Sometime in 2011, long-range plans were being made deep inside Ace Hardware’s Oak Brook, Ill., headquarters. Brian Cronenwett, vice president of international logistics, and his team were busy working out plans to revamp the company’s international distribution network. The project involved creating three far-flung distribution centers to service more stores in 62 countries.
“We had a major distribution center in Shanghai that we repurposed into a regional, then added centers in Panama and Dubai within six months,” Cronenwett says. “Needless to say, I didn’t see much of my family during that year.”
The goal of the project was to smooth out kinks in a supply chain that circled the globe and serviced more than 600 franchised stores outside the United States. “I had moved over from our U.S. side and I quickly got a feel for the issues of operating this type of business in other parts of the world,” Cronenwett says. “It requires a great deal of patience and an ability to think quickly on your feet.”
Managing product demand
Regulations regarding shipping within the United States are usually enacted after plenty of discussions; affected businesses know what’s coming up and have developed plans to conform to them. In other countries, however, rules can be more arbitrary.
“We might wake up and find there’s a significant change in customs that makes a SKU non-feasible,” Cronenwett says. “It can create volatility if you’re not prepared.”
One of the biggest issues was managing product demand. Differing seasons and cultures played a part, as did the economic conditions in individual countries.
“When we would look at demand history, these variations would create a great deal of noise, much more than you’d see in a U.S. distribution network,” he says. “The tools we were using to figure out demand patterns were somewhat hit or miss. It was obvious we needed to get a better handle on our forecasting.”
After working for a few years with supply chain software improvements, Ace found that keeping up was difficult. Deciding to start fresh in 2016, the company sought out cloud-based supply chain planner Blue Ridge.
“Like a number of retailers in their situation, they were sharing data among different systems and trying to analyze it, which led to some problems with forecasting,” says Greg White, CEO of Blue Ridge. “Keeping the right products in stock in stores around the world at the right time is one of the biggest challenges for supply chain management, and you need forecasting that can keep up.”
The ability to accurately predict demand pattern spikes and dips has become critical for retailers with international operations, and increasingly they’ve sought out cloud-based solutions. “A cloud system can give you that kind of 360-degree overview,” White says. “You’re able to get a forecast based on overall retail trends, as well as weather and country conditions. It’s hard to do that with a system that’s based on just your own company’s sales history.”
Supply chain management is basically about preventing both overstocks and understocks, a daily challenge for any retailer, let alone one with global operations.
“To get product to many of our stores internationally requires shipments that can take weeks or months. That can put a wrench in your forecasts.”Brian Cronenwett
“The complexity of shipping overseas is where you see much of your issues,” Cronenwett says. “In a U.S. network each store is within days or hours of a distribution center. To get product to many of our stores internationally requires shipments that can take weeks or months. That can put a wrench in your forecasts.”
“When you’re thinking about stocking, you’re looking at how much you need but you also have to figure in the shipping restrictions,” White says. “How many of the products are in a box? How many boxes fill a pallet? You have to fill a shipping container, so will you have enough of product A, or what other products should be added to the shipment?”
The method Ace used for monitoring demand and lead times involved collecting data from one system, moving it to another, then manually configuring it into a spreadsheet. “It created several points where we could have errors,” Cronenwett says.
“It was not only time consuming, but the manual process of reviewing and making decisions based on the data we were collecting could be risky. We didn’t have that universal view.”
In early 2016, Blue Ridge sent a four-man implementation team to Ace’s headquarters in Illinois. “Because it’s cloud-based, we’re not onsite with a lot of people creating a bunch of custom code or building interfaces and workflows,” White says. “It’s more configured than customized. We take the data and feed it into our system.”
Working with the Blue Ridge system requires changing how demand forecasting is viewed. “Instead of entering that you need this many units or that number of days of safety stock, you say you want a 99.8 fill rate and the system shows what needs to be done to achieve that,” White says.
“That kind of goal-oriented nature of the system’s automation gives retailers a big advantage. You’re not spending hours calculating what would make the best and most efficient order.”
“Past performance is no indication of future results” is a common axiom about the stock market, but it also applies to demand forecasting. “You can’t just take some snapshots of what the demand history has been for a product and assume that’s going to be the case tomorrow,” White says.
“You’ve got to look at the things that affect demand and see how they are changing. A forecast might tell you how many units of a product you’ll need, but it doesn’t necessarily tell you how many will fill a shipping container, and that’s real information retailers need.”
In addition to manufacturing and shipping dates, the Blue Ridge system takes weather, national holidays and other factors into account when calculating its forecasts. “These are details that are a huge factor over time,” Cronenwett says. “It can make the difference in getting seasonal products on the shelves when you need them to be there.”
Ace is closing in on its goal of 40 percent fewer stockouts a year after implementing Blue Ridge’s system.
Ace’s biggest goal was having 40 percent fewer stockouts a year after implementing Blue Ridge’s system; Cronenwett says the company is closing in on that figure.
“It’s a fairly significant improvement,” he says. “Normally when you roll out new software, you have that period where you think, ‘This isn’t going the way I thought it would.’ We haven’t had that with this project. It’s been a smooth transition.”
The future of demand forecasting for Ace is focused on incorporating big data analytics into supply chain logistics.
“Eventually, the focus of our attention will be on the individual consumer more than the product in demand planning,” White says. “We’ll have a better view of customers for each item and that will sharpen our analysis and make it more exact.”