CLUSTERING TRIZ ANALYSIS METHOD

• US 20080294658A1
• Filed: 01/04/2008
• Published: 11/27/2008
• Est. Priority Date: 05/21/2007
• Status: Active Grant

• Current Assignee: National Taiwan University of Science & Technology (Government of Taiwan)
• Original Assignee: National Taiwan University of Science & Technology (Government of Taiwan)
• Inventors: Lin, Zone-Ching, Chen, Mi-Yung
• Granted Patent: US 7,958,078 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2468 Fuzzy queries
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor

Abstract

The TRIZ decision process of the clustering method proposed by this invention uses the characteristics and invention rules from the contradiction matrix table resulting from massive quantities of patent inferences to find a similar or approximate character group and invention rule group of the physical meanings, and also applies statistics to calculate the number of display times of the groups to be the basic foundation.

Apart from the number of display times, Bayes probability, fuzzy object oriented method and Bayes probability combined with fuzzy object oriented method can be used as the system.

The reading value is utilized as a foundation for prioritizing the sequence of consideration for the groups, in which the system reading value constructed by different models gives designers lots of options to perform the reading, so as to acquire the undesired result features of the prioritized consideration.

Claims

1. A clustering TRIZ analysis method, comprising:

(1) constructing cluster elements according to features and inventive principles of a TRIZ matrix by associating items with similar physical meanings as one cluster;

(2) calculating display times of each cluster according to improved feature cluster and inventive principle cluster, calculating the display times of the corresponding inventive principle cluster and the no-worsening feature cluster according to a TRIZ contradiction matrix, and storing results of the calculation in a database;

(3) determining discrimination values of each model according to the display time results in step (2), and using the discrimination values to determine a priority order of the clusters.

2. The clustering TRIZ analysis method of claim 1, wherein the model comprises a display time model, Bayes probability model, fuzzy value model, and a combined fuzzy value and Bayes probability model, and the method determine discrimination values according to a fuzzy relation value defined by the models.

3. The clustering TRIZ analysis method of claim 1, further comprising:

providing a design direction;

selecting one of the improved feature clusters according to the design direction;

determining the priority of inventive principle cluster according to the selected improved feature cluster, and generating referential numerical values and discrimination values corresponding thereto;

determining the priority order of the no-worsening feature cluster according to the selected inventive principle clusters, and generating referential numerical values corresponding thereto;

determining a general table of discrimination values of the entire inventive principle cluster and the no-worsening feature cluster, and probability values corresponding to the display times, wherein determination is made as to whether the selected no-worsening feature clusters corresponding to the highest priority order is in the general table of discrimination values of the inventive principle cluster and the no-worsening feature cluster, whether the roll is the highest priority order through the vertical examination of the selected inventive principle clusters or whether the column is the best priority order when looking at the horizontal examination of the selected no-worsening feature clusters, and whether the result conforms with the design direction, limitation conditions and the technological requirements, and if so, deciding on a desired no-worsening feature cluster, otherwise, adopting the inventive principle cluster again;

selecting, from the selected improved feature clusters and the determined no-worsening feature clusters, a single improved feature and a single worsened feature conforming to the design direction and limitation conditions, and searching the original TRIZ matrix to identify a corresponding inventive principle.

4. The clustering TRIZ analysis method of claim 1, further constructing cluster elements associating similar items as one cluster.

5. The clustering TRIZ analysis method of claim 4, further associating 39 features of the TRIZ matrix into he following clusters—

moving object cluster, non-moving object cluster, object service life cluster, energy consumption cluster, substance amount cluster, manufacturing and measurement accuracy cluster, control cluster, operation and recovery, system efficiency and quality cluster, physical property cluster, wherein;

the moving object cluster comprises features of weight of the moving object, length of the moving object, area of the moving object, and volume of the moving object;

the non-moving object cluster comprises features of weight of the non-moving object, length of the non-moving object, area of the non-moving object, and volume of the non-moving object;

the object service life cluster comprises features of durability of the moving object, durability of the non-moving object, and waste of the substance;

the system efficiency and quality cluster comprises features of loss of the information, reliability, harmful factors acting on the object, and harmful side effects; and

the physical property cluster comprises features of speed, force, tension/pressure, shape, stability of the object, and hardness.

6. The clustering TRIZ analysis method of claim 4, further associating 40 inventive principles of the TRIZ matrix into the following clusters—

object segmentation and extraction cluster, prior handling cluster, counteraction cluster, cluster of partial action or structural change, balance object cluster, vibration and structural dynamicity cluster, harmful factor cluster, mechanical system replacement cluster, material cluster, physical change cluster, chemical action cluster, rejection and maintenance cluster, and system combination cluster, wherein;

object segmentation and extraction cluster comprises inventive principles of segmentation, extraction, asymmetry, prior action, cushion in advance, and dynamicity;

the cluster of partial action or structural change comprises inventive principles of local quality, spheroidality, partial, overdone or excessive action;

the mechanical system replacement cluster comprises inventive principles of moving to a new dimension, copying, replacement of a mechanical system, use of a pneumatic or hydraulic construction, and color change;

the material cluster comprises inventive principles of substituting an expensive, durable object by an inexpensive, non-durable one, flexible film or thin membrane, and use of porous material, homogeneity, and composite material.

7. The clustering TRIZ analysis method of claim 2, wherein the display time model comprises:

selecting the improved feature clusters and the inventive principle clusters having similar physical meanings;

calculating display times of each inventive principle cluster corresponding to each improved feature cluster according to the TRIZ matrix;

determining the priority order of the inventive principle clusters according to the calculated display times;

calculating display times of each of the no-worsening feature clusters according to the TRIZ matrix and the inventive principle cluster corresponding to the improved feature cluster; and

determining priority order of the no-worsening feature clusters according to the calculated display times.

8. The clustering TRIZ analysis method of claim 2, wherein the Bayes probability model comprises:

selecting the improved feature clusters and the inventive principle clusters having similar physical meanings;

determining display probability according to the contents of the TRIZ matrix, and the display times of each corresponding inventive principle cluster (I_j) made by each improved feature cluster, and the Bayes probability (P(C_i|I_j)) of each inventive principle cluster is determined according to the Bayes theorem equation as follows;

9. The clustering TRIZ analysis method of claim 2, wherein the fuzzy value model comprises:

selecting the improved feature clusters and the inventive principle clusters having similar physical meanings;

defining the fuzzy relationship value of each element in each feature cluster by 10 feature clusters having similar physical meanings and using a feature in the cluster as a major consideration factor for defining the fuzzy relationship value (F_value) between the major considered feature factor and other related features in accordance with the concept of the fuzzy object oriented model,;

determining the display times (I_F_value_i) of the added fuzzy value of each cluster and the total display times (S_F_value_j) of the added fuzzy values of the entire cluster required by the system in accordance with the TRIZ matrix and the display times (N_i) of each corresponding inventive principle cluster made by the improved feature cluster, and from the fuzzy relationship value (F_value_i) between the display times (N_i) and the elements in the improved feature cluster acquired through the fuzzy object oriented model, as follows;

10. The clustering TRIZ analysis method of claim 1, wherein the combined fuzzy value and Bayes probability model comprises:

selecting the improved feature clusters and the inventive principle clusters having similar physical meanings;

defining the fuzzy relationship value of each element in each cluster by 10 feature clusters having similar physical meanings and designating a feature in the cluster as a major consideration factor for defining the fuzzy relationship value (F_value) between the major considered feature factor and other related features in accordance with the fuzzy object oriented model;

determining the display times (I_F_value_i) of the added fuzzy value of each feature factor and the total display times (S_F_value_i) of the added fuzzy values of the entire cluster in accordance with the TRIZ matrix and the display times (N_i) of each corresponding inventive principle cluster made by the improved feature cluster (C_m), and from the fuzzy relationship value (F_value_i) between the display times and the elements in the improved feature cluster acquired through the fuzzy object oriented model, as follows;