System for using subsets of rules applied to a database for updating and generating the rule knowledge base and forecasts of system demand

• US 5,446,890 A
• Filed: 10/25/1994
• Issued: 08/29/1995
• Est. Priority Date: 11/27/1991
• Status: Expired due to Term

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Anderson, Douglas A., Boliere, Elaine R., Renslo, Erland, Collins, Clayton M., Harker, Elizabeth M., Gardner, Kimberly
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): AMSBURY, WAYNE P
• Application Number:
  US08/328,934
• Time in Patent Office: 308 Days
• Field of Search: 395/600, 395/20, 364/408
• US Class Current: 1/1
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A forecasting system predicts product demand. A database stores information related to the volume of orders for a plurality of products. A knowledge base stores rules relevant to the products. A single program interface accesses the database and the knowledge base, including access screen means for monitoring and for updating the database and the knowledge base, the single program interface also including access screen means for generating forecasts of the future product demand for the products.

Claims

1. A forecasting system for predicting demand for products produced by a manufacturing line, the forecasting system comprising:

a database for storing information related to the volume of orders for products produced by the manufacturing line, wherein there is a plurality of types of products produced by the manufacturing line;

a knowledge base for storing rules relevant to forecasting demand for the products produced by the manufacturing line;

a test knowledge base for storing proposed rules for forecasting demand for the products produced by the manufacturing line; and

a program interface, coupled to the data base, the knowledge base and the test knowledge base, for providing a user interface to the database, the knowledge base and the test knowledge base, the program interface including

generating means, coupled to the data base and the knowledge base, for generating forecasts of the future product demand for the products based on the rules stored in the knowledge base and the information stored in the data base,

displaying means, coupled to the generating means, for displaying the forecasts generating by the generating means,

simulation means for allowing generation of test forecasts using the proposed rules stored in the test knowledge base operating upon the information stored in the database, and

updating means, coupled to the data base and the knowledge base, for updating the database and for updating the rules stored in the knowledge base, wherein updating the rules results in changing the forecasts generated by the generating means.

2. The forecasting system of claim 1 wherein the database includes storage means for storing forecasts made for future product demand and for storing the related product data used to make such forecasts.

3. The forecasting system of claim 1 wherein the knowledge base stores standard rules used by the generating means to generate forecasts of the future products, and the test knowledge base stores simulation rules which are used to simulate forecasting scenarios.

4. A method of using a knowledge base to forecast product demand for products produced in a manufacturing environment, including the steps of:

(a) collecting in a computer database, sales volume information for the products;

(b) generating a reference forecast for the product demand using the sales volume information for the products collected in step (a) and using a first set of rules within a knowledge base;

(c) storing the reference forecast; and

,

(d) generating a test forecast for the product demand using the sales volume information for the products collected in step (a) and used in step (b) to generate the reference forecast, the test forecast being generated using a second set of rules within the knowledge base.

5. A method as in claim 4 additionally comprising the step of:

(e) comparing the reference forecast and the test forecast to actual product orders received in order to determine which of the reference forecast and the test forecast was a more accurate forecast.

6. A method as in claim 5 additionally comprising the step of:

(f) when the comparison in step (e) indicates that the test forecast was the more accurate forecast, using the second set of rules to generate future reference forecasts.

7. A method as in claim 4 additionally comprising the steps of:

(e) storing the test forecast; and

,

(f) comparing the reference forecast and the test forecast to actual product orders received in order to determine which of the reference forecast and the test forecast was a more accurate forecast.

8. A method as in claim 7 additionally comprising the step of:

(g) when the comparison in step (f) indicates that the test forecast was the more accurate forecast, using the second set of rules to generate future reference forecasts.