Computer program for and method of discrete event computer simulation incorporating biological paradigm for providing optimized decision support

• US 20020095393A1
• Filed: 06/06/2001
• Published: 07/18/2002
• Est. Priority Date: 06/06/2000
• Status: Abandoned Application

• Current Assignee: Kansas State University Research Foundation
• Original Assignee: Kansas State University Research Foundation
• Inventors: McHaney, Roger
• Application Number:
  US09/875,058
• Publication Number:
  US 20020095393A1
• Time in Patent Office: Days
• US Class Current: 706/13
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

A computer program for and method of discrete event computer simulation incorporating a biological paradigm, wherein the simulation provides decision support and the biological paradigm contributes optimization of the simulation such that the invention is able to devise and test its own potential solutions to a given problem. The biological paradigm or genetic algorithm is based upon the mechanics of natural selection and population genetics, combining machine-learning with the principle of “survival of the fittest”, including such biological concepts such as reproduction, crossover, and mutation. Potential uses include a programming language preprocessor; a simulation language preprocessor; and an embedded component providing decision support for controlling a system or process.

Claims

1. A computer program for facilitating decision making, the computer program being stored in a computer-readable memory and executable by a computing device, the computer program comprising:

a code segment operable to perform discrete event computer simulation for testing one or more encoded potential solutions by modeling a system wherein any changes take place substantially instantaneously and in countable increments in response to discrete events; and

a code segment operable to provide a biological paradigm involving the creation, evaluation, and propagation of a population of the encoded potential solutions,

wherein the computer program is operable to both devise and test one or more of the encoded potential solutions.

2. The computer program as set forth in claim 1, wherein creating, evaluating, and propagating the population of encoded potential solutions involves:

a code segment for evaluating a fitness parameter of each of the encoded potential solutions using an objective fitness test;

a code segment for terminating any of the encoded potential solutions that fail the objective fitness test;

a code segment for propagating any of the remaining encoded potential solutions to produce a second population of encoded potential solutions.

3. The program as set forth in claim 2, wherein propagating any remaining encoded potential solutions involves—

combining two or more of the remaining encoded potential solutions to produce a single member of the second population of encoded potential solutions; and

introducing random change into the population of encoded second potential solutions by randomly changing the encoding of one or more encoded potential solutions.

4. The computer program as set forth in claim 1, wherein the computer program is used as a simulation language preprocessor.

5. The computer program as set forth in claim 1, wherein the computer program is used as an embedded component operable to provide decision support for controlling a system and a process.

6. A computer program for facilitating decision making, the computer program being stored in a computer-readable memory and executable by a computing device, the computer program being operable to perform discrete event computer simulation for testing one or more encoded potential solutions by modeling a system wherein any changes take place substantially instantaneously and in countable increments in response to discrete events, the computer program comprising:

a first code segment for creating a population of encoded first potential solutions;

a second code segment for performing an evaluation of a fitness level of each encoded first potential solution, and for producing a fitness result representative of the evaluation;

a third code segment for terminating one or more of the encoded first potential solutions based upon the fitness result;

a fourth code segment for propagating any remaining encoded first potential solutions to create a population of encoded second potential solutions, the code segment for propagating including—

a fourth_A code segment for combining two or more of the remaining encoded first potential solutions to produce a single member of the population of encoded second potential solutions, and

a fourth_B code segment for introducing random change into the population of encoded second potential solutions by randomly changing the encoding of one or more encoded second potential solutions; and

a fifth code segment for reporting one or more fittest encoded potential solutions.

7. The computer program as set forth in claim 6, wherein each of the encoded first and second potential solutions comprise one or more input parameters, with each input parameter being encoded in a data structure using one or more bits.

8. The computer program as set forth in claim 7, wherein the fourth_B code segment involves randomly flipping one or more bits of one or more encoded second potential solutions.

9. The computer program as set forth in claim 6, wherein the computer program is used as a simulation language preprocessor.

10. The computer program as set forth in claim 7, wherein the computer program is used as an embedded component operable to provide decision support for controlling a system and a process.

11. A method of facilitating decision making, the method comprising the steps of:

(a) creating a discrete event simulation for testing one or more encoded potential solutions by modeling a system wherein any changes take place substantially instantaneously and in countable increments in response to discrete events;

(b) incorporating a biological paradigm into the discrete event simulation;

(c) creating, evaluating, and propagating a population of the encoded potential solutions; and

(d) reporting one or more identified fittest potential solutions.

12. The method as set forth in claim 11, wherein step (c) includes the substeps of—

(c₁) evaluating each encoded potential solution with an objective fitness test;

(c₂) terminating any encoded potential solutions that fail the objective fitness test; and

(C₃) propagating any remaining encoded potential solutions to produce a population of encoded second potential solutions.

13. The method as set forth in claim 12, wherein substep (C₃) of step (c) further includes the substeps of—

(C_3.1) combining two or more of the remaining encoded potential solutions to produce a single member of the population of encoded second potential solutions; and

(c_3.2) introducing random change into the population of encoded second potential solutions by randomly changing the encoding of one or more encoded second potential solutions.

14. The method as set forth in claim 11, wherein the method is used to provide a simulation language preprocessor.

15. The method as set forth in claim 11, wherein the method is used to provide an embedded component operable to decision support for controlling a system and a process.

17. The method as set forth in claim 16, wherein step (f) includes combining two or more of the remaining encoded first potential solutions to produce a single member of the population of encoded second potential solutions.

18. The method as set forth in claim 16, wherein step (f) includes introducing randomness into the population of encoded second potential solutions by randomly changing the encoding of one or more of the encoded second potential solutions.

19. The method as set forth in claim 16, wherein the method is used to provide a simulation language preprocessor.

20. The method as set forth in claim 16, wherein the method is used to provide an embedded component operable to provide decision support for controlling a system and a process.

16. A method of facilitating decision making, the method comprising the steps of:

(a) creating a discrete event simulation for testing one or more encoded potential solutions by modeling a system wherein any changes take place substantially instantaneously and in countable increments in response to discrete events;

(b) incorporating a biological paradigm into the discrete event simulation,

(c) creating a population of encoded first potential solutions;

(d) evaluating each encoded first potential solution using an objective fitness test;

(e) terminating any encoded first potential solutions that fail the objective fitness test;

(f) propagating any remaining encoded parent potential solutions to produce a population of encoded second potential solutions;

(g) repeating steps (d)-(f), wherein in each repetition the population of encoded second potential solutions replaces the population of encoded first potential solutions; and

(f) identifying and reporting one or more fittest potential solutions.