An announcement last week from ERP giant SAP arguably marks the coming of age of an analytics technology generally credited with improving the statistical forecast error of product demand forecasts by 30 to 40 percent—at least, in specific, generally consumer-facing applications.
The analytics approach known as demand sensing, deployed by consumer goods companies such as Procter & Gamble, Kraft Foods, Diageo, Kimberly-Clark, and Unilever, combines improved mathematical techniques with consumer buying shifts and other data which is not normally available to traditional forecasting techniques, in order to deliver a genuine step change in forecast accuracy.
And late last week, SAP announced that it plans to acquire one of the small number of companies operating in the space: forecasting and inventory optimization specialist SmartOps Corporation based in Pittsburgh.
The two companies certainly aren’t strangers. Founded in 2000, SmartOps has had a formal business relationship with SAP since 2006, and specializes in the use of large‑scale, stochastic algorithms using predictive analytics to eliminate uncertainty and risk from supply chain management processes.
The benefits of the SmartOps algorithms are twofold. First, through demand sensing, businesses gain a better understanding of short-term demand. Second, by factoring that better understanding of short-term demand into inventory calculations, they can then optimize inventory levels in order to serve that demand.
The result: increased sales, in those scenarios where traditional forecasting techniques would have underestimated demand; but also lower inventory levels in those scenarios where those traditional forecasting techniques would have overestimated demand.
How does it work? In contrast to traditional time-series analysis forecasting techniques such as exponential smoothing, moving averages and Box-Jenkins models, demand sensing aims to incorporate a much broader range of demand signals, in as near real-time as possible.
The contrast is stark. Traditional techniques simply extrapolate the past into the future: trends in past sales, past seasonality, and past promotional impact carried forward to produce a calculated estimate of future demand. Demand sensing, on the other hand, takes the traditional forecast as an input, but adds to the mix real‑world events such as market shifts, weather changes, changes in consumer buying behavior, social network sentiment, and real-time point of sales data.
And it’s that real-time forecasting ability that drove not only the SmartOps acquisition, but also SAP’s plans for SmartOps’ analytics technologies in the future, says Hans Thalbauer, senior vice president of business solutions for supply chain at SAP.
In particular, says Thalbauer, SAP plans to migrate the SmartOps capabilities onto the recently-announced in‑memory HANA database and analytics technology platform that powers its flagship Business Suite ERP system. The goal: real-time demand sensing, even with the data volumes typifying a Procter & Gamble or Unilever.
“Our vision at SAP is to realize the real-time supply chain, but the closer you move to the real-time supply chain, the more the distinction between supply chain planning and execution blurs: it’s about planning better, and executing better—at the same time,” he says. “We have already moved our advanced planning and scheduling and sales and operations planning capabilities onto HANA, and putting demand sensing and inventory optimization on the same platform, using the same data model, makes perfect sense.”
The acquisition of SmartOps also makes perfect sense strategically, too, according to Lora Cecere, founder and CEO of Baltimore-based Supply Chain Insights. The reason: demand sensing, first pioneered by Terra Technology, genuinely delivers on its promise.
“Some previous mathematically-based supply chain innovations—such as multi-echelon inventory optimization—have arguably over-promised and under-delivered, but you can’t say that of demand sensing. The statistics are impressive: you really do get a 30 to 40 percent better forecast,” Cecere says. “SAP will have been looking at its own maturing ERP market, noticed its customers buying licenses from Terra, and seen the need to respond.”
And once the acquisition closes later this quarter, the mission that will be set to SmartOps’ management and developers is very simply stated, she asserts: build a Terra Technology-beating demand sensing application.
Will they succeed? Watch this space.
Malcolm Wheatley, a freelance writer and contributing editor at Data Informed, is old enough to remember analyzing punched card datasets in batch mode, using SPSS on mainframes. He lives in Devon, England, and can be reached at email@example.com.
Home page photo of warehouse forklift operator by Flickr user toolstop.