Optimization adjusting method and optimization adjusting apparatus

• US 6,004,015 A
• Filed: 11/22/1995
• Issued: 12/21/1999
• Est. Priority Date: 11/24/1994
• Status: Expired due to Term

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Watanabe, Tatsumi, Takagi, Hideyuki, Maruno, Susumu, Obara, Kazuaki
• Primary Examiner(s): Louis-Jacques, Jacques H.
• Assistant Examiner(s): Pipala, Edward J.
• Application Number:
  US08/561,708
• Time in Patent Office: 1,490 Days
• Field of Search: 364/148, 364/151, 364/152, 364/153, 364/164, 364/578, 395/13
• US Class Current: 700/28
• CPC Class Codes: G06F 2111/06 Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
• CPC Class Codes: G06F 30/327 Logic synthesis; Behaviour synthesis, e.g. mapping logic, HDL to netlist, high-level language to RTL or netlist
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• 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: H04R 25/70 Adaptation of deaf aid to hearing loss, e.g. initial electronic fitting

Abstract

At the start of a process, an information presenting section presents information expressed by a solution vector, while an auxiliary information presenting section presents information for facilitating comparative evaluation of multiple pieces of information being presented simultaneously. Based on these pieces of information, the user evaluates each solution vector. A model estimation executing section generates an evaluation model for an adjustment process from a user evaluation history, and thereafter, a model evaluation calculating section performs evaluation using the evaluation model. A fitness calculating section calculates fitness from the evaluation made by the user or by the model, and in accordance with the thus calculated fitness, a recombination operation section performs an arithmetical recombination operation. The above process is iteratively performed to carry out optimum adjustment of solution vectors.

Claims

1. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, comprising:

a first step of selecting previously stored optimal solution vectors when setting an initial solution vector set;

a second step of setting an initial set of solution vectors to be adjusted, by using the solution vectors selected in the first step;

a third step of presenting information expressed by each solution vector to a user;

a fourth step of making the user evaluate each solution vector on the basis of his own criterion;

a fifth step of obtaining the fitness of each solution vector from the user'"'"'s evaluation value;

a sixth step of performing an arithmetical recombination operation based on genetic recombination, on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set; and

a seventh step of outputting an optimal solution vector after storing the same on a recording medium,

wherein the third through sixth steps are iteratively performed until a predetermined termination condition is satisfied, and in the seventh step, the resulting optimal solution vector is stored, so that a group of optimal solution vectors obtained in previous adjustments is used as an adjustment starting point when the user next performs an adjustment process, thereby enabling quick estimation of a solution vector optimum for the user.

2. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, comprising:

a first step of setting an initial set of solution vectors;

a second step of presenting information expressed by each solution vector to a user;

a third step of making the user evaluate each solution vector on the basis of his own criterion;

a fourth step of measuring physiological data when the user is making the evaluation;

a fifth step of estimating the user'"'"'s psychological state from the physiological data thus measured;

a sixth step of correcting the evaluation value obtained in the third step on the basis of the thus estimated psychological state;

a seventh step of obtaining the fitness of each solution vector from the corrected evaluation value;

an eighth step of performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set; and

a ninth step of outputting an optimal solution vector,

wherein by iteratively performing the second through eighth steps until a predetermined termination condition is satisfied, a solution vector optimum for the user is quickly estimated, while reducing the effect of the user'"'"'s psychological state.

3. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, the method being specifically concerned with an evaluation model for a user'"'"'s adjustment process, comprising:

a first step of setting an initial set of solution vectors;

a second step of presenting information expressed by each solution vector to a user;

a third step of making the user evaluate each solution vector on the basis of his own criterion;

a fourth step of recording the user'"'"'s evaluation and its corresponding solution vector as a history;

a fifth step of judging whether the history of the user'"'"'s adjustments satisfies a predetermined evaluation model estimation condition;

a sixth step of obtaining the fitness of each solution vector from the user'"'"'s evaluation value when it is judged in the fifth step that the history does not satisfy the predetermined condition;

a seventh step of performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set; and

an eighth step of estimating an evaluation model for the current user'"'"'s adjustment process from the history of the user'"'"'s adjustments when it is judged in the fifth step that the history satisfies the predetermined condition,

wherein in the eighth step the evaluation model for the user'"'"'s adjustment process is estimated from the history of adjustments obtained by iteratively performing the second through seventh steps until the predetermined condition is satisfied.

4. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, the method being specifically concerned with a common model representing a plurality of user'"'"'s adjustment processes, comprising:

a first step of setting an initial set of solution vectors to be adjusted;

a second step of presenting information expressed by each solution vector to a user;

a third step of making the user evaluate each solution vector on the basis of his own criterion;

a fourth step of recording the user'"'"'s evaluation and its corresponding solution vector as a history;

a fifth step of obtaining the fitness of each solution vector from the user'"'"'s evaluation value;

a sixth step of performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a seventh step of, when a predetermined user evaluation termination condition is satisfied, judging whether the number of users recorded satisfies a predetermined common model estimation condition; and

an eighth step of, when it is judged in the seventh step that the number of users satisfies the predetermined condition, estimating a common model for an adjustment process from the recorded history of the adjustments made by a plurality of users,

wherein the common model representing the adjustment processes by the plurality of users is estimated from the history of adjustments obtained by iteratively performing the second through seventh steps until the predetermined condition is satisfied.

5. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, comprising:

a first step of setting an initial set of solution vectors to be adjusted;

a second step of evaluation each solution vector by using a previously extracted common model representing a plurality of users'"'"' adjustment processes;

a third step of presenting information expressed by each solution vector to a user;

a fourth step of making the user judge whether the evaluation made in the second step using the common model is appropriate or not;

a fifth step of, when it is judged in the fourth step that the evaluation is not appropriate, making the user evaluate each solution vector on the basis of his own criterion;

a sixth step of, when the evaluation is made by the user in the fifth step, recording the solution vector and the user'"'"'s evaluation thereof as a history;

a seventh step of judging whether the recorded history of solution vector set updates satisfies a common model update condition;

an eighth step of, when it is decided in the seventh step that the common model be updated, updating the common model representing the plurality of users'"'"' adjustment processes by using the recorded history of data;

a ninth step of obtaining the fitness of each solution vector from the user'"'"'s subjective evaluation value or from the evaluation value obtained from the adjustment process common model;

a 10th step of performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set; and

an 11th step of outputting an optimal solution vector,

wherein by iteratively performing the second through 10th steps until a predetermined termination condition is satisfied, the common model representing the plurality of users'"'"' adjustment processes is constantly updated, while, at the same time, estimating a solution vector optimum for the user.

6. An optimization adjusting method for estimating an optimal solution, using a processing means, by successively updating solution vectors, arbitrarily set for a problem, in accordance with fitness thereof, comprising:

a first step of setting an initial set of solution vectors to be adjusted;

a second step of presenting information expressed by each solution vector to a user;

a third step of making the user evaluate each solution vector on the basis of his own criterion;

a fourth step of recording the user'"'"'s evaluation and its corresponding solution vector as a history;

a fifth step of judging whether the history of the user'"'"'s adjustments satisfies a predetermined evaluation model estimation condition;

a sixth step of estimating an evaluation model for the current user'"'"'s adjustment process from the history of a series of user adjustments when it is judged in the fifth step that the history satisfies the predetermined condition;

a seventh step of obtaining the fitness of each solution vector from each evaluation value;

an eighth step of performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a ninth step of, when the evaluation model for the user'"'"'s adjustment process is estimated in the sixth step, estimating each solution vector by using the evaluation model instead of going through the third step; and

a 10th step of outputting an optimal solution vector,

wherein the evaluation model for the user'"'"'s adjustment process is estimated from the history of adjustments obtained by iteratively performing the second through eighth steps until the evaluation model estimation condition is satisfied, and thereafter, the seventh through ninth steps are performed iteratively by using the evaluation model until a predetermined termination condition is satisfied.

7. An optimization adjusting apparatus comprising:

a subject data input section for inputting data which a target apparatus operates on;

an update region limiting section for limiting an update region that the solution vectors to be adjusted can take;

an initial set setting section for setting an initial set of solution vectors within the update limited region;

an information presenting section for presenting information expressed by each solution vector to a user;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

a fitness deriving section for obtaining the fitness of each solution vector from the evaluation value obtained in the user evaluation judging section;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector;

a set resetting section for constructing a new solution vector set by replacing, in the solution vector set generated by the recombination operation section, the solution vectors lying outside the limited range with arbitrary vectors that fall within the limited range;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

8. An optimization adjusting apparatus according to claim 7, as applied to the creation and adjustment of a vision-correcting lens, wherein

a set of lens parameters necessary for vision correction is treated as the solution vectors to be adjusted,

the subject data input section handles a test image for presentation to the user,

the information presenting section presents to the user the test image viewed through a corrective lens created by each solution vector, and

the apparatus adjustment executing section adjusts a vision-correcting lens optimum for the user in accordance with the obtained optimum solution vector.

9. An optimization adjusting apparatus according to claim 7, as applied to the creation and adjustment of a character font for personal user, wherein

a set of parameters describing a character font is treated as the solution vectors to be adjusted,

the subject data input section handles basic character font data for presentation to the user,

the information presenting section presents to the user a deformation of the basic character font as created by each solution vector, and

the apparatus adjustment executing section makes adjustments to create a character font that best matches the user'"'"'s preferences by using the obtained optimum solution vector.

10. An optimization adjusting apparatus comprising:

a subject data input section for inputting data which a target apparatus operates on;

a recording medium holding thereon a group of past optimal solution vectors;

a recorded information reading section for reading the group of past optimal solution vectors from the recording medium;

an initial solution vector selecting section for selecting initial solution vectors for the current adjustment from the solution vector group read by the recorded information reading section;

an initial solution vector replenishing section for setting an initial set of solution vectors by adding randomly generated initial solution vectors to the initial solution vectors selected by the initial solution vector selecting section;

an information presenting section for presenting information expressed by each solution vector to a user;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

a fitness deriving section for obtaining the fitness of each solution vector from the evaluation value obtained in the user evaluation judging section;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied;

an optimal solution vector recording section for recording the optimal solution vector on the recording medium; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

11. An optimization adjusting apparatus comprising:

a subject data input section for inputting data which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

an information presenting section for presenting information expressed by each solution vector to a user;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

a physiological data measuring section for measuring physiological data when the user is making the evaluation in the user evaluation judging section;

a psychological state estimating section for estimating the user'"'"'s psychological state from the physiological data thus measured;

an evaluation value correcting section for correcting an evaluation value on the basis of the user'"'"'s psychological state estimated by the psychological state estimating section;

a fitness deriving section for obtaining the fitness of each solution vector from the evaluation value corrected by the evaluation value correcting section;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

12. An optimization adjusting apparatus for an evaluation model for a user'"'"'s adjustment process, comprising:

a subject data input section for inputting data which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

an information presenting section for presenting information expressed by each solution vector to a user;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

an adjustment history recording section for recording an evaluation value obtained by the user evaluation judging section and its corresponding solution vector;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s evaluation value;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

an evaluation model estimation judging section for judging whether the history of the user'"'"'s adjustments recorded in the adjustment history recording section satisfies a predetermined evaluation model estimation condition;

a model estimation executing section for, when it is judged that the evaluation model estimation condition is satisfied, estimating an evaluation model for the current user'"'"'s adjustment process from the history of the user'"'"'s adjustments; and

an evaluation model output section for outputting information representing the user evaluation model estimated by the model estimation executing section.

13. An optimization adjusting apparatus for an evaluation model for a user'"'"'s adjustment process according to claim 12, wherein the model estimation executing section is constructed from a neural network.

14. An optimization adjusting apparatus for a common model representing a plurality of user'"'"'s adjustment processes, comprising:

a subject data input section for inputting data which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

an information presenting section for presenting information expressed by each solution vector to a user;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

an adjustment history recording section for recording an evaluation value obtained by the user evaluation judging section and its corresponding solution vector;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s evaluation value;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a user adjustment termination judging section for judging whether a predetermined user adjustment termination condition is satisfied;

a common model estimation judging section for, when the user adjustment termination condition is satisfied, judging whether the recorded history of a plurality of users'"'"' adjustments satisfies a predetermined common model estimation condition;

a common model estimation executing section for, when it is judged that the common model estimation condition is satisfied, estimating a common model for an adjustment process from the history of the adjustments made by a plurality of users recorded in the adjustment history recording section; and

a common model output section for outputting information representing the common model for the adjustment process estimated by the common model estimation executing section.

15. An optimization adjusting apparatus for a common model relating to a plurality of users'"'"' adjustment processes according to claim 14, wherein the common model estimation executing section is constructed from a neural network.

16. An optimization adjusting apparatus comprising:

a subject data input section for inputting data which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

a common model evaluation calculating section for evaluating each solution vector by using a common model representing a plurality of users'"'"' adjustment processes;

an information presenting section for presenting information expressed by each solution vector to a user;

a common model evaluation judging section for making the user judge whether the evaluation value calculated by the common model evaluation calculating section is appropriate or not;

a user evaluation judging section for making the user evaluate each solution vector when it is judged in the common model evaluation judging section that the evaluation value is not appropriate;

an adjustment history recording section for recording the evaluation value obtained by the user evaluation judging section and its corresponding solution vector;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s evaluation value or from the evaluation value obtained from the adjustment process common model;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a common model update judging section for judging whether the history of the user'"'"'s adjustments recorded in the adjustment history recording section satisfies a predetermined common model update condition;

a common model update executing section for, when the common model update condition is satisfied, updating the common model by using the recorded history of the user'"'"'s adjustments;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

17. An optimization adjusting apparatus comprising:

a subject data input section for inputting data which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

a method selection switch for making selection as to whether evaluation should be performed by a user or by using an evaluation model;

an information presenting section for presenting information expressed by each solution vector to the user when evaluation by the user is selected by the method selection switch;

a user evaluation judging section for making the user evaluate each solution vector;

an adjustment history recording section for recording the evaluation value obtained by the user evaluation judging section and its corresponding solution vector;

a method selection judging section for judging whether the history of the user'"'"'s adjustments recorded in the adjustment history recording section satisfies a predetermined evaluation model estimation condition, and for, when the evaluation model estimation condition is satisfied, directing the method selection switch to switch selection;

a model estimation executing section of the evaluation for, when the switching of selection is directed by the method selection judging section, estimating an evaluation model for the current user'"'"'s adjustment process from the history of a series of adjustments;

a model evaluation calculating section for, when evaluation by an evaluation model is selected by the method selection switch, evaluating each solution vector by using the evaluation model;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s subjective evaluation or from the evaluation by the evaluation model;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

18. An optimization adjusting apparatus for a common model representing a plurality of user'"'"'s adjustment processes in a problem dealing with time sequential information, comprising:

a subject data input section for inputting time sequential information which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

an information presenting section for presenting information expressed by each solution vector to a user;

an auxiliary information presenting section for creating and presenting information that assists the user'"'"'s memory when the user compares and evaluates the time sequential information;

a user evaluation judging section for making the user evaluate each solution vector on the basis of the information presented by the information presenting section;

a presentation order altering section for altering the order of presenting information expressed by each solution vector in accordance with an evaluation value obtained in the user evaluation judging section;

an adjustment history recording section for recording the evaluation value and its corresponding solution vector;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s evaluation value;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a user evaluation termination judging section for judging whether a predetermined user evaluation termination condition is satisfied;

a common model estimation judging section for, when the user evaluation termination condition is satisfied, judging whether the recorded history of a plurality of users'"'"' adjustments satisfies a predetermined common model estimation condition;

a common model estimation executing section for, when it is judged that the common model estimation condition is satisfied, estimating a common model for an adjustment process from the history of the adjustments made by a plurality of users recorded in the adjustment history recording section; and

a common model output section for outputting information representing the common model for the adjustment process estimated by the common model estimation executing section.

19. An optimization adjusting apparatus for a problem dealing with time sequential information, comprising:

a subject data input section for inputting time sequential information which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

a common model evaluation calculating section for evaluating each solution vector by using a common model representing a plurality of users'"'"' adjustment processes;

an information presenting section for presenting information expressed by each solution vector to a user;

an auxiliary information presenting section for creating and presenting information that assists the user'"'"'s memory when the user compares and evaluate the time sequential information;

a common model evaluation judging section for making the user judge whether the evaluation value calculated by the common model evaluation calculating section is appropriate or not;

a user evaluation judging section for making the user evaluate each solution vector when it is judged in the common model evaluation judging section that the evaluation value is not appropriate;

a presentation order altering section for altering the order of presenting information expressed by each solution vector in accordance with an evaluation value obtained in the user evaluation judging section;

an adjustment history recording section for recording the evaluation value and its corresponding solution vector;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s evaluation value or from the evaluation value obtained from the adjustment process common model;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

a common model update judging section for judging whether the history of the user'"'"'s adjustments recorded in the adjustment history recording section satisfies a predetermined common model update condition;

a common model update executing section for, when the common model update condition is satisfied, updating the common model by using the recorded history of the user'"'"'s adjustments;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

20. An optimization adjusting apparatus for a problem dealing with time sequential information, comprising:

a subject data input section for inputting time sequential information which a target apparatus operates on;

an initial set setting section for setting an initial set of solution vectors to be adjusted;

a method selection switch for making selection as to whether evaluation should be performed by a user or by using an evaluation model;

an information presenting section for presenting information expressed by each solution vector to the user when evaluation by the user is selected by the method selection switch;

an auxiliary information presenting section for creating and presenting information that assists the user'"'"'s memory when the user compares and evaluates the time sequential information;

a user evaluation judging section for making the user evaluate each solution vector;

a presentation order altering section for altering the order of presenting information expressed by each solution vector in accordance with an evaluation value obtained in the user evaluation judging section;

an adjustment history recording section for recording the evaluation value and its corresponding solution vector;

a method selection judging section for judging whether the history of the user'"'"'s adjustments recorded in the adjustment history recording section satisfies a predetermined evaluation model estimation condition, and for, when the evaluation model estimation condition is satisfied, directing the method selection switch to switch selection;

a model estimation executing section of the evaluation for, when the switching of selection is directed by the method selection judging section, estimating an evaluation model for the current user'"'"'s adjustment process from the history of a series of adjustments;

a model evaluation calculating section for, when evaluation by an evaluation model is selected by the method selection switch, evaluating each solution vector by using the evaluation model;

a fitness deriving section for obtaining the fitness of each solution vector from the user'"'"'s subjective evaluation or from the evaluation by the evaluation model;

a recombination operation section for performing an arithmetical recombination operation on the solution vectors in the solution vector set in accordance with the fitness of each solution vector, and thereby constructing a new solution vector set;

an optimal solution vector output section for outputting an optimal solution vector when a predetermined termination condition is satisfied; and

an apparatus adjustment executing section for adjusting the apparatus operating on the data input from the subject data input section, in accordance with the solution vector from the optimal solution vector output section.

21. An optimization adjusting apparatus according to claim 20, wherein the user evaluation judging section obtains graduated evaluation based on relative evaluation within the solution vector set presented to the user.