5.  Various methods for improving the user's system
           (Prediction Module: Part 2. Method tree)

Let us go ahead further towards more concrete methods useful for solving our own
prolems.  Next is the function for assisting the users to improve their own object systems;
it is implemented in the [Prediction Module] of the [TechOptimizer Pro].  Though not
described in Section 2, the principal function of [Prediction Module] is to guide users to
consider applying a large number of alternative "standard means" for improving their own
problems.  For this purpose, it has a database of typical examples of applying standard
means of problem solving, and shows them to the users.  The implementation of this
function is also much sophisticated.  Let us first examine the contents of the database,
and then discuss the ways of information retrieval of the database.

5.1  List of standard means for system improvement and
       the database of application examples

The [Prediction Module] has a database containing 248 cases of problem solving by using
"standard means".  Each case is composed of a display of the old design of the system
(text with an illustration) and another display of the new design (text with an illustration)
after solving the problem.  The method applied to the problem solving is characterized
and illustrated with a simple scheme of objects and functions.  Such "standard means"
(which corresponds to the 76 "standard means" of the TRIZ methodology) are listed in
the [Prediction tree] window of the [Prediction Module].

[Operation:  When the [Prediction Module] is opened, the display appears like the one
shown in Fig. 5.1.  The module expects that the user inputs, in the [Initial model]
window at the middle left of the display, the description of the object system user wants
to improve and of a candidate method for improving it.  You may leave columns of the
system description blank, but have to click either one in the [Define required change of
action] window (click any, othewise the [Database] menue cannot be opened in the next
step).  In response to these inputs, a message containing "IM recommends ..." is shown
in the upper right of the display; but you may discard it at moment for going to the next
step.  You should open the [Database] menue in the menue bar and click the [Prediction
tree] command.  The [Prediction tree] window is now open to show the standard means
of system improvement in a hierarchical way.  The [Prediction tree] is actually composed
of two parts; they can be switched to the other by clicking either [Transform] button or
[Measure] button in the [Lines] window at the left end of the display. ]



Fig. 5.1  The initial, main display of the [Prediction Module]

In the [Prediction tree], standard elementary means for problem solving  are represented
in a simplified scheme of the object system (or its component system), as shown below in
Fig. 5.2.


Fig. 5.2  Simplified object system to be handled in the [Prediction tree]

In such a simplified scheme, [TechOptimizer Pro] proposes 67 elementary methods as
TRIZ's standard means of problem improvement.  The standard means are classified
hierarchically as shown in Table 5.1.  The means at the top levels are classified as:

  method of problem solving
     --  methods of transformation
          ---  by use of interaction with additives
          ---  by use of boosted interactions
          ---  by use of trimmed interactions
          ---  by use of multiplied interactions
     --  methods of measuring
          --- methods of using marks
          --- methods of roundabout


Table 5.1 Standard means of problem solving:
                 [Prediction tree] in the [Prediction Module]
 
Transform:
Interaction with 
   additives:
Using new substances: in obj1, in obj2, between obj1 and obj2, 
on obj1, on obj2, around obj1,2, 
instead of obj1, instead of obj2
Using modified substances:  in obj1, in obj2, between obj1 and obj2, 
on obj1, on obj2, around obj1,2, 
instead of obj1, instead of obj2
Using voids:  in obj1, obj2, between obj1 and obj2 
Using fields, forces, :  in obj1, in obj2, around obj1,2
Boosted 
interactions:
Segmenting:  obj1 into parts, obj1 down to particles, 
obj1 into porous obj, 
obj2 into parts, obj2 down to particles, 
obj2 into porous obj
Dynamization of : obj1 by making hinged, obj1 by making flexible, 
obj1 by phase transitions, 
obj2 by making hinged, ovj2 by making flexible, 
obj2 by phase transitions, 
action by making pulsed
Structurization of : obj1, obj2, action
Coordinating Rhythm of :  obj1 and action, obj2 and action, 
different actions,  action's pauses
Trimmed 
interactions:
Combining obj1 and obj2
Removing obj1 (obj2)
Self-serving obj1 (obj2)
Multiplied
interactions:
Multiplying:  obj1, obj2
Multiplying and connecting: obj1, obj2
Multiplying and differing:  obj1, obj2 
Multiplying and combining:  obj1, obj2
Redistributing features :  obj1, obj2
Measure:
Using marks: Using new substances: in obj, on obj, around obj
Using modified substances: in obj, on obj, around obj
Roundabout: Roundabout:
Discrete detecting
Indirect measuring 
Measuring derivatives

[Prediction Module] has a database of application examples for demonstrating these
standard means.  (Note that this database is different from the [Example] database of the
[Effects Module].)

[Operation:  Click a method in the [Prediction tree] window, and then click an example in
the list of examples shown in the [Examples] window at the lower right of the display.
The display now appears as [Examples of Current Concept] and shows the [new design]
of the system after improvement.  By clicking the [Show old design] button at the lower
right of the dispaly, you should better read first the [old design] of the system before
improvement. ]

These examples of system improvement (or problem solving) are very illustrative and
useful.  In total, 248 cases are installed.  You should read the [old design] first; it
describes the purpose and structure (with an illustration) of the present system and the
problem to be solved.  You shoud challenge to think of any possible solution of the
problem for a while by yourself; and then read the [new design] as an answer.  This
serves you as a good self-training.  Since these application cases are classified with the
TRIZ's standard means, they form a good textbook to understand the standard means.

An example of system improvement is demonstrated in Figs. 5.3 and 5.4.  "We want to
form a very thin film on the surface of a substrate, and to examine the film after
removing from the substrate.  To protect the thin film from any damadge during the
separation, what can we do?"  The solution of this problem is shown:  "Before forming
the thin film, let the substrate absorb halogen gas to the saturation.  After the film is
formed, warm the substrate; then, together with the halogen gas coming out, the thin film
will be separated from the substrate without damage".  (It is not clear in the description
what kinds of substrate and films this case actually handles.  The present author is not
familiar how practical this method is and how widely this technique can be applied.)



Fig. 5.3  Example of the display of [Example to current concept] of the [Prediction Module]:
              Method of separating a thin film (old design)



Fig. 5.4  Example of the display of [Example to current concept] of the [Prediction Module]:
              Method of separating a thin film (new design)

The present author has read all the 248 cases.  They cover the areas of science and
technology similar to those the database of the [Effect Module] covers.  The cases,
however, are mostly very old.  Moreover, the present author feels that a considerable
number of them show just the ideas for improvement and not yet actually used in the real
applications.  Most of such examples are taken from "patents" of the former USSR in the
old days of 1950s and 1960s.  The "patent" of USSR in those days were simply new ideas
registered after reviewing but not well examined from the technical effectiveness.  In the
recently published textbook "An Introduction to TRIZ" (Hatamura et al; Nikkan Kogyo
Shimbun, 1997, in Japanese), Professor Hatamura of the University of Tokyo made this
sort of criticisms in detail; the present author agrees in them.  In this point, it is very
much desirable to modernize the contents of [TechOptimizer Pro].

5.2  Thinking-experiments with the TRIZ' standard methods for improvement

The TRIZ methodology recommends to consider to apply the standard means discussed
above one by one to your own problem.  [TechOprimizer Pro] offers this function in the
[Prediction Module] as the module's main feature by using  the following mechanism:

[Operation:  In the main display of the [Prediction Module] (see Fig. 5.1), you need to
input a simplified description of your current object system in the middle left windows.
(Let us assume your object system is composed of "Obj1", "Action P", and "Obj2".)
Next, in the [Define required change of action] window at the lower left of the display,
you must select a verb for showing the direction of improvement on the action.  The
alternatives of the verb are "generate", "separte", "make pulsed" etc.  (Here for the
explanation we take a general verb "modify", even though it is not in the list.)  Then, the
software shows, in the [Current concept] window at the upper right of the display, a
statement saying "IM recommends ..." together with a schematic diagram, and in the
lower right of the display a list of examples selected from the database.  Thus, as
explained in the previous subsection, you should think over whether and how the IM's
recommendation and the displayed examples can be applied to your own problem.

Note that the statement and the list depend on your choice of the standard means
specified in the [Prediction tree].  You can specify your choiceof the standard means in
the way as described in Section 5.1 or by clicking the downward/upward arrow at the
bottom right of the [Current Concept] window.  ]

When you input a brief scheme of your system and the direction of improvement on the
action, the software tool outputs the statement "IM recommends ..." as explained in the
above operation note.  This very likely makes you think that the software must have high
intelligence with some AI logic and is recommending a thoughtful suggestion.

However, this is not the case.  Actually, [TechOptimizer Pro] outputs the statement and
the list of examples just mechanically with the following mechanism:

[Current Concept]
   "IM recommends:   You may 'modify' 'Action P'
                                                           (1)            (2)
       by 'introducing new substance' 'between' 'obj1' 'and' 'obj2'. "
                             (3)                                  (4)         (5)      (4)      (6)

  (1)  <=  The verb you have specified in the [Define required change of action] window
  (2)  <=  The interaction you have specified in the [Action] window for the description of
                  the system components
  (3)  <=  The standard means you have specified in the [Prediction tree] window; its
                   upper-level part representing the main function.
  (4)  <=  The standard means you have specified in the [Prediction tree] window; its
                   lower-level part representing the target object or location on which the standard
                   means is applied.
  (5)  <=  The object you have specified in the [Object 1] window for the system
                  description
  (6)  <=  The object you have specified in the [Object 2] window for the system
                  description

  [Schematic]  <=  The schematic diagram representing the standard means you have
                                    specified in the [Prediction tree] window
  [List of examples]  <=  The list of examples for the standard means you have specified
                                             in the [Prediction tree] window

This mechanism is simple but so well formulated that the statement of "IM recommends
..." may look likely and attractive for you.  However, the software assumes that you try
the standard means in the [Prediction tree] one by one on your problem, and it simply
repeats the information you have specified, as shown above.  The examples are also
displayed corresponding simply to the standard means you have specified in the
[Prediction tree] and with no relation to your system description.

At the bottom right of the [Current Concept] window, there may appear the following 4
buttons depending on the means  specified in the [Prediction tree]:

  [Substance] button:   a list of various states of substance
  [Modified substance] button:  a list of various modified states of substance
  [Void] button:  a list of various forms of void/empty/hole
  [Field] button:  a list of various types of field/forces/interactions

By clicking one of these buttons, you can see a detailed list of substances or "Fields"
containing information similar to Figs. 2.2 and 2.3.  If you click a substance or Field in
the list, a part of the recommendation statement in the [Current Concept] is replaced and
made more specific; for instance, the phrase 'new substance' may be replaced with 'porous
substance'.

For searching the example database of the [Prediction Module], you may also use the
keyword search.  In the main display of the module, open the [Database] menue in the
menue bar, and click the [Find] command.  Then by inputting the keyword you can
retrieve the relevant examples.  For example, by using the keyword of "'joint' or 'hinge'",
a list of 7 examples are obtained.

In order to use this software function in real porblem solving, it shoud be important to
understand the mechanism of this function as described here and to try to examine only
the cases worthy of consideration.  Otherwise, you will be either impressed by the
function to think it great without understanding its nature, or exhausted and tired of
examining so many cases of alternative possible means.

  6.  To learn the "40 Principles of Invention"
        (Principles Module: Part1. Principles and Examples)

Let us master the [Principles Module] next.  This module contains the core part of the
TRIZ methodology.  In the manner similar to the preceeding sections, let us examine the
databases and then learn how to use them.  In the database of this module, the "40
Principles of Invention" and their application examples, i.e. 210 cases in total, are
implemented.  For using the database, in addition to the keyword search, the method
using the "Contradiction Solving Matrix" is most important and uniquely sophisticated in
the TRIZ methodology.

[Operation:   In the main display of the [Principle Module], open the [Database] menue in
the menue bar and then click the [Principles list] command.  Thus the display appears as
shown in Fig. 6.1.  In the [Principles] window at the lower part of the display, the "40
Principles of Invention" are listed in its standard order of numbering.  When you click a
Principle in the list, explanation of the principle appears at the upper right of the display.
The explanation includes schematic figure(s) and concise description in at-most several
lines.  By clicking the [Examples] button at the upper right of the display, you can read
several (usually from 3 to 8) examples of applying the principle, each  illustrated with a
figure.  ]

The "40 Principles of Invention" were established by Altshuller, the Founder of TRIZ, by
extracting essences of inventions from a huge number of patents world-wide and
condensing them into practical guiding principles for solving difficult, inventive
problems.  In the TRIZ textbooks written by Altshuller, these principles of invention are
explained one by one with illustrative application examples.  If you fully understand
these principles and are able to readily use them in appropriate situations, you may say
you have mastered the essence of the TRIZ methodology.  Most of the functions in
TechOptimizer Pro serve the users to let them understand and utilize these principles of
invention.

A table of the 40 Principles of Invention is very useful; you should have one at hand for
ready reference.  In Table 6.1, the 40 principles are listed as written in the [Principles
Module].  (For reading them in Japanese, one should refer to a textbook, such as the one
published by Nikkei BP in Dec. 1997.  (The quality of translation into Japanese in this
textbook was improved from that in the article appeared in Nikkei Mechanical in 1996.)


Table 6.1   Table of the "40 Princioles of Invention"
 
1.  Segmentation Divide object; 
Disassemble; 
Divide into smallest possible 
   parts 
21.  Skipping Skip harmful actions
2.  Taking out Take out harmful part; 
Single out useful part
22.  'Blessing in 
       disguise'
Rearrange harmful actions 
  to get a useful result; 
Compensate two harmful 
   actions 
3.  Local Quality Make object non-uniform 23.  Feedback  Use feedback loop; 
Regulate feedback loop 
4.  Asymmetry Make object asymmetric 24.  'Intermediary' Control action using an 
   intermediary
5.  Merging  Merge objects 25.  Self-service Use self-performed 
    actions
6.  Universality Make object universal  26.  Copying Make copy of the object; 
Optical copies; 
Invisible copies
7.  'Nested doll' Place many in one; 
Move one inside another 
27.  Cheap 
       short-living
Use less-costly disposable 
  objects 
8.  Anti-weight Compensate weight using 
    another object; 
Using a medium
28.  Mechanics 
       substitution
Replace mechanical 
   devices with physical 
   fields
9.  Preliminary 
      anti-action
Perform anti-action in 
   advance
29.  Pneumatics and
       hydraulics 
Make part of object 
   gaseous or liquid 
10.  Preliminary 
       action
Perform partial action in 
   advance 
30.  Flexible shells
       and thin films
Make part of object out 
   of shells; 
Isolate using shells 
11.  Beforehand
       cushioning 
Use protective and curative 
   measures
31.  Porous materials Make object porous; 
Fill pores
12.  Equipotentiality Eliminate the need to 
   elevate object 
32.  Color changes Change object's color; 
Make object transparent 
13.  'The other way 
      around'
Replace the action eith its 
   converse; 
Invert movement/position
33.  Homogeneity  Use similar features for 
   contiguous objects 
14.  Spheroidality Use no-flat shapes; 
Rollers; 
Rotation 
34.  Discarding and
      recovering 
Eject used parts; 
Rebuild object during 
  action 
15.  Dynamics  Optimize object's 
   parameters 
Movable parts; 
Adaptive parts 
35.  Parameter 
       changes
Use different phases; 
Consistency; 
Flexibility; 
Temperature 
16.  Partial or 
      excessive actions
Make action excessive; 
Make action 'deficient'
36.  Phase
       transitions 
Use phase transitions 
17.  Another 
       dimension
Move object in 3 dimensions; 
Use 'multistories' and 
   'back sides' 
37.  Thermal 
       expansion
Use thermal expansion 
   and contraction; 
Use poly-expansion 
18.  Mechanical 
      vibration
Shake object ...; 
using ultrasonic, el-magn. 
   fields; 
using resonance 
38.  Strong oxidants Use strong oxidants
19.  Periodic 
      action
Make action pulsatory; 
Change period; 
Use pauses
39.  Inert 
       atmosphere 
Use an inert medium; 
Use inert parts 
20.  Continuity of 
       useful action
Continuously perform same 
   action
40.  Composite 
       material
Use 'armature'

To understand the nature and philosophy of the 40 Principles of Invention, let us read the
Principle No. 1 "Segmentation" as the most important example.  The display of the
principle is shown in Fig. 6.1.  This principle is the concept of "to segment an object into
smaller ones".  It says that the segmented parts should better be made as independent as
possible with each other.  Segmentation so as to easy to assemble/disassemble is also
useful.  It is also advised to segment an object into much smaller ones.  (It should be
noted that the same idea is also emphasized in the trends of technical evolution as
discussed in Section 2.  The trends No. 1 through No. 4 are related with the segmentation
principle and discuss about segmentation of tools, working parts, surface, and space,
respectively. )



Fig. 6.1  Example display of the Principles of Invention in the [Principles] window of the
[Principles Module]

One of the application examples of Principle No. 1 is shown in Fig. 6.2.  Similarly to the
example displays in other modules, the display is composed of a illustaration and concise
text.  The text for explanation is organized systematically with the following items:

  Principle of Invention:                  Brief statement in one line.
  Title of the example:                     Purpose of application.
  Problem to solve in this example:   Explanation of the current system and its  problems.
  Names of principles of invention applied in this example:   multiple principles are often
                                                            used in one invention
  Concrete description of the technology which solves the problem:  new idea/design
                                                             which is proposed here
  Effects, advantages and improved points of the new technology:
                     It should be noted that new problems and tasks associated with the new
                     technology are not described.
  source of reference:             Patent number, handbooks, paper references, etc.



Fig. 6.2  Demonstration of an application example in the [Examples] display of the
[Principles Module]

The database of these application examples contains 210 cases in total, and covers
essentially the same areas as the databases in other modules.  For each Principle of
Invention, around 5 (usually from 3 to 8) examples are demonstarted; such examples do
not overlap among multiple principles of invention.

This application-example database can be searched with the keyword indexing, as well.

[Operation:  In the main display of the [Principles Module], open the [Database] menue
in the menue bar, and click the [Find] command.  Then input the keyword for searching
with, and click the [Find] button for start searching. ]

For instance, when the database was searched with the keyword "Segment" (i.e. the word
stem of "Segmentation"), 14 examples were retrieved.  In addition to the 5 examples
assigned to the Principle No. 1 (i.e. "Segmantation"), there are 9 examples which are
assigned to other principles but include the keyword "Segment(ation)" in its text.  In the
displays of examples found with the keyword searching, the matched words are shown in
blue fonts.

  7.  To obtain suggestions of Principles of Invention for solving
          user's contradiction    (Principle Module)

The typical and maker-expected usage of the [Principles Module] proceeds in the
following three steps:

(1)  First, you clarify your own problem which you want to solve and input the
description of the difficulty or contradiction of the problem.

(2)  Then, the software tool shows you a few "Principles of Invention" which were used
in former inventions having similar contradiction problem, and also shows application
examples of such principles.

(3)  Finally, you are expected to solveyour own problem by using such displayed
information as the hints.

In Step (2), the software uses the "Contradiction Solving Matrix", the core knowhow of
the TRIZ methodology, as the internal logic.

7.1  Express the problem and its contradiction

Now let us examine the main display of the [Principles Module] shown in Fig. 6.1 in the
preceeding section.  At the top of this main display, [TechOptimizer Pro] requests you to
describe your own problem briefly.

      What do you want to do?
      What kind of method/means are you going to use?
      What is the problem in the process?

These descriptions are requested to input in the top three lines of the display.  However,
they are asked simply for helping you make your thought clearer by the software tool and
are never used in its internal processing logic.

Then TRIZ further requests yu to think over your problem more deeply and to clearly
find the "contradictions" in your problem.  It requests you to describe which feature (or
aspect, attribute, parameter) of the system you want to improve for your target and which
other feature of the system gets worse because of such improvement and consequently
forms a contradiction.  These features are requested to be specified in the [Improving
feature] window (at the upper left of the display) and in the [Worsening feature] windows
(at the lower left of the display), by selecting fom a list.

The TRIZ methodology is unique at the point that it does not allow you to describe these
features freely but enforces you to select the features among the preset 39 alternatives.
Both the improving feature and the worsening feature are to be selected from the same set
of features.  The list of the 39 features is shown in Table 7.1.


Table 7.1  The 39 Features (or aspects, attributes) for describing the "Contradiction
Solving Matrix"
1.  weight of moving object 
2.  weight of stationary object 
3.  length of moving object 
4.  length of stationary object 
5.  area of moving object 
6.  area of stationary object 
7.  volume of moving object 
8.  volume of stationary object 
9.  speed 
10.  force (intensity) 
11.  stress or pressure 
12.  shape 
13.  stability of object's composition 
14.  strength 
15.  duration of action of moving object 
16.  duration of action of stationary object 
17.  temperature 
18.  illumination intensity 
19.  use of energy by moving object 
20.  use of energy by stationary object 
21.  power 
22.  loss of energy 
23.  loss of substance 
24.  loss of information 
25.  loss of time 
26.  quantity of substance 
27.  reliability 
28.  measurement accuracy 
29.  manufacturing precision 
30.  object-affected harmful factors 
31.  object-generated harmful factors 
32.  ease of manufacture 
33.  ease of operation 
34.  ease of repair 
35.  adaptability or versatility 
36.  device complexity 
37.  difficulty of detecting and measuring 
38.  extent of automation 
39.  productivity 

[Operation:  Usually, first select the [Improving Feature] in the list shown in the upper
left window and click it.  Then, select the [Worsening Feature] in the lower left window.
If you watch the tool's behabior carefully, you may recognize that the order of the
[Worsening Feature] changes in responce to the selection of the [Improving Feature].
The software tool apparently uses the knowledge of frequency of these features learned
from analyses of former examples of problems, such as patents.

If appropriate, you may select the [Worsening Feature] first.  For this purpose, you
should click the small button at the left of the [Worsening feature] and select a feature in
the window. ]

7.2  Obtain "Principles of Invention" as the hints

When you specify the [Improving Feature] and the [Worsening Feature] as described
above, the software tool shows you, in the [Principles] window at the upper right of the
display, the recommendation with the heading "IM recommends:" and up-to-four
recommended "Principles of Invention".  At the lower right of the display, a list of
examples of applying the selected Principle of Invention are also shown.  You are
advised to read the Principles of Invention one by one, and to read the corresponding
application examples carefully.

Here the software tool [TechOptimizer] is saying "IM recommends: ..." on the basis of
the valuable knowhow in the TRIZ methodology.  (Note that this point is quite different
from the similar statements of "IM recommends ..." in the [Prediction Module] as
discussed in Section 5.2.)  The TRIZ methodology have analyzed a huge number of
patents; it has described the problems in the framework of the 39 Improved Features
against the 39 Worsening Features, and has found which Principles of Invention are
frequently used in innovative patents.  By the accumulation of such analyses, TRIZ has
recorded the top four Principles of Invention for each element of the 39 x 39 matrix,
which is called "Contradiction Solving Matrix" (or "Contradiction Matrix").  The table of
the matrix has been published in the textbooks, e.g. in Japanese, the Introduction of
TRIZ, Nikkei BP.  The advantage of software tool like [TechOptimizer Pro] is its
convenience in reading the TRIZ recommended Principles of Invention without referring
the large table with numbered principles.
 
 
Top of this page 5-1 Prediction module Table 5-1  Prediction tree 5-3 Example of application (old design) 5-4 Example of application (New design) Next page
6-1 principles of invention 6-2 Example of applying a principle of invention Table 7-1   39 Features

 
1. Overview 2. Trends in Tech. Evolution 3. System of Science & Technology 4. Implementing 
Tech. Goals
5. Improving User's System 6. 40 Principles 
of Invention
7. Solving Contradiction 8.  Describing User's System 9.  Functional Analysis 10.  Trimming & 
Feature Transfer
11. Recording 
& Reporting
12. Conclusion

 
Home Page Software Tools Papers and Tech. Reports Top of this Tech. Report Japanese page Japanese 
Home Page

Last updated on Feb. 25, 1999.     Access point:  Editor: nakagawa@utc.osaka-gu.ac.jp