Computer implemented method and system for demand forecast applications

• US 20030144855A1
• Filed: 01/30/2002
• Published: 07/31/2003
• Est. Priority Date: 01/30/2002
• Status: Abandoned Application

• Current Assignee: Oracle Technology Company (Oracle Corporation)
• Original Assignee: Oracle Technology Company (Oracle Corporation)
• Inventors: Cohen, Nir, Zvi, Amit Ben, Adoram, Meirav, Kaplan, Mervyn
• Application Number:
  US10/058,830
• Publication Number:
  US 20030144855A1
• Time in Patent Office: Days
• US Class Current: 705/1
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

Computer implemented method and system for multi-server computing of demand forecast information for a demand forecast tree to shorten run time.

Claims

1. Computer implemented method for computing demand forecast information for a demand forecast application capable of being graphically represented by a demand forecast tree having a single top level node with at least two branches directly emanating therefrom each having at least one node with a time series of observations associated therewith, the method comprising the steps of:

(a) providing a database for storing time series of observations;

(b) providing at least two computer servers each independently capable of computing demand forecast information for an entire branch of the demand forecast tree; and

(c) simultaneously computing demand forecast information for at least two branches of the demand forecast tree on two different computer servers of the at least two computer servers.

2. The method according to claim 1 and further comprising the steps of:

(d) allocating each and every branch of the demand forecast tree to a task including one or more branches for execution on a single computer server of the at least two computer servers where the same computer server computes the demand forecast information for an entire task.

3. The method according to claim 2 and further comprising the steps of

(e) determining the number of tasks as the product of the number of computer servers available for computing demand forecast information and a user entered value.

4. The method according to claim 2 wherein the total number of bottom level nodes of the branches of each task is substantially equal for all the tasks.

5. Computer implemented system for computing demand forecast information for a demand forecast application capable of being graphically represented by a demand forecast tree having a single top level node with at least two branches directly emanating therefrom each having at least one node with a time series of observations associated therewith, the system comprising:

(a) a database server for storing time series of observations; and

(b) a forecast engine including two or more computer servers each independently capable of computing demand forecast information for an entire branch of the demand forecast tree for simultaneously computing demand forecast information for at least two branches of the demand forecast tree on two different computer servers of the at least two computer servers.

6. The system according to claim 5 and further comprising a computer manager for allocating each and every branch allocating to a task including one or more branches for execution on a single computer server of the at least two computer servers where the same computer server computes the demand forecast information for an entire task.

7. The system according to claim 6 wherein the computer manager determines the number of tasks as the product of the number of computer servers available for computing demand forecast information and a user entered value.

8. The system according to claim 7 wherein the total number of bottom level nodes of the branches of each task is substantially equal for all the tasks.