Demand forecast system and method
AssigneeTaiwan Semiconductor Manufacturing Company, Ltd.
InventorYuan-Fu Liao
Life of this patent
Estimating the term…
Abstract
A demand forecast system and method. A planning module normalizes historical forecast data, and generates a forecast error interval according to the historical forecast data and a Grey forecast model. An adjustment module adjusts a target forecast value of a target-planning bucket according to the forecast error interval and/or the distance between the forecast error interval and a benchmark.
Claims7
A computer-implemented demand forecast method for use in a computer, wherein the computer is programmed to perform the steps of inputting historical forecast data of a target planning bucket to a Grey forecast model to calculate two forecast values for the target planning bucket;
- forming a forecast error interval according to the two forecast values by the computer;
- adjusting a target forecast value of the target planning bucket by the computer according to a distance between the forecast error interval and a benchmark, wherein the target forecast value is adjusted according to
- ATFV=TFV×(1−(s1×FEI))×(1−(s2×D)) when the forecast error interval is above the benchmark, and is adjusted according
- to ATFV=TFV×(1+(s1×FEI))×(1+(s2×D)) when the forecast error interval is under the benchmark,
- wherein TFV is the target forecast value, ATFV is the adjusted target forecast value, FEI is the forecast error interval, D is the distance between the forecast error interval and the benchmark, s1 and s2 are adjustable constants.
A machine-readable storage medium comprising a computer program, which, when executed, causes a computer to perform a demand forecast method, the method comprising:
- inputting historical forecast data of a target planning bucket to a Grey forecast model to calculate two forecast values for the target planning bucket;
- forming a forecast error interval according to the two forecast values;
- adjusting a target forecast value of the target planning bucket according to a distance between the forecast error interval and a benchmark, wherein the target forecast value is adjusted according to
- ATFV=TFV×(1−(s1×FEI))×(1−(s2×D)) when the forecast error interval is above the benchmark,
- wherein TFV is the target forecast value, ATFV is the adjusted target forecast value, FEI is the forecast error interval, D is the distance between the forecast error interval and the benchmark, s1 and s2 are adjustable constants, and the demand forecast method is used to forecast a supply chain which supports material purchase, fabrication of materials into intermediate and finished products, and distribution of finished products to clients.
Dependencies are inferred from the claim text.