Manufacturers often face difficulties balancing inventory with the needs of their customers in the most profitable manner possible. For example, many manufacturers are often faced with out-of-stock situations where they don't have enough inventory to meet customer demands, which could cause great customer dissatisfaction.

The best way for manufacturers to solve this problem is to integrate their supply chain management (SCM) and advanced planning and scheduling (APS) applications with a central data warehouse that can be accessed in real time and used for more accurate forecasting.

By leveraging a central data warehouse, manufacturers can optimize their demand and supply chains, allowing them to become more responsive to constantly changing demand and supply conditions.

Inventory highs and lows
APS, a key component of SCM, is a manufacturing management process by which raw materials and production capacity are optimally allocated to meet demand. Because it can analyze large amounts of data, APS is especially well-suited to environments where simpler planning methods cannot adequately address complex trade-offs among competing priorities.

Traditional planning and scheduling systems, such as manufacturing resource planning, use a stepwise procedure to allocate material and production capacity. This approach is simple but cumbersome, and it does not readily adapt to changes in demand, resource capacity or material availability. Materials and capacity are planned separately, and many systems do not consider limited material availability or capacity constraints. Thus, this approach often results in plans that cannot be executed.

Unlike prior systems, APS simultaneously plans and schedules production based on available materials and capacity. APS software enables manufacturing scheduling and advanced scheduling optimization within the following environments or product types:

	Products that require a large number of components or manufacturing tasks
	Inventory that necessitates frequent schedule changes that cannot be predicted before an event
	Capital-intensive processes, where plant capacity is constrained
	Facilities in which multiple types of inventory are produced

Demand planning is a key function that starts the APS process. It develops a detailed estimate of future demand, monitors and improves the accuracy of this estimate, and moves the forecast to the planning system. The planning system then presents a highly accurate supply forecast. These steps effectively drive the planning process and enable forecasters to easily and quickly update the operation to reflect changing business conditions.

Companies with optimum demand planning or demand forecasting perform better. However, for demand planning to work, manufacturers must have access to large amounts of data that relate to the amount of products that have been sold to their consumers. This data could include detailed point-of-sale (POS) data, wholesaler or other channel inventory and sales data, trade promotion data, customer forecasts, and one to three years of detailed sales history.

Most popular APS software systems on the market today, such as the mySAP Supply Chain Management application from SAP, offer business processes such as demand planning, supply planning, supply and demand matching, and production planning/detailed scheduling.

Up-to-the-minute developments
To optimize demand planning, manufacturers can implement an enterprise data warehouse (EDW). The EDW collects all of the relevant data and becomes the data source for all reporting and analytics related to the demand planning process. As its name suggests, the EDW is tightly integrated with the enterprise resource planning and forecasting system, as well as with the APS system.

The ability to access the EDW actively, or in real time, and monitor the planning system throughout the day offers greater value to the manufacturers. The theory is this: If users have the up-to-the-minute data available immediately, they can react more quickly and ultimately improve their inventory and customer service levels with more accurate planning and forecasting.

Unfortunately, many manufacturers do not have access to real-time data, as most advanced planning systems are run once a month or once a week. Additionally, the information gleaned is often inaccurate. Industry studies show that a typical forecast is off by 35%—meaning the manufacturer produces too much of a product (resulting in inventory overloads) or not enough (resulting in missed sales opportunities).

Demand plans, however, are improving, thanks to Teradata's active data warehouse technology. For example, at two multinational consumer product goods (CPG) manufacturers using Teradata, the actual demand feeds (such as POS data) are loaded into their APS systems and compared ad hoc with the predicted demand. If the actual demand does not match the forecast demand within a specified amount, an alert can be sent to the product planner and a new plan can be triggered.

So, why should manufacturers optimize demand planning by integrating it with an active data warehouse? Because doing so can improve forecast accuracy, which is correlated with better financial performance and better order fulfillment rates. But perhaps the biggest benefit to optimizing demand planning is balancing inventory with customer service levels in the most profitable manner possible. T

Benefits of integrating an enterprise data warehouse with a supply chain planning system Speed > Eliminates multi-hour wait times to analyze planning results > Enables what-if analysis, scenario modeling and forecast comparisons > Identifies discrepancies in time to make a change Detail > Uses detail-level point-of-sale (POS) and transaction data for improved accuracy > Performs highly intelligent aggregations and disaggregations > Enables insight into price elasticity, uplift effects, affinity and cannibalization > Analyzes planned versus actual by store/stock keeping unit (SKU)/day for optimized inventory and service History > Uses years of detailed data for accurate assumptions > Proves assumptions for seasonality, promotion uplift and event demand > Enables insights into product life cycle, new product introduction and end-of-life trends Frequency > Responds faster to unexpected deviations from plan > Triggers alerts or re-plans in real time > Delays production and distribution until promotion or event uplift is confirmed

Lisa Campbell is a freelance marketing and technology writer in New York.

Photograph by Corbis

Teradata Magazine-June 2007