will be achieved at current inventory levels and with your current inventory policy.
2. Identify items that will achieve high service levels (98%+) but at prohibitively high cost.
3. Identify items that are at high risk of stockout (service levels < 75%).
4. Run multiple what-if scenarios based on a different prioritization of service levels by item or item groups. Choose the scenario that optimizes financial constraints with service objectives.
5. Quantify cash savings from reducing overstocks and the costs to increase inventory when service levels are unacceptably low.
6. Take action to establish new service level-driven reorder points, order quantities and inventory levels to meet your service targets and budget.

To view the webinar replay, please click here and complete the registration request.

Gregory Hartunian serves as President of Smart Software and as a member of the Board of Directors. A graduate of The F.W. Olin School for Business at Babson College, he formerly served as Vice President, Sales and Operations.

Does your extended supply chain suffer from extreme seasonal variability? Does this situation challenge your ability to meet service level commitments to your customers? I have grappled with this at Rev-A-Shelf, addressing unusual conditions created by Chinese New Year and other global events, and would like to share the experience and a few things I learned along the way.

First, let me explain our situation. We import 60% of the parts we use to build our kitchen and bath accessories from China and Europe. Most of the year we were able to plan our inventory needs using a spreadsheet-based min/max approach. But not during Chinese New Year, which drives the planet’s greatest annual population migration. Chinese New Year shuts down production for up to two months, creating significant supply risk as we strive to meet our three day order fulfillment commitment.

We solved our problem, introducing statistical demand forecasting with the flexibility to extend lead times when necessary, the ability to reliably establish safety stocks that achieve our required service levels and a continuous reporting system that lets everyone know exactly where we stand. However, success required much more than a new piece of software. We needed to change the way we view future demand, supply risk and safety stock. Here are a few key things we did that made all the difference.

Stakeholder education and buy-in

Regardless of the project, it’s always best to enlist the buy-in of all stakeholders. We knew we had to do something to solve our problem, but there was bound to be resistance. Senior managers, for example, had developed a healthy distrust of software and wondered whether demand forecasting software could help. Our buyers had developed their own perspectives and procurement methods, and felt personally at risk as we considered new approaches.

People came around as they developed a common understanding of the problem and how we would address it. Education was a big part of the solution. We explained how forecasting works and key factors we should all understand: how to analyze trends, how to use “what if” scenarios, impact of shifting lead times, how to relate service levels to supply risk and safety stock and key performance indicators like inventory turns. Going through this process together, we all became stakeholders in the solution.

Use the Right software

When you have lots of part numbers and any sort of supply or demand variability, you just cannot forecast effectively with a spreadsheet. With our min/max forecasting system, we were planning to an average, and it wasn’t working. Average usage has inherent flaws for planning purposes—it’s always looking backward!

You need software that looks ahead, recognizes seasonal patterns and enables you to determine how much stock you’ll need to meet required service levels over varying lead times.

Fine-tune processes

When the old ways don’t work, you need to be open to adjusting your assumptions. Think less about where you’ve been, and more about where you want to be. Take a look at your lead times and plan to your desired service level. Last year’s history may not be the best predictor of this year’s demand. The same forecast horizon may not be appropriate for all products or certain time of the year.

Make the Forecast Actionable

It’s not enough to produce an accurate forecast and estimated inventory stocking levels. You’ve got to develop a way to make the information actionable for those tasked with using it. We developed a set of reports that enabled buyers to leverage better forecast and safety stock information. Now, at the end of every month, we produce a forecast report that provides a clear picture of current inventory, safety stock, past usage, forecasted usage, incoming deliveries (PO’s) and recommended order quantities.

Validate Results

You can, and we did, test our new methods against our own demand history. Still, an authoritative outsider can make acceptance easier. We commissioned a study by a professor at Louisville University’s College of Business who set one of her graduate students to the task. Through them we were able to reinforce what we saw happening from our results, and feel comfortable that we were on a good path.

All of these factors helped Rev-A-Shelf transform its demand planning process, to great effect. Today we are exceeding our service level targets, and our fill rate, based on a three day ship cycle, is showing steady improvement, and trending up. Overall, units-in-stock have stayed flat while supporting a 13% increase in sales.

John Engelhardt is currently Director of Purchasing and Asian Operations for Rev-a-Shelf, LLC in Louisville, KY. He has held a variety of management positions both in private business and public organizations. At Rev-A-Shelf he held the position of International Sales Manager and Director of Sales Support before assuming his current position. He can be reached at johne at rev-a-shelf dot com.

Smart Software recently announced a Software as a Service (SaaS) option for SmartForecasts—SFCloud™. Premises-based perpetual licenses will continue to be the preferred software implementation method for many organizations, but there are many reasons why demand for cloud-based solutions is taking off. A vintage post by Bill Richardson at ApplicantStack Team Blog summarizes key benefits of the SaaS model.

At year’s end, we are often caught up in thinking and planning for the coming year. Did 2013 turn out the way you expected? Will 2014 be dramatically different? Are there other factors—things we are planning to do; things we think our competitors might do; outside forces like changing taste, demographics or economics—that might change the course of business in the coming year?

Most companies that do a formal forecast start out with a statistical projection of past sales patterns into the future. Your forecast model should detect and characterize any seasonality inherent in your markets and include that in the projection. But that’s just the first step.

The next thing to consider is product lifecycle. Nearly all products go through a predictable cycle of introduction, acceptance and growth, maturity (demand levels off) and finally decline to obsolescence. These cycles can be as short as weeks or as long as decades. Clothing fashions and consumer electronics would be on the shorter end of the scale, while products like plumbing fixtures and construction equipment would experience longer cycles. In specialized situations like bus fleet management, entire fleets may be replaced over defined transition periods. In any case, the demand forecast should be adjusted to reflect increasing or decreasing demand according to the product’s position in its lifecycle.

Now comes the hardest part—predicting the unpredictable. In general, the future is likely to look a lot like the recent past in a similarity to Newton’s first law of motion: a body in motion tends to stay in motion unless acted upon by an external force. But it’s those external forces that can send your carefully calculated forecast right into the gutter. A competitor might slash their prices to take away some of your market share. New technologies might obsolete your product before the end of its expected life span. Changing tastes or new regulations might stop sales in their tracks.

But good things might happen as well. You might be the one to slash prices or improve your product and take away a competitor’s business. Your product may catch the fancy of the market and sales will skyrocket. A competitor may abandon the business or go bankrupt, leaving you with more opportunity.

Should you plan for these kinds of things? Certainly, to the extent that you can. You may know when you’ll run promotions or phase in the next product line. But the future, by nature, is uncertain. History and your business knowledge of the past lay the foundation for your view of the future. Statistically-based tools can help you create a risk-adjusted forecast, with safety stock recommendations that correspond with the level of risk you are willing to take. Beyond this, your key to success is agility—the ability to adapt to changing conditions. Prepare the best forecast you can, build your plans around that forecast—then monitor sales and market conditions closely and continuously. Look for early warning that things may be going in a direction other than you predicted.

You must be willing to recognize and adapt to changing conditions—in other words, don’t fall in love with your forecast and ignore evidence that it may be wrong. Pride of authorship in this case can be deadly to the business.

It is also important to have contingency plans in place so you will be prepared to make the necessary changes to procurement, production and inventory to respond to the new estimate of demand. The best tools for this are the shortest possible lead times (both production and supplier), good supplier relationships and a clear view of the world and your markets.

Forecasting is difficult mainly because people know it is likely to be wrong and nobody likes to be publicly and visibly wrong. Nevertheless, a good forecast is necessary to position the resources necessary to satisfy customer demands. Just be open to the first signs of change and be prepared to react quickly and decisively.

Dave Turbide, CFPIM, CIRM, CSCP, CMfgE is a New Hampshire-based independent consultant and freelance writer. He can be reached via e-mail at dave at daveturbide dot com.