Feed Energy Overcomes "Data Dump"

In spring 2011, the American Feed Industry Association and Feed & Grain magazine partnered to create the Information Technology Innovation Award, honoring the feed and grain industry’s leaders in finding technology-based solutions.

Feed Energy Company, a further refiner of vegetable oil as a high-energy, liquid feed ingredient, was named runner-up for its “Nth Load” entry. The Nth Load is a computational, Dynamic Risk-Based Modeling^™ tool that objectively analyzes large data sets to provide comprehensible information for making critical and timely operating decisions.

Supply management trouble

Alan Richardson, executive director for strategy and processes at Feed Energy, said the need for Nth Load grew as the company transitioned from a niche feed ingredient supplier to a leading provider of energy products to turkey, poultry, cattle and swine producers throughout the Midwest. Data overload combined with erratic spikes and dips of oversupply and shortages in raw material inventories caused frustration and headaches for leaders and staff at Feed Energy.

“As a smaller company, we’d have violent swings either filling our plants up with material or starving it,” said Richardson. “We’d always met our customers’ demands, but if we were going to achieve operations excellence and growth, we knew we needed to tame down the supply and demand teeter-totter.”

So in 2008, Robert G. Riley Jr. Feed Energy’s CEO, introduced Richardson and the senior leadership team to Spencer Parkinson, executive director, Decision Innovation Solutions (DIS), who develops proprietary tools to assist agricultural executives, business owners and producers.

“Everybody on the staff at DIS has a combination of agriculture and economics in their background,” says Parkinson. “We work with smaller companies that don’t have the need or resources for a full-time data analysis team, and we help them get their data into a format they can understand.”

From April 2008 through June 2009, the DIS team developed a a dynamic risk-based model to fit Feed Energy’s business processes. The model projects a five-week planning horizon using the previous 140 weeks’ historical data combined with current sales volume, prices, ingredient inventory levels, contracted purchases, anticipated demand and supply, and other variable inputs and constraints from Feed Energy’s three locations. Feed Energy began using the model’s projections in Q3 2009.

Quick results

The benefits of Nth Load quickly revealed themselves to Feed Energy. Within the first year of implementing the model, Feed Energy increased its annual average pounds of product sold per week by 7%.

Additionally, annual total sales pounds increased by 18%.

Bringing leadership together

Perhaps an unintended benefit of using the Nth Load was increased communication within the company’s leadership.

“The Nth Load doesn’t only help us understand the interaction of data better,” says Richardson, “but it’s also helped our team make better decisions because instead of merely two or three people making choices, we have an entire operations team collectively making decisions. Now we have a stronger, healthier team that relies on each other and is dependable.”

Feed Energy’s weekly Nth Load meetings are where the collective decision making takes place, bringing together leadership from each department — purchasing, supply, sales, production and financial — to discuss the model’s next five weeks’ projections.

“Every Friday afternoon we update the model to the extent we can, and on Monday [inventory day] we add the rest of the most current data available,” says Parkinson. “Then we send management a report prior to the meeting and address any questions beforehand, so during the meeting we aren’t discussing merely what the projections are, but what the plan is for the coming weeks.”

Based on the forecast, the team discusses plans for everything from ingredient procurement, to pricing adjustments and product formulation options.