Data processing: financial – business practice – management – or co – Automated electrical financial or business practice or... – Discount or incentive
Reexamination Certificate
2000-05-04
2004-11-09
Hafiz, Tariq R. (Department: 3623)
Data processing: financial, business practice, management, or co
Automated electrical financial or business practice or...
Discount or incentive
C705S007380, C705S028000, C705S029000
Reexamination Certificate
active
06816839
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to a method for forecasting the demand of products with multiple options, to be used for establishing an efficient supply chain management framework by providing an accurate demand forecast of each attachment.
BACKGROUND OF THE INVENTION
New technology advance, rapid market condition changes, and growing customer demands have dramatically changed the landscape of the electronic industry. While product life cycles get shorter, prices continue to drop, and profit margins become paper-thin, customers are demanding faster and more reliable delivery of highly customized products. On the supply side, a globalized market, coupled with long component lead times, only adds to the challenges. In order to succeed in this complex environment, companies have to improve their speed, efficiency, and customer experiences by exploring different business models and increasing supply chain flexibility. This has resulted in recent trends in direct market business model, global information technology systems, mass customization, configure-to-order (CTO) operation, supply chain collaboration, and end-to-end supply chain optimization. In the new economy, a supply chain can be viewed as a network of enterprises linked either loosely or tightly together to produce and move the goods through the pipeline to reach customers. By coordinating and managing the flow of information, material, and control in a supply chain, total supply chain uncertainty can be reduced dramatically. Reports have shown that leading companies with best-in-class supply chain management can have a 7% cost advantage, a 40% to 65% advantage in cash-to-cash cycle time, 50% to 80% inventory reduction, and at the same time, achieve better customer service level and satisfaction.
SUMMARY OF THE INVENTION
A critical aspect in the supply chain uncertainty reduction is the demand forecast. The bull-ship effect examined by Lee et. al. demonstrated how forecast errors were amplified through the supply chain and caused supply chain inefficiency. Studies by Ett et. al. and Lin et. al. also showed the significant impacts of the forecasting accuracy to the overall supply chain performance. It is clear that forecasting should be an integral part of the supply chain management.
The electronic industry traditionally has adopted a build-to-forecast process. The process is sometimes referred to as the “machine-type model” (MTM) operation. In this process, there is a set of predefined end products, or MTM's. Demand forecasts over a future planning horizon are generated for each MTM, and updated periodically for each planning cycle (this is typically a weekly cycle). A “materials requirements planning” (MRP) type explosion technique is then used to determine the requirements for the components over the planning horizon, based on the bill-of-materials structure of each end product. The entire supply chain planning and manufacturing plan are based on the forecasts and the interactions between the forecast. The supply chain operations are limited to the periodical forecast adjustments. Products are then built according to the forecast. Because of the random variation and unavoidable error involved in demand forecasts, excessive inventory is usually kept at the end product level. However, holding finished goods inventory for any length of time is very costly for products with a short life cycle and a frequent price reduction. This type of MTM operation is being migrated to a web-based configure-to-order operation where customer orders will be taken from the internet. Instead of a set of predefined machine type models, customers can configure the products with multiple options over the web based tool on a set of components or Building Blocks (BB) without violating the technological constraints. This model offers higher product variety, and hence, can result in a better customer satisfaction, broader market coverage, and higher demand volume. There is also a potential revenue and profit benefit from the increased market coverage and the leverage of profit multipliers. In the CTO/BB environment, no finished goods inventory will be kept and the inventory management focus is shifted to the components of “building blocks”, which are still built based on the build-to-forecast scheme, due to their long lead-times. The number of different finished products based on the feasible combinations of the building blocks is quite large. It is very difficult, if not impossible, to forecast well the demands for each feasible combination of the building blocks as it is done in the simpler MTM environment. In this invention, we present a method to accurately forecast the demand of building blocks, in a CTO/BB environment, to be used for establishing efficient supply chain management.
Accordingly, in a first aspect of the present invention, we disclose a method for demand planning of products, the method comprising the steps of:
1) constructing a configure-to-order operation/multiple building block environment; and
2) forecasting the demand of the building blocks within this environment for establishing an efficient supply chain management.
Preferably, the demand for forecasting building blocks comprises a step of focusing on the sales volume of each building block.
Preferably, the demand for forecasting building blocks comprises the step of considering a sales volume of a product family and the feature ratios of the building blocks. For example, step 2 may comprise normalizing feature ratios within a building block family. Preferably, step 2 comprises using dependencies among building block families comprising a pairwise correlation between any two building block families. For example, this may comprise a step of approximating the pairwise correlation relationship with a spanning tree. The method, in this regard, may advantageously comprise a step of propagating an adjustment by tracing the spanning tree by starting from the most dominant building block family. For example, the method usefully includes the steps of using mutual information as an index for the pairwise correlation and fitting a joint distribution for the building blocks to be consistent with the pairwise correlation.
In a second aspect of the present invention, we disclose a program storage device readable by a machine, tangibly embodying a program of instructions excutable by the machine to perform method steps for demand planning of products, the method comprising the steps of:
1) constructing a configure-to-order operation/multiple building block environment; and
2) forecasting the demand of the building blocks within this environment for establishing an efficient supply chain management.
With respect to the invention as just summarized in both its first and second aspects, we now provide the following additional instruction.
There are two approaches for forecasting the demands of building blocks. One is the plain approach, which just focuses on the sales volume of each building block, regards it as a scalar value independent of other values, and applies a statistical forecast to it. The other is the top-down approach, which consider the sales volume of the product family and the feature ratios of the building blocks. The sales volume of a building block is obtained by multiplying the total sales volume and the ratio corresponding to the building block. In this case, we need to forecast the sales volume of the product family and the feature ratios for the building blocks.
On the other hand, there are dependencies among the choices of building blocks from multiple building block families, for example, a customer who selects a fast CPU tends to select a large memory module. These have never been addressed so far within the two approaches above, because existing statistical forecasting technologies mainly focus on one scalar value. Further, the forecasted demand is not always accepted as it is, rather adjusted afterwards based on external constraints including price change, inventory status, and competitors performance. Here the dependencies are not sufficien
Gung Roger R.
Hosking Jonathan R.
Lin Grace Y.
Tajima Akira
Boyce Andre
Hafiz Tariq R.
Kaufman Stephen C.
Whitham Curtis & Christofferson, P.C.
LandOfFree
Demand planning for configure-to-order and building... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Demand planning for configure-to-order and building..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Demand planning for configure-to-order and building... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3359339