The Supply Side: Harnessing Big Data for supply chain efficiency

Big Data feeds analytics well beyond point of sales data that used to be the be-all, end-all in retail forecasting. Today, retailers and suppliers are tapping “Big Data,” which encompasses market loyalty, store and site traffic, social media and mobile shopping in addition to point-of-sale and marketshare data.

Fred Baumann, vice president at JDA Software Group, recently spoke at the University of Arkansas Supply Chain Conference about how advanced analytics fed by Big Data are changing the supply chain. He said tapping smart analytics are the key to solving out-of-stock issues at retailers, because by leveraging information across social media more real-time demand can be accessed.

Baumann said category managers have the opportunity to integrate relevant real time data into their forecasting. He said advanced analytics also may improve assortments and space planning for the retail store. He said analytics within the supply chain have to be a connected collaborative framework between retailer and supplier.

“Manufacturers and retailers struggle to try and solve their problems themselves, when they both want the same thing — productive shelves,” Baumann said.

Forecasting inventory to the shelf is the future, according to Baumann.

“Planing shipments to the shelf creates a ton more data than shipping to the distribution center. It is very granular and requires an interconnected collaborative framework with total transparency,” he said.

REAL TIME DEMAND
Ryan Frazier, CEO of DataRank in Bentonville, is a Big Data analytic service firm for retailers and suppliers. He said many suppliers are using analytics gleaned from Big Data to better understand real time demand.

Frazier said the present system used in retail to track product from the warehouse to the shelf is largely managed by employees using a paper trail. He said companies like Movista have sprung up to offer better solutions by helping retailers and suppliers save money on implementing efficient data collection and delivery practices.

The Big Data evolution has three distinct stages at this point, according to Frazier. He said it’s a very broad term and analytics can help with visualization and grasping the relevant information. He said knowing what is occurring is step one. Next, he said Big Data is typically used in forecasting demand with predictive and proactive analytics. The final stage is working with different “levers” to manipulate and make changes in real time data to obtain desired results.

Frazier and Baumann agreed that it’s important to find the right partners to work with in the Big Data space because being able to make changes in real time requires fast analysis. Frazier said moving from step two to step three is cumbersome to do alone.

Going it alone can be cost prohibitive given the talent needed in terms of engineers and hardware servers to process the data. Frazier said as searches against the data are needed, high powered servers and tech savvy are essential.

TRANSFORMATIONAL CHANGES
Dhiraj Rajaram, CEO of Mu Sigma, also was one of the featured speakers at the recent supply chain conference. Rajaram said the information age requires one to think in terms of the decision supply chain, moving away from the physical supply chain.

Rajaram defined Big Data by saying it is data engineering, data science with applied technology and math then layered with business context to create a stack. 

“Big Data will both drive and require transformation from businesses,” he said.

Keith Mercier, global retail leader at IBM Watson, said the impact Big Data and social media are having on retail is transformational. He said Pinterest, a four-year old company that is a virtual bulletin board, has reshaped the fashion merchandising sector. A case in point is Nordstrom, he said.

“They know merchandising, as a 100 year-old company, and Nordstrom is modeling its own merchandising after a four-year social media site (Pinterest) with its Pinned Shoes campaign,” Mercier said.

Nordstrom features the shoes on its site in order of popularity noted by the number of pins or likes gleaned from crowd sourcing.