This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commerci

このページでは http://www.triz.com.cn/docs/20160920142947956033.pdfをHTMLに変換して表示しています。

変換前のファイルはこちらから確認できます。

※Netscape4.0と4.7では正しく表示されない場合があります。ご了承ください。

※HTMLバージョンとして表示する際、レイアウトが崩れたり、文字が読めなくなる場合があります。ご了承ください。

Yahoo! JAPANはページ内のコンテンツとの関連はありません。

Page 1

This article appeared in a journal published by Elsevier. The attached

copy is furnished to the author for internal non-commercial research

and education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

http://www.elsevier.com/copyright

Page 2

Author's personal copy

UXDs-driven conceptual design process model for contradiction solving using CAIs

Runhua Tan *, Jianhong Ma, Fang Liu, Zihui Wei

Institute of Design for Innovation, Hebei University of Technology, Tianjin 300130, China

1. Introduction

It is generally agreed that the conceptual design is the most

critical phase in the design processes [1]. The inputs for this phase

are product or design specifications, and the outputs are principle

solutions or concepts [2]. There are many research results in this

field, which may be divided into three types: studies on process

models; methods for generating ideas or concepts; and computer

applications. The process models [2–5] for conceptual design are

descriptions at some level for real conceptual design processes in

industrial firms. Pahl and Beitz’s model [2], as a case in point,

includes seven steps. Many methods to generate ideas or concepts

[6] have been developed, such as brainstorming, mind mapping,

lateral thinking, etc. These methods are basically brainstorming-

driven. Different types of methods [7,8] to assist designers to

generate concepts are continuously studied. Computer systems [9–

11] have been applied in conceptual design processes. In these

systems the lack of formal product representations of function,

behaviour and structure [12], which are the knowledge base for

conceptual designs, have been identified as a kind of shortcoming.

Contrary to the brainstorming-driven methods, theory of

inventive problem solving (TRIZ) [13] is a systematic method to

guide designers on a high plane to generate ideas for inventive

problems [14]. Today, several computer-aided innovation systems

(CAIs) [15], such as Goldfire Innovator, IWB, and InventionTool,

have been developed and commercialized to support designers in

conceptual design. The basic theory for developing these systems is

TRIZ. Several knowledge bases are included in the CAIs, which are

abstracted from patent analysis or different scientific branches. By

these knowledge bases, designers have a chance to apply the cases

of other designers and the effects from the scientific world.

Nevertheless, how to integrate the TRIZ and CAIs into a conceptual

design process is also a problem.

Design situation is a particular state of interaction between

designers and the environment at a particular point in time [16].

CAIs are new systems for most designers and will change the

design situations when they are applied. In this new situation, the

interactions mainly happen among designers and a serial of

interfaces of CAIs. The reason that the interfaces will support

designers to generate ideas is needed to be studied.

Contradictions are a kind of inventive problems in conceptual

design. Contradiction matrix in TRIZ is a tool for solving that kind of

problems. The matrix was developed many years ago [13], but it is

still applied today [17,18]. The matrix does help designers to solve

some contradictions faced during the design. This study will be

restricted to find solutions for contradiction solving. The target is

to form a new conceptual process model under the environment of

CAIs.

2. Contradictions in conceptual design and solving them

using CAIs

Pahl and Beitz [2] divided a design process into four phases. The

conceptual design is a phase or process to develop the principle

Computers in Industry 60 (2009) 584–591

A R T I C L E I N F O

Article history:

Available online 4 July 2009

Keywords:

Unexpected discovery

Contradiction solving

Conceptual design

Computer-aided innovation

A B S T R A C T

Design is situational, which means the explicit consideration of the state of the environment, the

knowledge and experience of the designer, and the interaction between the designer and the

environment during designing interact. When computer-aided innovation systems (CAIs) are applied to

the design, the environment and the situation are different from the traditional design process and

environment. The basic principles of some CAIs available in the world market are directly related to

theory of inventive problem solving (TRIZ). Special TRIZ solutions, which have a few inventive principles

and the related cases for contradiction problem solving, are medium-solutions to domain problems. The

second stage analogy process is used to generate domain solutions and in this process, the TRIZ solutions

are used as source designs of analogy-based process. Unexpected discoveries (UXDs) are the key factors

to trigger designers to generate new ideas for domain solutions. The type of UXDs for the specific TRIZ

solutions is studied and a UXDs-driven contradiction solving for conceptual design is formed. A case

study shows the application of the process step by step.

* Corresponding author. Tel.: +86 22 60204037; fax: +86 22 60204037.

E-mail address: rhtan@hebut.edu.cn (R. Tan).

Contents lists available at ScienceDirect

Computers in Industry

journal homepage: www.elsevier.com/locate/compind

doi:10.1016/j.compind.2009.05.019

Page 3

Author's personal copy

solution. Several sub-processes are divided, which are an abstract

of the essential problems, establishment of the function structure,

search for suitable working principles, and combinations of these

principles, to form a working structure. Some difficult problems,

which exist during conceptual design, may be as follows:

(1) Adoption for a suitable principle solution of a function in a

function structure that will result in the existence of a new

harmful function, or intensifying an existing harmful function.

(2) Elimination or reduction of a harmful function in a function

structure will deteriorate a useful function.

(3) Intensification of a useful function or reduction of a harmful

function will cause the unacceptable complication of the

design.

All of the above problems are technical contradictions from the

point of view of TRIZ [13]. As a result, designers in conceptual

design face some difficulties to solve these contradictions. The tool

in TRIZ for solving technical contradictions includes 39 generic

engineering parameters to represent contradictions; 40 inventive

principles to solve the contradictions; and a matrix to find a few

suitable inventive principles. Under each inventive principle there

are several design cases abstracted from patent bases of the world.

Every case shows a result to solve a contradiction by former

designers. Fig. 1 shows the structure of the tool.

Every case under an inventive principle in Fig. 1 is the result of

analyzing patent bases from outside world. The knowledge, which

is tacit in different domains of a patent base, is difficult to have

them applied by designers because it is a problem to find a useful

one from different domain. If a patent abstracted from any domain

is stored in the case base of TRIZ it becomes explicit knowledge or

codified knowledge, which can be found following the TRIZ

problem solving routine and applied for idea generation.

The TRIZ world in Fig. 1 has been programmed as a kind of

arithmetic and a module of CAIs, such as in the Goldfire Innovator

and InventionTool, which are cases of CAIs. Interaction between

TRIZ world and outside world is realized by the interfaces of the

CAIs. Fig. 2 shows a used cases model to describe an interface of a

module for contradiction solving in InventionTool. The application

of different CAIs based on TRIZ has made the TRIZ more powerful

and applicable. There are a few knowledge bases in CAIs. The

knowledge is arranged by the framework of TRIZ. By applying the

knowledge base, the design cases from different industrial fields

are accessed by designers.

In the knowledge base of CAIs, a case for a technical contra-

diction solving is described using a sketch with text to explain the

working principle of that sketch. Fig. 3 shows an example of a case

in which the sketch shows the working principle for cutting with

microwave heating and the text is the explanation. When one

principle as a TRIZ special solution is selected, all the cases relevant

to that principle can be browsed one by one. New ideas for the

domain solutions may be formed from designers’ mind during the

browsing process.

3. An analogy-based concept generation for contradiction

solving

The process of solving inventive problems using TRIZ is shown

in Fig. 4. The TRIZ process is supported by CAIs, in which the TRIZ’s

special solutions, inventive principles and the cases related to the

principles, are produced automatically after TRIZ mapping. The

designer can browse them one by one as shown in use case model

in Fig. 2.

From the TRIZ special solutions to domain solutions is an

analogy-based process. Analogies are partial similarities between

different situations that support further inference. Analogy-based

conceptual design (ABCD) means the application of an analogy to

generate a new concept in the process of the conceptual design. In

this process, the existing designs and the designs to be carried out

are source designs and goal designs, respectively. One of the

Fig. 1. Contradiction solving model in TRIZ.

Fig. 3. A cases in knowledge base of CAIs.

Fig. 2. Use cases for contradiction solving in InventionTool.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

585

Page 4

Author's personal copy

conditions to carry out ABCD is the existence of source or base

designs in different domains in large number. In the situation of the

conceptual design supported by CAIs, there are many source

designs because every case included in a TRIZ special solution,

which is stored in the case bases of CAIs, is a source design.

Fig. 4 shows two mappings from a domain problem to domain

solutions, which are from domain problem to TRIZ special solution

and from TRIZ special solution to domain solution. Both mappings

are analogy processes [19], which are the first stage and the second

stage analogy process, respectively [19]. The first stage analogy

process is completed by the application of CAIs and the outputs are

TRIZ special solutions. The second stage analogy process is a process

of analogy-based conceptual design, which is a human-based

process.

Suwa et al. [20] have developed a concept ‘‘situated-

invention (S-invention)’’, which means a designer generates

the issue or requirement for the first time in the current design

task in a way situated in the design setting. Gero et al. [16,21–

23] have studied the generation of S-invention, and summarized

a design process. Firstly, the designer apperceives the domain

problem and determines source design and goal design. Then the

designer finds unexpected discoveries (UXDs) through the

matching of source design and goal design; UXDs are transferred

to goal design by mapping, and a new goal is generated, and

then the modified goal design is produced. There may be multi-

source designs, and the last modified goal design is the concept

of solving domain problems through modifying goal designs

continually.

Gero and his group have majored in architectural design, so the

source designs are drawings of different kinds of architectures. If

the source designs are substituted by TRIZ special solutions the

design process for generation of S-invention can be applied to

generate the domain solutions in TRIZ-based design processes.

Designers find several UXDs and modify goal design depending on

their design experience, their comprehension of domain problems,

and the situation. At times some modified goal designs are domain

solutions. The macro-process of ABCD for contradiction solving

using CAIs and TRIZ is shown in Fig. 5.

The contradiction analogs are the technical contradictions

selected from the 39 engineering parameters, which have similar

meanings to the domain contradictions. The UXDs solving

contradiction analogs are found from the principles solving

contradiction analogs and cases.

The main processes of implementation of the process in Fig. 5, is

how to find UXDs in solving the contradiction analogs, and

converting these UXDs into ideas for solving the domain contra-

dictions. UXDs enlighten designers on invention and make new

ideas appear, so discovering and transferring UXDs is the key to

success. According to the concept of constructive memory [21], the

memory is not a direct reappearance of a former experience but a

function of a former experience, which changes after producing

these experience and situation of memory requirement. An UXD is a

‘‘new’’ perceptual action that has a dependency on ‘‘old’’ physical

action(s) [20]. This means that if a designer traces or pays attention

to the existence of source designs, the perceptual action is an

instance of UXD. Experiences and UXDs drive designers to generate

new ideas. The new ideas are mapped to the goal design to produce a

new goal design. The last few goal designs are the solutions for a

conceptual design.

Perceptual actions [20] are operated for architectural drawings.

They must be extended to TRIZ special solutions for TRIZ-based

design. Ideas from the constructive memory driven by UXDs are

developed to form concepts. Concepts are described using working

principles, combinations of these principles to form a working

structure or system working principles, or sketches.

4. UXDs from TRIZ special solutions

Suwa et al. [20] have divided UXDs into three types, depending on

the types of visuo-spatial feature, which are the discovery of a visual

feature; a spatial or organizational relation among more than one

previously drawn element; and a space that exists in previously

drawn elements. The types are suitable for the design of an

architecture, in which the basic elements are dots, lines, rectangles,

circles, arrows and so on. For complex system design, such as

complex mechanical system design, the basic elements are more

complex. The types divided are not suitable for them. New types are

needed.

Currently, there are several well-known design theories and

methodologies, such as Systematic Design Methodology [2],

Axiomatic Design Theory [24], and General Design Theory [25].

The common feature for all these theories and methodologies is the

function-based design. The first step for the design is to transform

the design specifications or the product needs to a function model,

such as a function structure. Then, the structure model of the

design, such as a working principle or a sketch, is developed from

function model by mapping.

There are two kinds of mapping: function–structure mapping

and function–behaviour–structure mapping, as shown in Fig. 6.

The former is suitable for the mapping of extrinsic functions and

the latter is suitable for intrinsic functions. There are different

kinds of functions, behaviours and structures [26–28]. Functions

are divided into atomic, source, destination and transfer functions.

Behaviours are divided into three kinds, continuous-time-beha-

viour, discrete-time-behaviour, and state-transition-behaviour.

According to the specific design context, a structure may refer

to a sub-system, a sub-assembly, a component, a feature, or a

geometric entity, and a physical relationship.

Fig. 4. The process of solving problem using TRIZ.

Fig. 5. The macro-process of ABCD for contradiction solving using TRIZ.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

586

Page 5

Author's personal copy

A structure of a design implicates all the information about the

behaviour and the functions for that design. If a TRIZ special

solution with some cases is applied as source designs for analogy,

the designers will find the explicit behaviours and functions,

except for structures. As a result, some functions, behaviours, or

structures stimulate the designers and the new concepts are

appeared in their minds. Therefore, the implicit functions,

behaviours and structures in TRIZ special solutions in different

levels are UXDs for designers when TRIZ and CAIs are applied. The

UXDs here are different from types of the definitions applied in

Gero’s group.

For technical contradiction problem solving, forty the inventive

principles in TRIZ will be applied. One principle selected from the

forty has implicit meaning, which will be a kind stimulus for

designers. For example, if a circle is applied for a product design, in

order to make some modifications by the inventive principle 6

‘asymmetry’, designers may image different forms to substitute

the circle, as shown in Fig. 7. As a result an inventive principle for

technical contradiction solving is also a UXD, when it is selected.

There are four types of UXDs when TRIZ special solutions are

applied as resource designs under the situation applying CAIs. The

first type is to find an inventive principle and the others are to find

a function, a behaviour, or a structure in different levels. Table 1

shows the types, definition of each type, and the instances of each

type and how to find an UXD.

For the designers using CAIs, the physical actions to find a UXD

are only looking, which means designers viewing the computer

screen: the principles of contradiction solving and the cases shown

by pictures and contexts. By viewing actions, designers discover

the UXDs implied.

5. An UXDs-driven process model for conceptual design

A top-model for conceptual design is shown in Fig. 8. The inputs

are design specifications and outputs are selected solutions.

However, the process is supported by CAIs. This implicates that

the new ideas for the solutions are developed by the stimulation of

TRIZ special solutions.

There are several models for conceptual design process from

literatures. Refs. [2,3], and [6] are examples of these. In this study,

Pahl and Beitz’s model [2], as a traditional and human-oriented

process model, is selected as a base to extend from. Fig. 9 shows a

new model of conceptual design process which is driven by UXDs

and supported by CAIs.

According to Fig. 9, the process is divided into eight steps, which

are as follows:

Step 1: Establish or modify function structures. For original

design [2], the function structure of the design is established

from the design specification directly. For adoptive design [2],

the existing function structure is modified to adopt the changed

specifications.

Step 2: Search for working principles for sub-functions. The

working principles for every sub-function in the existing

function structure are found by searching.

Step 3: Analyze the working principles of sub-functions and find

whether there are technical contradictions or not. If there are no

contradictions turn to step 7. Otherwise continue the process.

Step 4: Identify contradictions. Analyze all the working

principles again and identify all the contradictions clearly.

There may be multi-contradictions and they need to be

identified one by one.

Step 5: Find TRIZ special solutions. For each contradiction, the

TRIZ special solutions are found under support of CAIs.

Step 6: Find UXDs for every contradiction. Designers analyze

the selected inventive principles, browse the cases one by one

and find UXDs implied.

Step 7: Form solutions. There are two sub-processes in this step.

One is following step 3 and the other is continuing from step 6.

(1) Combine every working principle of sub-functions into a few

system working principles, named working structures.

(2) Find new working principles for sub-functions which relate to

the contradictions and eliminate all the contradictions. Then,

combine every principle into a few system working principles.

Step 8: Evaluate solutions. Identify one or two working

structures as the outputs of conceptual design processes.

Table 1

Types of UXDs.

Types

Definition

Instances

How to find

UXD-1

An inventive principle which is suitable

for solving the contradiction faced

One to four inventive principles

Check the matrix using two

engineering parameters

UXD-2

A function that one case implicated

Atomic, source, destination and transfer functions

Physical actions

UXD-3

A behaviour that one case implicated

Continuous-time-behaviour, discrete-time-behaviour,

state-transition-behaviour

Physical actions

UXD-4

A structure that one case implicated

A sub-assembly, a component, a feature,

or a geometric entity, and a physical relationship

Physical actions

Fig. 7. Some images from a source using an inventive principle.

Fig. 6. Function–behaviour–structure and function–structure mapping.

Fig. 8. Conceptual design supported by CAIs.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

587

Page 6

Author's personal copy

Step 5 belongs to the first stage analogy process and step 6 to 7(2)

belongs to the second stage analogy process. The second stage

analogy process is a UXDs-driven process. The evaluation is included

in step 8, and many methods have been developed for this

application.

6. Case study

Dropping pills, produced by dropping pill machines, are a kind

of Chinese traditional medicine. After dropping and drying, the

pills should be put into little bottles for selling in the market. The

machine to put the pills into bottles is a kind of packaging machine.

There are no standard machines of this kind. The machines have

been developed by one or two firms in China. New operating

principles for the machines are needed to be produced by other

firms of medicine production.

The sub-functions of the machine are distributing pills, dischar-

ging pills, bottling and lidding. The working principle of existing pills

packaging machines is shown in Fig. 10. Certain amount of pills are

carried by pills board fixed on a tumbling cylinder. When running

the tumbling cylinder, the pills are transferred, discharged, and then

put into bottles. The pills board which is not full of pills will be

detected by a counting sensor array. The bottles, which are filled

with unqualified amount of pills, will be removed after lidding.

There are two problems in the existing design of pills packaging

machines:

(1) The sub-function of distributing certain amount of pills is

implemented by a pills board. This results a difficulty in

changing the amount of packaging. The flexibility of the

machine is poor.

(2) The amount of pills is detected twice in the original design,

which leads to a large number of photoelectric counters.

The model in Fig. 9 is applied to form a new working principle

for this kind of machines.

Step 1: Establish or modify function structures.

To solve the above problems, the original function structure for

the existing product is modified as shown in Fig. 11. In order to

express the modified structure clearly, some sub-functions of

converting energy is omitted. In Fig. 11 the sub-function of

distributing pills is substituted for the sub-function of distributing

certain amount of pills, the pills are detected once, and the sub-

function of counting pills is added.

Step 2: Search for working principles for sub-functions.

To implement the modified functions, i.e. distributing pills, a pills

board is added, and a tumbling cylinder in Fig. 12 is adopted to

distribute pills. To adapt to this change, counter array correspon-

dence with pills board is replaced by a single counter. But the

counter in the tumbling cylinder cannot implement the sub-

function of counting pills. New position of the counter is set on top of

filler as shown in Fig. 13.

Step 3: Find contradictions.

Contradiction 1: distribute pills and discharge pills.

Fig. 9. An UXD driven process model for conceptual design.

Fig. 10. The working principle of existing pills package machine.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

588

Page 7

Author's personal copy

The pills are distributed in the several circles. The implementa-

tion of discharging pills is based on the principle of distributing

bills. Fig. 14 shows a principle for discharging pills that is used at

present. When the roller is redesigned and used inside the cylinder,

a transmission mechanism is also needed, which leads to the

complexity of the device.

Contradiction 2: distribute pills and transfer bottles filled with

pills.

Transferring bottles requires a pause in distributing and

discharging pills. Tumbling cylinder must be stopped accurately

and frequently. This is difficult and harmful.

Step 4: Identify contradictions.

Contradiction 1 defined as a technical contradiction between

adaptability (No. 35) and complexity of a device (No. 36).

Contradiction 2 defined as a technical contradiction between

adaptability (No. 35) and complexity of control (No. 37).

Step 5: Find TRIZ special solutions.

InventionTool 3.0, which is CAIs, is applied in this step. The

module ‘contradiction solving’ in InventionTool 3.0 is used to find

TRIZ special solutions. Select the improved parameter ‘adapt-

ability’ and worse parameter ‘complexity of a device’, then, the

interfaces shows the TRIZ special solutions of contradiction 1,

which are No. 29 (Pneumatic or hydraulic construction), No. 15

(Dynamicity), No. 28 (Replacement of mechanical system), No. 37

(Thermal expansion). The four principles and the relevant cases in

the case base are the TRIZ special solutions.

By the same process, special solutions of contradiction 2 can be

found, which is No. 1 (Segmentation) and No. 15 (Dynamicity).

Step 6: Find UXDs for every contradiction.

For the solutions of contradictions the designers may find several

UXDs and generate several ideas under the stimulations of UXDs by

browsing the cases in InventionTool 3.0. Fig. 15 is a case, which

shows the principle of a component used in a kitchen machine. The

ball moves up under the pressure of air flow to releases the

Fig. 11. Modified function structure.

Fig. 15. A case in no. 15.

Fig. 12. Tumbling cylinder.

Fig. 13. New position of the counter.

Fig. 14. A principle for discharging pills.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

589

Page 8

Author's personal copy

exhausted gas produced during frying. The principle implies an UXD,

which is the ball moving under the pressure of the exhausted gas.

The UXD is a kind of behaviour, which is an UXD-3 in Table 1.

Fig. 16 is a case in InventionTool 3.0 that shows adjusting the

pipe’s hydraulic pressure by using valves dynamically. This case

implies an UXD for the solutions of contradiction 2.

Step 7: Form solutions.

Solutions of contradiction 1: transfer UXDs in Fig. 15 to goal

designs. Convert the UXD into the new ideas and generate a

principle for discharging pills. The pill moves under the pressure of

airflow. A possible structure for this principle is shown in Fig. 17.

Solutions of contradiction 2: to solve contradiction between

distributing pills and transferring bottles, a valve is used to control

the falling of pills, which is similar to the valve in Fig. 16. When the

required amount of pills fall into the bottle, the valve turns off to

ensure that no pills fall during the shifting of bottles. A possible

structure for this principle is shown in Fig. 18.

7. Conclusions

CAIs show all inventive principles and cases, which are the

source designs of ABD. The databases of CAIs are a fruit of TRIZ

researchers for many years, which have broad applicability. The

application of the database will improve the validity of ABD that

has been extensively accepted by designers.

When TRIZ is applied to solve a contradiction in design the first

and second stage analogy processes exist. The results of the first

stage analogy process are source designs of the second stage

analogy process. To find UXDs from the sources is the key step to

generate successful ideas for innovation. Four types UXDs have

been divided. The physical action for finding UXDs on the

computer screen of CAIs is by viewing.

A human-oriented eight-process step model is formed for

conceptual design, in which UXDs are the driving force for

generating new ideas. Designers find UXDs from the TRIZ special

solutions and react with experiences that designers have to

construct memories suddenly. Then new ideas for domain solutions

are formed.

The model put forward is only related to contradiction solving

of TRIZ. It needs to be extended to technological evolution, effects,

and standard solutions of TRIZ, in order to be effective in future

application of CAIs.

The eight-step model is suitable for the experienced designers.

For the inexperienced designer, a detailed or formal model is

needed. This is a future work.

Acknowledgements

We would like to thank the reviewers for their valuable

comments, which have greatly improved the presentation of this

paper. The research is supported in part by the Chinese Natural

Science Foundation (grant numbers 50675059) and by National

Innovation Project (grant number 2008IM30100). No part of this

paper represents the views and opinions of any of the sponsors

mentioned above.

References

[1] H. Lipson, M. Shpitalni, Conceptual design and analysis by sketching, Journal of

Artificial Intelligence in Design and Manufacturing 14 (2000) 391–401.

[2] G. Pahl, W. Beitz, Engineering Design—A Systematic Approach, 2nd edition,

Springer, London, 1996.

[3] Y. Jin, P. Chusilp, Study of mental iteration in different design situations, Design

Studies 27 (2006) 25–55.

[4] O. Shai, Y. Reich, D. Rubin, Creative conceptual design: extending the scope by

infused design, Computer-Aided Design (2007), doi:10.1016/j.cad.2007.11.004.

[5] J.S. Gero, U. Kannengiesser, The situated function–behaviour–structure frame-

work, Design Studies 25 (2004) 373–391.

[6] K.N. Otto, K.L. Wood, Product Design, Prentice Hall, Upper Saddle River, NJ, USA,

2001.

[7] O. Benami, Y. Jin, Creative stimulation in conceptual design, in: ASME 2002 Design

Engineering Technical Conference (DETC2002/DTM-34023), Montreal, Canada,

2002.

[8] G.R. Weber, S.S. Condoor, Conceptual Design using a Synergistically Compatible

Morphological Matrix, in: Proceedings of the 28th annual frontiers in Education,

1998.

[9] P. Ociepka, J. Swider, Object-oriented system for computer aiding of the machines

conceptual design process, Journal of Materials Processing Technology 157/158

(2004) 221–227.

[10] S.C. Feng, E.Y. Song, Information modeling of conceptual design integrated with

process planning, in: ASME Int Mechanical Engg Congress and Exposition, vol. 19,

Orlando, Florida, (2000), pp. 123–130.

[11] K. Yoshinobu, N. Washio, Y. Koji, Functional metadata schema for engineering

knowledge management, in: First Workshop FOMI 2005: Formal Ontologies Meet

Industry, Castelnuovo del Garda (VR), Italy, 2005.

[12] S. Szykman, R.D. Sriram, W.C. Regli, The role of knowledge in next-generation

product development systems, ASME, Journal of Computing and Information

Science in Engineering 1 (2001) 1–11.

[13] G. Altshuller, The Innovation Algorithm, TRIZ Systematic Innovation and Tech-

nical Creativity, Technical Innovation Center Inc., Worcester, 1999.

[14] D. Cavallucci, P. Lutz, F. Thie�baud, Methodology for bringing the intuitive design

method’s framework into design activities, Proceedings of the Institution of

Mechanical Engineers. Part B: Journal of Engineering Manufacture 216 (2002)

1303–1307.

[15] S. Kohn, S. Husng, A. Kolyla, Development of an Empirical based categorisation

scheme for CAI software, in: 1st IFIP TC-5 Working Conference on CAI, ULM,

Germany, 2005.

[16] G.J. Smith, J.S. Gero, What does an artificial design agent mean by being situated?

Design Studies 26 (2005) 535–561.

[17] G.C. Robles, S. Negny, J.L. Lann, Case-based reasoning and TRIZ: a coupling for

innovative conception in Chemical Engineering, Chemical Engineering and Pro-

cessing 48 (2009) 239–249.

Fig. 18. Structure for controlling falling of pills.

Fig. 16. Another case in no. 15.

Fig. 17. Structure for discharging pills.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

590

Page 9

Author's personal copy

[18] G. Cascini, P. Rissone, Plastics design: integrating TRIZ creativity and semantic

knowledge portals, Journal of Engineering Design 15 (4) (2004) 405–424.

[19] R.H. Tan, Process of two stages analogy-based design employing TRIZ, Interna-

tional Journal of Product Development 4 (1/2) (2007) 109–121.

[20] M. Suwa, J.S. Gero, T. Purcell, Unexpected discoveries inventions of design

requirements, Design Studies 21 (2000) 539–567.

[21] J.S. Gero, Constructive memory in design thinking, in: G. Goldschmidt, W. Porter

(Eds.), Design Thinking Research Symposium: Design Representation, MIT, Cam-

bridge, (1999), pp. 29–35.

[22] J. Kulinski, J.S. Gero, Constructive representation in situated analogy in design, in: B.

Vries, H. Achten (Eds.), CAAD Futures 2001, Kluwer, Dordrecht, 2001, pp. 507–

520.

[23] J.S. Gero, Concept formation in design: towards a loosely wired brain model, in: L.

Candy, K. Hori (Eds.), Strategic Knowledge and Concept Formation Workshop,

Loughborough University of Technology, 1997, pp. 135–146.

[24] N.P. Suh, Axiomatic Design—Advance and Application, Oxford University Press,

New York, 2001.

[25] H. Yoshikawa, General Design Theory and a Cad System, North Holland Publish-

ing Company, Amsterdam, 1981.

[26] F. Zhang, D. Xue, Distributed database knowledge base modeling for concurrent

design, Computer-Aided Design 34 (1) (2002) 27–40.

[27] D. Xue, H. Yang, A concurrent engineering-oriented design database representa-

tion model, Computer-Aided Design 36 (2004) 947–965.

[28] R.B. Stone, K.L. Wood, Development of a functional basis for design, Transactions

of the ASME, Journal of Mechanical Design 122 (4) (2000) 359–370.

Runhua Tan is currently a Professor in the School of

Mechanical Engineering and Vice-President of Hebei

University of Technology. He received his MS and PhD,

both in Mechanical Engineering from HUT in 1984 and

from Zhejiang University in 1998, respectively. He

worked as a Visiting Scholar at Brunel University (UK)

from 1994 to 1995 and had a 3-month stay at Munich

University of Applied Science (Germany) in 2001. He

has authored 2 books, and authored or co-authored

over 210 refereed papers. He holds eight patents and

seven software copyrights. His research interests

include design theory and methodology and innova-

tion management.

Jianhong Ma is a Professor at the School of Computer

Science and Software Engineering of Hebei University

of Technology in Tianjin, China. Her research fields

include software engineering, software design technol-

ogy, and AD in software design. She has presented more

than 20 high level papers during the last 3 years. Prof.

MA and her software development group have been

working in developing Innovation Tools for more than

10 years, and have released Computer-Aided Invention

Design Software System-InventionTools version 1.0,

2.0, and 3.0.

Fang Liu is a member of the Institute of Design for

Innovation, School of Mechanical Engineering, Hebei

University of Technology. She received her MS in

Mechanical Engineering from Hebei University of

Technology (China) in 2005. Currently, she is pursuing

a PhD in Mechanical Engineering, at Hebei University of

Technology. Her research interests include mass

customization, product platform and innovation

design.

Zihui Wei is a doctoral candidate of the Institute of

Design for Innovation, School of Mechanical Engineer-

ing, Hebei University of Technology. He received his MS

in Mechanical Engineering from Naval University of

Engineering (China) in 2002. His research interests

include innovation design and electromechanical inte-

gration.

R. Tan et al. / Computers in Industry 60 (2009) 584–591

591