Planning and Forecasting 5 – The First Shall Be First

Not surprisingly, technology has a lot to do with defining best practices within IBM's supply chain planning processes. By integrating demand fulfillment capabilities with its enterprise resource planning (ERP) system, IBM can schedule an order throughout its supply chain within milliseconds, says Joe DiPrima, manager of supply chain planning and optimization with the ISC group. "That is a best practice because you can have the best planning tools m the world, but if you can't pull the data into the planning tools with integrity so that people trust the data and know that it's current, you're not going to use the planning tool," DiPrima observes. It took a while to convince people that the planning tools were accurate, he admits, but the sheer speed of the forecasts has won over those skeptics who were still relying on their trusty spreadsheets.

Enterprise resource planning (ERP) software ties together manufacturing, sales, distribution, and finance by collecting data from each area and using it to plan a company's resource use—everything from employees to raw materials.

IBM used to manually schedule orders, which became a problem when the company began to dread the arrival of unexpected orders. In normal circumstances, getting new business is good news, but IBM's visibility into its supply lines was less than ideal. There was a fear within some quarters that a new order would divert supply from a high-priority customer that hadn't actually placed its order yet but was expected to. "We didn't want to schedule a lower-priority customer in the hopes that a high-priority order would come in," DiPrima remembers.

To get past that mindset, IBM has done away with those manual processes and replaced them with new processes and new tools based on streamlining the order receipt to delivery time. In the past, order entry to delivery could take anywhere from 15 to 20 days; that process is now down to 5 to 10 days.

How did IBM pull that off? As DiPrima explains, the company instituted a business policy of first in, first out (FIFO). "Orders are now scheduled FIFO. If a customer wants supply, they need to get their orders in first. Very simple. We have exception processes that we invoke occasionally, but if a product is deemed to be FIFO—and over 95 percent of our products are FIFO—they're scheduled first in, first out."

Additionally, IBM has enabled direct shipment to customers from suppliers as they've gone global. "We've outsourced manufacturing to China, Eastern Europe, and Mexico," DiPrima observes, "and as a result, we've enabled these companies to direct ship on behalf of IBM. It looks like an identical order whether we ship it to the customer from our warehouse or whether the manufacturer ships it." This postponement strategy includes some subtle back-office processes such as enabling the outsourcers to print invoices with the IBM logo. The goal, DiPrima says, is to postpone the building of the product until an order is received from a customer.

"From a demand planning standpoint," he continues, "we used to have to be able to forecast each end item a customer would buy." That was no small task since IBM had tens of thousands of end items. "If a customer wanted to buy a standard ThinkPad, but with his corporate logo on the start-up screen, that was a new model number. So while we might only have 300 or 400 core models, it would turn into tens of thousands of models when we actually built them. We used to forecast demand that way, and it was extremely difficult to do. It was never accurate. We would always be chasing and remixing supply from what we had forecast to what actually got ordered."

IBM's solution was to move to a sales building block model, based on a best practice known as attach rate planning. "We have tens of thousands of components and tens of thousands of end items," DiPrima states, "but if you look at the sales building blocks, we only have several hundred to a couple thousand of those. So we find the pinch-point in the development of a product by asking: Where can I have the fewest planning items in the plan, not only because it's easier, but also because I'll get all the advantages of risk pooling by doing it at that level? So we went to a forecast attach rate approach."

IBM's forecasting accuracy at the sales building block level is 80 to 90 percent, a marked improvement from the 50 to 60 percent accuracy it had when it was planning at the end item level. "We always knew how many units in aggregate we would sell, but where we would get it wrong was in trying to figure out the mix," he says. "Now that we know what the percentage mixes are, the planning process is a lot simpler."

Another best practice at IBM has been moving from a monthly planning cycle to a weekly S&OP process. "We also have an ad hoc process running daily to share our demands, including orders, with our suppliers via the web, so they can respond back to us with their capabilities every day," DiPrima explains. "We used to only share that information with a supplier once a week. Now they see it every day, which is critical when you're trying to bring your order and delivery cycle times down below 10 days. We're a lot more collaborative today with our suppliers. Our supply chain is not limited to what happens within the four walls of manufacturing, or even inside of IBM. We extend it out to our suppliers, and even our suppliers' suppliers, so we can have Tier 2 visibility as well."

A HAPPY ENDING

Improving its supply chain visibility has proven to be the key to Cisco Systems' rebound from its forecasting nightmares, which were described at the beginning of these articles. The company's turnaround began with a dramatic paring back of suppliers (from 1,300 down to 600) and the concurrent outsourcing of logistics, subassembly manufacturing, and materials management. All suppliers and distributors can now tap into the same supply chain network, dubbed eHub, and as a result everybody has access to the same forecasts and is working off the same demand assumptions.

Not only does eHub save Cisco millions of dollars by eliminating paper-based purchase orders and invoices, but it also has improved on-time shipment performance. And by applying "analytical rigor" to its supply chain plan, the company can make better decisions sooner in the process, such as what to do if a key supplier can't meet its commitments. By optimizing its supply chain plan, "we find you can remove emotions and bias from decision-making processes," explains Jim Miller, Cisco's vice president of manufacturing operations. "Supply chain has become a science now."