I had the pleasure of attending JDA Software’s user conference late last month. I found a few user presentations to be especially informative. Perhaps most informative was a presentation by David Dornseif of Altera on the building of a customer-driven supply chain at his firm. I considered this presentation to be exceptional due to the logical and structured process Altera developed to better align its operations with a changing and more demanding operating environment.
Altera is a leading supplier of programmable logic semiconductors. Over the years, Altera’s product portfolio expanded, product complexity and manufacturing cycle time increased, and customers’ lead times became more demanding. Altera responded to these market pressures by embarking on a supply chain reengineering initiative that included improvements to its internal S&OP process, increased forecast collaboration with key customers, and tighter collaboration with suppliers. The increased sophistication of these processes required high quality data, robust supply chain software systems, and improved processes within the company. Perhaps most interesting is the well-structured process Altera used to determine business segments that ideally support more tailored supply chain efforts. The company analyzed numerous dimensions by which it could segment its business. It considered criteria such as geography, customer type, product, manufacturing process, and profitability. The company decided to categorize by Customer, Product, and Risk – dimensions that would best enable the alignment of business processes with segments to arrive at the desired operating improvements.
Altera established four categories of customers. Category A included Altera’s strategic, high volume customers. On the other end of the spectrum, Category D included customers such as hobbyists and academics that made infrequent, low volume purchases. Customers in Categories B and C had qualities or attributes somewhere in the middle. Altera then cross-referenced these customer categories with relevant business functions and determined the most appropriate processes for each combination. For example, Altera determined that engaging in collaborative forecasting at frequent intervals with strategic accounts would enable it to better anticipate unexpected demand swings between forecast cycles. In contrast, Altera concluded that aggregate statistical forecasting would be effective for estimating Category D customer demand. Similarly, Altera saw potential benefits to allocating work in process (WIP) to strategic customers and holding dedicated safety stock for them as well. Meanwhile, these cautionary steps were not considered ideal for Category D customers. Similar trade-off analysis was applied to efforts such as customization and order promising. Of course, the decisions were also based on the relative importance of these functions to the business relationships.
The proliferation of products within Altera’s portfolio was a contributing factor to the increase in the complexity of its operations. Altera therefore decided to consider segmenting its business by product category as well as customer group. Important product considerations included factors such as yield management, manufacturer relationship management, cost/benefit of forecast accuracy, and inventory buffer policies. Altera ultimately broke its products into four main segments. For each product segment, the company determined appropriate business process rules to maximize product availability and cost savings. For example, Altera applied substantial oversight of contract manufacturing efficiencies, detailed demand forecasting, and strict yield management to its high-cost, high-density products. In contrast, Altera relaxed its monitoring of manufacturing efficiencies, yield input, and forecast accuracies for less complex products, but increased safety stock allowances to meet demand fluctuations for these products.
Finally, Altera overlaid a risk management element to its segmentation process. To reduce risk, Altera implemented a postponement strategy by holding a product at the die bank for as long as possible. This provided the flexibility to utilize work in process for the greatest range of possible end products. For high risk products they also planned for back-up manufacturing in case primary sites went off line. Altera’s intelligent segmentation of its business across these three dimensions has enabled it to operate more efficiently in an increasingly demanding and complex market environment.