Caught in a Perfect Storm, SmartForecasts Helps Rev-A-Shelf Weather the Crisis

The Smart Forecaster

Pursuing best practices in demand planning, forecasting and inventory optimization

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.

Leave a Comment

Related Posts

Caught in a Perfect Storm, SmartForecasts Helps Rev-A-Shelf Weather the Crisis

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.

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. […]
Forum Energy Streamlines Demand Planning, Improves Market Responsiveness with SmartForecasts

Forum Energy Technologies is a $1.7 billion global oilfield products company, serving the oil and natural gas industry. With over 3,500 employees, Forum is headquartered in Houston, TX and has manufacturing and distribution facilities strategically located around the globe. Forum’s Valve Solutions Division in Stafford, Texas, uses SmartForecasts in combination with its Epicor ERP system for planning at the finished goods and component levels.

The Challenge

Forum faced a complex demand planning challenge. Its business, highly cyclical and sensitive to market changes, required a responsive planning process. This was complicated by the fact that the majority of its stocked items exhibit sporadic, seemingly unforecastable intermittent demand. This is characterized by many zero or low volume values interspersed with random spikes that are often many times larger than average demand.

Forum relied upon a cumbersome excel-based forecasting process that generated estimates of average demand but did not provide guidance on appropriate levels of stock. Determining stock levels was mostly guesswork, based on intuition. The large number of items and lack of effective tools resulted in an inefficient planning process where only a small portion of items could be reviewed on a regular basis. It was not unusual for items to be forecast once per quarter.

Given six to ten month lead times for imported items and a cyclical business climate, Forum needed to forecast more frequently in order to react faster to changing market conditions. This would require a data-driven solution that could determine appropriate stock levels to reduce stockout risk while keeping inventory at an affordable level.

The Solution

Forum Energy purchased SmartForecasts for its abilities to automatically apply appropriate forecasting methodologies to specific items, handle intermittent demand, and interface with its Epicor ERP.

SmartForecasts enabled Forum to make several changes to the way it managed its inventory:
• automate forecasting, increasing efficiency to the extent that planners could forecast products on a monthly basis and have the time and ability to review exceptions;
• forecast at both the finished goods and component levels, using SmartForecasts intermittent demand forecasting for all components;
• perform a meaningful S&OP process spanning supply chain management and sales; and,
• model different inventory policy and service level scenarios to better allocate inventory investment and react more quickly to market changes.

The Results

Like other companies in the oil and gas industry, Forum has experienced a slowdown in its business. SmartForecasts enabled Forum to quickly adjust its inventories by running what-if scenarios and evaluating the risks involved with different actions. While it has only recently started to use Smart’s “Service Level Demand Planning” methodology, according to Rod Cardenas, purchasing manager, “the service level planning method has led to productive conversations between sales and supply chain and given us a platform or common ground from which we base our discussions.”

Forum now does intermittent demand forecasting for all of its components. That capability has been an important factor in increasing confidence in forecasting results. According to Cardenas, “SmartForecasts gives us good information to work with. People are feeling comfortable with numbers, and through our S&OP process we’ve been able to create buy-in across the company.”

Over four years Forum experienced revenue growth of 15-20% per year, and was able to do so without increasing inventory. Using SmartForecasts it has also been able to increase inventory turns from 1.8 times to an estimated 3.0 times this year.

Recently Forum began a vendor-managed inventory (VMI) program for its fasteners. SmartForecasts is being used to help vendors manage their bin sizes. Shifting inventory risk to its vendors, Forum has been able to reduce in-house inventory from 700,000 units to 100,000, and is expected that it will be able to turn its in-house inventory 12 times.

“We’re just scratching the surface of taking advantage of SmartForecasts’ capabilities, added Cardenas. “There’s a lot more we can do.”

SmartForecasts Helps Metro-North Railroad Keep the Trains Running On-Time


Metro-North Railroad (MNR) is part of the New York Metropolitan Transit Authority, and services commuters and other travelers in the New York metropolitan area. It’s the second largest commuter railroad in the U.S. serving 300,000 passengers each weekday, while operating 1,266 engines and rail cars over 775 miles of track.

The Material Management division is charged with stocking spare parts inventory. It maintains 37,500 parts valued at $107 million. More than 80% of its inventory has extremely volatile and hard-to-forecast intermittent demand.

Challenge: Safety Stock for Long-Lived Assets

MNR’s prime objective is to move people in a timely manner. For service outages, it must have the right items in stock to fix the problem.

Service level is MNR’s most important key performance indicator. In 2008, it found itself with too much expensive inventory sitting idle and an unacceptably low 95.8% service level.

To remedy this situation, Material Management needed to be able to analyze all of MNR’s inventory, including items with intermittent demand, and then optimize them while simultaneously increasing service levels. MNR did not have appropriate planning tools to do the job.

MNR also faced several other challenges: ridership was increasing, and the agency was expanding its fleets, retiring old trains, and introducing new ones. With a five-year planning horizon, MNR needed to maintain adequate safety stock for an asset service life of 20 years or more, and replacement lead times, for some parts, of more than a year.

Solution: SmartForecasts

After a thorough evaluation process, MNR selected Smart Software’s SmartForecasts®. Smart’s solution uses patented bootstrapping technology that accurately forecasts intermittent demand. It also provides inventory managers with accurate estimates of the optimal inventory and safety stock required to meet future orders and achieve desired service levels.

MNR realized a payback after only 9 months. SmartForecasts was used to identify overstocked items, many of which had pending orders. By putting holds on some orders, and canceling others, MNR realized immediate improvements in cash flows. Management was also able to identify items at risk of stocking out and re-deployed resources that buoyed immediate service level improvements.

Results:Accurate Intermittent Demand Forecasts

MNR’s multi-year project continues to provide improving results.Recently, some of the company’s notable successes and approaches for utilizing SmartForecasts, were cited in Material Management’s 2015 report to its Board of Directors on 2014 operations:

  • Parts Inventory to support its equipment has declined 8% from $128 million since 2007.
  • In 2014, following a continuing trend, customer service levels reached a record high 98.7%. This was accomplished with a new fleet growth of over 2%
  • SmartForecasts was instrumental in reducing inventory growth for new equipment from a projected 10% to only 6%.
  • MNR uses SmartForecasts to identify multi-year service part needs for long-term contracts. By contracting with suppliers on a multi-year agreements, administrative lead times were greatly reduced. This enabled further reductions in stock without compromising service.
  • Smart Forecast was critical in the formulation of a stock reduction plan for MNR’s retiring fleet, and helped identify inventory retirement reserves, enabling MNR to maximize disposal benefits and warehouse bin space reclamation.
  • MNR identified $1.6 million in inactive inventory for final disposal, and provided availability of excess parts to Long Island Railroad, thus potentially minimizing additional MTA inventory investments.

According to Rich Price, Chief Material Officer at MNR, “SmartForecasts does the job. It’s the only solution out there I’m aware of that has really licked the intermittent demand modeling challenge. It gives us accurate information and more importantly at the lowest dollar. That’s the bottom line we’re really interested in. Without SmartForecasts, I think we would have seen a continued growth in our inventory and not necessarily gotten any greater benefit for service.”


SmartForecasts helps Vicor Increases Service Levels 10 Percentage Points and Reduces Inventory 20 Percent

Vicor Corporation, a publicly traded, $200 million company with headquarters in Andover, Massachusetts, is a build-to-order manufacturer of power conversion products for the electronics OEM market. Vicor produces 30,000 product items grouped into fifteen product families. The items are made from 10-15,000 components. As a build-to-order manufacturer, Vicor does both assembly and fabrication. It also tries to run a lean operation, keeping a very low backlog of finished good items and only ordering component parts on an as-needed basis. This strategy tends to lower inventory costs but also heightens the company’s risk of stocking out of key items.

The Challenge

While a few hundred of Vicor’s products are sold on a regular basis, some are ordered as little as every two years-a factor that makes demand forecasting and inventory planning more difficult. Yet Vicor’s business success depends critically on having the right component in stock when it’s needed to make a product item. So, having an accurate forecasting process in place is a major contributor to the company’s smooth operations.

As CIO of Vicor Corporation, Doug Richardson was charged with helping the company achieve two strategic objectives: improve on-time delivery of the company’s products and reduce inventory levels.

The Solution

Starting in 1999, Richardson led a team that replaced Vicor’s old manufacturing systems and a homegrown forecasting system with the PeopleSoft ERP system and SmartForecasts Enterprise. In May 2002, Vicor completed a tight integration of SmartForecasts with PeopleSoft’s Production Planning module, and by July 2002, the company started seeing positive results. Mark Vernaglia, senior operations planner at Vicor, makes the point that the ability to interactively adjust the statistical forecast results and understand seasonality were two big selling points in SmartForecasts’ favor.

Vicor has a two-phase demand forecasting process that combines both sales force and statistical forecasting techniques. In the first phase, the sales force obtains judgmental predictions directly from Vicor’s largest customers regarding their demand for each of the company’s product items. These customers account for about 40 to 50 percent of Vicor’s shipment volume. In the second phase, the shipment history for these customers is excluded, and SmartForecasts’ Automatic Forecasting capabilities are used to forecast statistically the product demand expected from all of the other customers. By summing these two sets of results, Vicor generates the total production forecast for each product item.

Because Vicor needs to assign future revenue goals for its product families as part of its sales and operations planning (S&OP), the company also makes use of SmartForecasts’ top-down modeling capabilities. It uses SmartForecasts to impose the build plan for each family and to intelligently distribute that goal among the items in the family.

The Results

In less than a year following integration with the PeopleSoft system, SmartForecasts uncovered a critical upward trend in Vicor’s business. “At first we thought something was wrong,” stated Richardson. So Vicor did the forecast again with more demand history as input. Sure enough, SmartForecasts had spotted a trend that began in late 2001.

“The SmartForecasts results provided evidence that the post-9/11 business downturn had hit bottom,” Richardson continued. “If SmartForecasts hadn’t uncovered the upward trend, we wouldn’t have had enough inventory on hand to meet demand.”

The system has continued to deliver substantial results that put the company well on the road to achieving its strategic objectives. According to Richardson, on-time delivery performance (customer service level) has increased by 10 percentage points and, at the same time, inventories are down by 20 percent. The improvements are the results of hard work by a team focused on improving order fulfillment, a key element of which is the company’s ability to better forecast inventory needs and understand demand patterns with SmartForecasts Enterprise.

The forecasting process has been greatly enhanced since the integration of SmartForecasts Enterprise with Vicor’s PeopleSoft system. Before the implementation, Vicor only forecasted about twenty percent of its products, in part because the old homegrown forecasting system was cumbersome to use and didn’t have the necessary horsepower or data capacity. Because of the speed of SmartForecasts and its ease of use, the company can now forecast more than eighty percent of its items. The ability to forecast more products has been a big factor in reducing Vicor’s inventory requirements and improving service levels.

Mark Vernaglia feels very comfortable using SmartForecasts and claims that he was able to be proficient with little training. “We’re getting great results,” stated Vernaglia, “and I haven’t even started taxing the software’s capabilities.”

The results have made Vernaglia more confident in his job. “With SmartForecasts we’ve been able to be much more aggressive with our inventory planning,” he continued. “You’ve got to have confidence in your forecast, and we now have the confidence to stock certain component parts that we didn’t have before.”

In the future, Vernaglia plans to expand the forecasting process to the component level and to start using SmartForecasts’ patented intermittent demand forecasting capabilities. He’s confident that these unique capabilities will help Vicor improve its ability to set stocking levels appropriately for slow-moving items.

Vicor looks forward to continuing improvements. “I’m willing to bet that our numbers will continue to go in opposite directions-inventory down and service levels up,” wagered Vernaglia. These are the kinds of changes that continue to help Vicor’s management achieve its strategic goals.

Chemical Manufacturer, Bedoukian Research, Uses SmartForecasts to Transform its Supply Chain Planning

Bedoukian Research, Inc., located in Danbury, Connecticut, is a specialized process manufacturer producing high quality ingredients used in flavors and fragrances for customers worldwide. Bedoukian makes 350 specialty chemicals, in small batches, from 2000 raw materials. In addition, the company manufactures fifty insect pheromones used to attract and control insect pests, and resells 100 other products. Two-thirds of its products have hard-to-forecast, intermittent demand.

A Smart Software customer since 1987, Bedoukian Research not only uses SmartForecasts to forecast products with normal demand, but became an early adopter of SmartForecasts intermittent demand forecasting technology.

The Challenge

Many of Bedoukian’s raw materials are sourced in Asia, and are shipped by sea. Lead times can run 2-3 months and affect equipment usage and the company’s ability to meet the company’s high customer service level expectations. Many of the company’s products are built-to-order, and the same raw material might be a component in numerous products. When an ingredient is out-of-stock, it can lead to expensive emergency shipments or lost sales. In addition, Bedoukian’s planning process must account for product shelf lives and changes in commodity pricing.

To achieve high customer service goals, the company needed a solution that would provide accurate stocking level estimates for all of its products, including those with intermittent demand.

The Solution

Bedoukian uses SmartForecasts integrated with Chameleon, an ERP system specifically designed for chemical manufacturers.

Each month, SmartForecasts forecasts the demand for every finished goods item four months into the future, and estimates optimal inventory stocking levels for its raw materials too. The company plans to be one of the first users of Smart Software’s “Bill-of-Materials” forecasting capabilities that should simplify the planning process across product lines, while ensuring that its standard costs keep pace with the escalating costs of raw materials.

The Results

“The intermittent demand forecasting capability has been very useful,” claims Leona Eggleston, who leads the forecasting process. “We’ve gone from forecasting our slow-moving items by hand with poor results to automatically generating highly accurate stocking level estimates.”

In addition, SmartForecasts helps the company see the patterns in its product demand and better track sales performance. When buying patterns change, salespeople can swiftly take remedial action and avoid losing a customer’s business.

In the past five years, Bedoukian has been able to support increased sales with less inventory. While its sales increased 15%, finished goods inventory increased only 4% and raw materials inventory 5%. Additionally, it has improved its customer service by reducing late shipments to customers due to insufficient stock from 16% to 3%.

Improved planning has also enabled Bedoukian to consolidate ocean shipments for almost every raw material which decreases total freight expenses, and cuts down on emergency airfreight costs which saves over $1,000 for every drum.

Bob Bedoukian, president of the company, is more than satisfied and claims, “The more we use SmartForecasts, the more important it becomes for our business and the more efficiently our business runs.”

Driving Better Auto Aftermarket Parts Distribution at Prevost Car

SmartForecasts’ forecasting solutions increase parts availability and reduce inventory costs

Prevost Car, Inc. is a leading Canadian manufacturer of intercity buses and coach shells for high-end motor home and specialty conversion. Prevost Parts, a division of Prevost Car, serves the North American auto aftermarket distributing original coach and urban bus parts for Prevost Car and Nova Bus, as well as replacement parts for other models. To serve its market, the company maintains two distribution centers, one in Canada and one in the US; five service centers; and over 25,000 active parts. 70 percent of those parts have an intermittent, irregular demand profile, which makes them very hard to forecast and manage.

The Challenge
In 2002, Prevost Parts’ management started a comprehensive program to improve the company’s parts distribution system. Prevost needed inventories at its warehouses to more accurately reflect demand at those locations, and required more accurate forecasting inputs to its production planning processes. 25 percent of its orders were not being shipped from the distribution point closest to the customer or were backlogged. This situation caused increased transportation costs, delivery delays, and unacceptable customer service levels. In addition, because Prevost’s SAP Min/Max system didn’t compensate for seasonality or trends, the company had too much stock in the off-season and too little when the need was greatest.

To respond to these issues, Prevost’s management conducted a formal evaluation of six demand planning and forecasting systems, including SAP’s demand planning module, to find a solution that would overcome the limitations of the Min/Max system and enable it to use the full potential of SAP’s distribution resources planning (DRP) module. Because the DRP system wasn’t in sync with the demand forecast, product orders weren’t being issued in a timely and accurate manner.

In addition, Prevost Parts was lacking effective demand planning within its multiple site/multiple level distribution network. To accomplish this, the company needed a forecasting solution that could accurately forecast demand for all of its products and deliver accurate safety stock estimates at the lowest level in the network. It also required a system that could provide the flexibility of cumulative lead time measurements, isolate and identify extreme values, recognize seasonal patterns, and provide top-down and bottom-up product group forecasts.

The Solution
Of the six applications considered, Prevost Parts selected SmartForecasts Enterprise from Smart Software, Inc. SmartForecasts scored highest in the evaluation of more than eight functional criteria. The key deciding factor, however, was SmartForecasts’ ability to generate accurate sales forecasts and safety stock requirements for intermittently demand products. Products of this type are very difficult to forecast because of the slow-moving, irregular nature of their demand patterns.

As Dave Gilbert, logistics director at Prevost Parts, noted, “With most of our parts having intermittent demand, the ability to solve that problem was very important to us, and SmartForecasts’ unique solution in that area was a major advantage.”

In addition, the test results showed that SmartForecasts’ inventory stocking recommendations could reduce required safety stock levels as much as eighteen percent more than the next best application. There were other factors that also weighed in the decision, including SmartForecasts’ ability to interface easily with SAP’s DRP system and simplify the process of managing safety stocks.

The Results
Early in 2004, SmartForecasts was installed at corporate headquarters in Ste. Claire, Quebec, where the software is directly linked to an Oracle database used by Prevost Part’s SAP system to store data and forecasting results. Since implementation, the company’s demand planners have used the system to make a number of planning changes, and those changes have begun to pay off. Backorders of the company’s most frequently demanded parts have been reduced 65%, lost sales are down 59%, and fill rates increased from 93% to 96% in just 3 months.

But the implementation of these changes required Prevost Parts to creatively adjust the way it operates. The company’s strategy for improving its distribution activities required a major shift in the way it forecasted demand. Rather than simply forecasting overall demand, the company needed to forecast demand for each product item at each of its distribution and service centers. This would give a more accurate picture of local demand and enable the company to better align its inventory where it was needed most. Local forecasts could then be rolled up into a company-wide forecast for planning purposes.

At the beginning of every month, Prevost Parts transfers the past month’s consumption data directly from SAP tables in the Oracle database into a Demand History table maintained by SmartForecasts in the same Oracle database. Using SmartForecasts, Prevost automatically produces a forecast of demand and safety stock requirements for all products at each of its distribution and service centers using thirty-six months of history. These forecasts are validated by a demand planner and branch parts managers using audit reports and graphical adjustment facilities in SmartForecasts and then passed to the SAP tables for direct use by the SAP DRP program.

Based on these results, Prevost Parts started stocking its warehouses on a monthly basis, but management quickly found that monthly shipments weren’t meeting its replenishment needs. For this reason, demand is now broken down into weekly buckets, resulting in a Just-in-Time ordering process that has lowered stocking levels at its warehouses.

Prevost plans to continue to measure its progress and savings, and fully expects that it will continue to see major improvements in reducing costs, increasing customer service, and effectively deploying its production and distribution assets. According to Gilbert, “We need to have the right parts in the right place to support our customers. SmartForecasts helps us to not only improve our inventory allocation but also significantly reduce transportation and inventory costs.”

Read the case study on Prevost Parts published in APICS Magazine.

Read the article on Prevost Parts written by Alex Daudelin in the Canadian logistics journal, Gestion Logistique.