A few weeks ago, I had a face-to-face conversation with an executive from a consumer electronics accessories manufacturer. Since January 2007 this company has implemented five Oracle supply chain solutions: Demantra Demand Management, Oracle Advanced Supply Chain Planning (ASCP), Global Order Promising, Oracle Inventory Optimization (IO), and the Oracle iSupplier Portal (note: Oracle is an ARC client). Considering the size of this company and the scope of its implementation, this rollout was amazingly fast.
The implementation of these products was driven by poor inventory visibility (the company has multiple, separately planned inventory organizations). Inventory turns were getting worse and there was a disconnect between what commodity managers had negotiated and the purchase orders (POs) placed by the factories. Other issues included poor accuracy in the company’s order date promising system and a manual and bureaucratic Sales and Operational Planning (S&OP) process that did not provide enough opportunity for analysis and learning.
In its current S&OP process, the company still comes up with a consensus supply and demand plan and then matches this with its financial plan. But now the demand plan provides some of the parameters required for its inventory optimization plans. The company’s planning horizon is 12 months with minimums (mins) and maximums (maxs) adjusted monthly. The min/max outputs from Oracle IO are dependent upon target service levels. Interestingly, this company lets sales set the service levels, but the IO solution tells the sales organization the cost impact of moving from, for example, a 95 percent service level to a 99 percent level. Some of the sales organization metrics that are also shared with operations include goals and objectives pertaining to excess and obsolete inventory that affect quarterly bonuses.
The company manages its supply plan every day to account for new sales orders and to allow its production schedules to adjust to cancelled/augmented orders and to changes in request dates. This data gets mirrored in its order promising solution and it allows the company to make reasonably accurate commitments to suppliers and customers (the company is able to ship 89 percent of its orders within 72 hours of the customer request date).
The company uses postponement as a key manufacturing strategy. A single work-in-process (WIP) component can turn into as many as six different finished goods products. Postponement is also important in allowing the company to compete on service while running an efficient supply chain.
An important source of savings for the company has been the iSupplier portal. The procurement team creates blanket agreements that define, by supplier, the price for an item. The iSupplier portal helps to ensure there are no overrides to those agreements. If the supplier accepts an order, the system locks in the agreed to ship date and the predefined price. Suppliers, in turn, get visibility to their purchase orders and collaborative forecast information. The combination of iSupplier and IO min/max targets has allowed the company to switch many of its suppliers to Vendor Managed Inventory (VMI) programs. In the future, the company hopes to transition many key VMI suppliers to a pure consignment model (i.e. the suppliers won’t get paid until their inventory is consumed).
Implementing supply chain solutions immediately before a recession can make it difficult to calculate their ROI. In this case, however, these solutions helped the company to weather the global economic downturn effectively. When the recession hit, the company implemented daily inventory planning meetings. In December 2008, it told all of its suppliers to stop all shipments for four weeks until it had a chance to recalibrate its demand plan. The company also moved quickly to resize the company. As a result of these actions and its new supply chain applications, the company was able to come within five percent of its financial plan every quarter, even during the worst of the downturn. The company’s stock price, like many others, tumbled with the recession. But it has now more than tripled since its low point. Wall Street has treated this company better than many of its peers.
I ended our conversation by asking this executive what advice he would give to others contemplating a SCP implementation. Here is his advice:
- Don’t underestimate the importance of data maintenance! His company created alerts to help keep its planning data accurate and complete. For example, an alert is sent to a buyer to make a correction if a report shows that an item has demand but no preferred sourcing site, or if there is an SKU with demand but no planner.
- Follow the software vendor’s default process flows to the maximum extent possible! If managers want to diverge from the default process in implementation, make them provide a very strong justification for that divergence.
- Come back a year after the implementation and look at the lead time and time fence assumptions you made! Once you are a year smarter, many of those assumptions will be proved wrong.