System and method of solving optimization problems using prestored advanced bases

• US 7,424,451 B1
• Filed: 10/10/2000
• Issued: 09/09/2008
• Est. Priority Date: 10/10/2000
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kraemer, James Richard, Moore, Bruce Wayne
• Primary Examiner(s): Poinvil; Frantzy
• Assistant Examiner(s): GRAHAM, CLEMENT B
• Application Number:
  US09/685,838
• Time in Patent Office: 2,891 Days
• Field of Search: 705/36, 705/7, 705/37, 705/35, 705/38, 703/6, 706/13, 707/2, 707/4, 395/674
• US Class Current: 705/37
• CPC Class Codes: G06Q 10/02 Reservations, e.g. for tickets, services or events
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

A system, computer code, and method are disclosed for solving optimization problems using prestored data groups pertaining to a plurality of anticipated optimization problems, whereby optimization problems can be solved much faster with fewer iterations. The method may comprise storing a plurality of data groups each associated with one of a plurality of anticipated optimization problems, each of the data groups including optimal solutions to a corresponding anticipated optimization problem; and solving the current optimization problem using the stored data groups.

Claims

1. A computer-implemented method for solving a current financial portfolio optimization problem inputted by a user of a computer system comprising the steps of:

storing, on a computer, a plurality of data groups each associated with one of a plurality of anticipated financial portfolio optimization problems, each of the data groups including solutions to a corresponding anticipated financial portfolio optimization problem, each of the data groups further including input values and intermediate calculation values associated with the corresponding anticipated financial portfolio optimization problem, wherein said storing occurs before said current financial portfolio optimization problem is input;

compiling, using said computer, said solutions to produce a set of results from said solving anticipated financial portfolio optimization problems;

preparing and storing, on said computer, a plurality of look-up tables for identifying each of said results, the plurality of look-up tables containing equation names, RHS (Right Hand Side) values, and objective values pertaining to the plurality of anticipated financial portfolio optimization problems;

solving, using said computer, the current financial portfolio optimization problem using the stored results from said solved anticipated financial portfolio optimization problems, the solving step including the steps of;

selecting, using user-defined functions, at least one of the stored results using the look-up tables; and

determining whether or not the selected result contains solutions to the current financial

portfolio optimization problem;

wherein, if the determining step determines that the selected result contains solutions to the current financial portfolio optimization problem, then the optimal solutions included in the selected result are output as optimal solutions to the current financial portfolio optimization problem; and

wherein, if the determining step determines that the selected result does not contain solutions to the current financial portfolio optimization problem, then the selected result is modified using a search method, and the current financial portfolio optimization problem is iteratively solved using the modified data group to obtain solutions to the current problem.

2. A system for solving a current financial portfolio optimization problem inputted by a user of a computer system comprising:

a storage unit, in a computer, storing a plurality of data groups each associated with one of a plurality of anticipated financial portfolio optimization problems, each of the data groups including solutions to a corresponding anticipated financial portfolio optimization problem, each of the data groups further including input values and intermediate calculation values associated with the corresponding anticipated financial portfolio optimization problem,

wherein said storing occurs before said current financial portfolio optimization problem is input;

and an optimization unit in said computer, said optimization comprising;

means for compiling a set of results from said solving anticipated financial portfolio optimization problems;

means for preparing and storing a plurality of look-up tables for identifying each of said results, the plurality of look-up tables containing equation names, RHS (Right Hand Side) values, and objective values pertaining to the plurality of anticipated financial portfolio optimization problems;

means for solving the current financial portfolio optimization problem using the stored results from said solved anticipated financial portfolio optimization problems, the solving means including;

means for selecting, using user-defined functions, at least one of the results using the look-up tables; and

means for determining whether or not the selected result contains solutions to the current financial portfolio optimization problem;

wherein, if the determining means determines that the selected result contains solutions to the current financial portfolio optimization problem, then the solutions included in the selected result are output as solutions to the current financial portfolio optimization problem; and

wherein, if the determining means determines that the selected result does not contain solutions to the current financial portfolio optimization problem, then the selected result is modified using a search method, and the current financial portfolio optimization problem is iteratively solved using the modified data group to obtain solutions to the current problem.