Supply Chain Trends and Issues: Our Weekly Feature Article on Important Trends and Developments in Supply Chain Strategy, Research, Best Practices, Technology and Other Supply Chain and Logistics Issues  
  - February 10, 2009 -  

Supply Chain News: How Do Forecasting Processes Need to Change in Tough and More Dynamic Times?

pdf of this article

Getting more Granular and Urgent are Key, our Experts Say; Judgment has never been more Important


SCDigest Editorial Staff

SCDigest Says:
More problematic is that this approach increases, sometimes significantly, the planner’s workload. With many planning staffs already operating very thinly, something will have to give somewhere to enable the extra time gathering and analyzing still more data.

Click Here to See Reader Feedback

These tough economic times, combined with the growing volatility of demand, are making demand planning and forecasting very difficult processes right now.

Many pundits have commented in the press or elsewhere on the challenges of demand planning in this environment, or stated that existing processes may have to change, but few have offered any real details as to what those changes might really be.

We decided to ask some experts ourselves, and obtained some excellent insight.

Dr. Larry Lapide is Director, Demand Management, at the MIT Center for Transportation & Logistics, and a recognized expert in areas such as forecasting, demand planning, and sales and operations planning.

“Since we are in economic conditions that we have not seen in quite awhile (if ever), forecasters will not be able to put as much weight on historical patterns as the basis for demand forecasting,” Lapide told SCDigest. “Demand during these economic times will likely be more volatile than they have seen and heavily dependent on what is happening to the economy. Meanwhile, the market will be more competitive and a forecaster’s company likely will be doing more promotions as well.”

Lapide says he has been advising companies to focus more on three things:

(1) Use downstream information, such as POS data, as an early indicator on the changes taking place in a product’s consumption (i.e., their shipment data lags ‘real’ consumption).

(2) Keep abreast on what is going on in the market by reading more of what economists are talking about in order to incorporate economic impacts into their demand forecasts.

(3) Stay on top of what their Sales and Marketing organizations are doing to incorporate promotional effects that will heavily impact demand.

(Supply Chain Trends and Issues Article - Continued Below)




Demand Planning World is Different Right Now

The current economic dislocation is changing demand planning processes and requirements in several ways.

Anand Iyer, a Fellow at i2, agrees with Lapide that a lack of visibility to end customer demand can play havoc with forecasting processes now more than ever.

“If you are used to looking at demand based on what you are shipping from your DC to the retailer’s DC, that may be different than actual consumer demand, because it is only based on what the retailer is willing to take right now,” Iyer observed.

“Right now, you have to pay a lot more attention to signals from the true customer, not the sales channels,” he said.

He noted some other anomalies that are occurring. For example, he said that in working with one of i2’s semiconductor customers, that company’s large tier 1 OEM customers were delaying orders until the last minute. Meanwhile, tier 2 customers were firming their orders earlier, better securing capacity and inventory, and hence receiving better service than the firm's largest customers.

Iyer also says that demand planners must develop a true sense of “urgency” to dig deeper into the numbers, find out what they don’t know that they could know, and get to a more detailed level of granularity in the numbers.

“Traditionally, companies have often used historical demand as the primary predictor of current demand, with some intuition and judgment thrown in from marketing and demand planners themselves,” Iyer says. “In this environment, that approach simply no longer works.”

He also says that the current environment is causing companies to do some things they should have been doing all along , such as better synchronizing demand plans with internal supply plans and taking inventory plans in many cases all the way back to suppliers.

“This become an “Aha” moment for many companies,” Iyer said, as the current challenges bringing the need for better integration back up the supply chain into stark relief.

He adds that companies are looking to achieve greater granularity in forecast processes in many ways. For example, he says many companies take regional forecast data and simply aggregate it up to a national number, again an approach that was workable in the past but dangerous today.

“I see companies now looking at POS or channel sell through data coming out of each region and even each retail store,” he said. “Today, you are seeing much greater differences in demand regionally for many products than you have in the past, so you really have to analyze the numbers at the lowest level of granularity possible.”

Iyer also noted that that the need for demand planner judgment is now more important than it has ever been in recent times.

“You have to get all the data you can, including collaborative forecasting, syndicated POS data as appropriate, what competitors are doing in terms of promotions,” he said. “But in the end, planners really have to hone their own judgment based on this expanded data set and be able to fully rationalize that judgment.”

But this more granular approach to forecasting doesn’t come free. In some cases, a company’s existing demand planners may lack the skill sets or orientation to approaching the task in this new way, though Iyer says most pick it up quickly based on a bit of training.

More problematic is that this approach increases, sometimes significantly, the planner’s workload. With many planning staffs already operating very thinly, something will have to give somewhere to enable the extra time gathering and analyzing still more data.

Do you agree with our expert’s insights into how demand planning needs to change in the current environment? What would you add, or what is your company doing? Are new skill sets required? Let us know your thoughts at the Feedback button below.

Send an Email