Unlock instant, AI-driven research and patent intelligence for your innovation.

Demand forecasting using weighted mixed machine learning models

A technology of demand forecasting and machine learning, applied in the field of computer systems, which can solve problems such as difficulty in adapting to market conditions, inaccurate forecasting, and forecasting methods not working very well

Pending Publication Date: 2020-06-16
ORACLE INT CORP
View PDF8 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this technique can be difficult to adapt to changing market conditions and can lead to inaccurate forecasts
Also, with more and more factors to consider, traditional forecasting methods such as time series analysis or regression do not work very well

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Demand forecasting using weighted mixed machine learning models
  • Demand forecasting using weighted mixed machine learning models
  • Demand forecasting using weighted mixed machine learning models

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0011] One embodiment generates multiple trained models by training multiple algorithms / methods and multiple features using historical sales data input, and then weighting each trained model based on an error value to predict demand for a product. Using weights and multiple models, a demand forecast is generated by combining the weighted forecasts generated by each trained model.

[0012] Sales and demand forecasting methods can be roughly divided into judgment methods, extrapolation methods, and causal methods. Extrapolation methods use only the time-series data from the activity itself to generate forecasts. Known specific algorithms / methods range from the simpler moving average and exponential smoothing methods to the more complex Box-Jenkins methods. While these known methods successfully identify and extrapolate trend, seasonality, and autocorrelated time series patterns, they do not account for external factors such as price changes and promotions.

[0013] Vector auto...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Embodiments forecast demand of an item by receiving historical sales data for the item for a plurality of past time periods including a plurality of features that define one or more feature sets. Embodiments use the feature sets as inputs to one or more different algorithms to generate a plurality of different models. Embodiments train each of the different models. Embodiments use each of the trained models to generate a plurality of past demand forecasts for each of some or all of the past time periods. Embodiments determine a root-mean-square error (RMSE) for each of the past demand forecasts and, based on the RMSE, determine a weight for each of the trained models and normalize each weight. Embodiments then generate a final demand forecast for the item for each future time period by combining a weighted value for each trained model.

Description

technical field [0001] One embodiment relates generally to computer systems, and in particular to computer systems that forecast demand. Background technique [0002] Products are typically delivered to consumers through a network of manufacturers, distributors, transporters, retailers, and the like. Such a network of facilities that together deliver products to consumers is often referred to as a "supply chain" network. [0003] Suppliers of products (eg, manufacturers, suppliers, retailers, etc.) are often faced with the task of forecasting product demand in order to provide a smooth and efficient flow of products through a supply chain network in the presence of changing market conditions. Overestimating demand can lead to overproduction and increased costs associated with holding inventory (eg, storage costs, obsolescence, etc.). On the other hand, underestimating demand results in lost revenue. [0004] Also, in the retail industry, retailers need to estimate their f...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/06G06Q30/02
CPCG06Q10/0631G06Q30/0202G06Q30/0201G06N20/00
Inventor 雷明C·波佩斯库
Owner ORACLE INT CORP