Image Image Image Image Image Image Image Image Image Image
Scroll to top


Voice of Customers Pushed By Directed Evolution

Voice of Customers Pushed By Directed Evolution

| On 21, Jun 2002

Voice of Customers Pushed By Directed Evolution

Tan Runhua
(School of Mechanical Engineering, Hebei University of Technology,
Tianjin, 300130, People’s Republic of China)

Abstract: The three categories of customer requirements and the patterns for them to ‘pull’ or ‘push’ product design are introduced first. Technology forecasting is discussed, in which it is pointed out that the knowledge of TRIZ technology forecasting should be used in ‘push’ pattern. Voice of the customers driven by directed evolution is formed to implement the ‘push’ pattern. An engineering case shows how the voice of the customer pushed by directed evolution is applied.

Key words: Voice of customers Directed evolution TRIZ Technology forecasting

1 Introduction

Future will be different from the present and changes will certainly be happen. Technology is responsible for many of the most important changes. Forecasting future advantages in technology and their impact can be vital for top management in any manufacturing firms.

The technology forecasting, which is developed in western countries, can be divided into normative and exploratory[1-3] . The exploratory covers those techniques based upon an extension of the past through the present and into future. The normative approach starts from the future and then traces backwards to present. The main characteristics of successful methods are quantification and timescales of the parameters to describe the technology.

The TRIZ technology forecasting, which is developed in USSR by Altshuller[4-5] has different versions now: evolution of technique (ET)[6] , guided technology evolution[7] , directed evolution (DE)[8] . The main characteristics of TRIZ methods are that several patterns and paths in structure forms for evolution of technology have been abstracted from researching a lot of patents. These patterns and paths are more easily followed and used by product designers and researchers.

One of the recent trends is to integrate TRIZ with other methods to strengthen their strong points. Terninko[9] integrates QFD, TRIZ and Tanuchi to form a customer-driven robust innovation. In his model, TRIZ is responsible for solving problems and generating creative solutions. Noel Leon[10] develops a conceptual design model using QFD/FA/TRIZ, in which TRIZ is also to solve problems. Schlueter[11] uses IWB2000, which is a TRIZ based software, to introduce structure into customer requirements of QFD. Savransky[6] mentions that the knowledge of ET, which is a branch of TRIZ, should be used as the input of QFD.

Voice of the customers (VOC) is first step for product design. A methodology to obtain VOC from the knowledge of TRIZ technology forecasting integrated with Kano’s model[12] will be studied in this paper.

2 Voice of Customers

Voice of the customers (VOC) is the words and phrases that customers use to describe their wants and needs. Customers will be satisfied only with products and services that meet or exceed their wants and requirements at a price that represents value. This means that firms must actively seek out and understand what their customers really want.

Kano’s model [12] in Fig.1 shows three categories of customer requirements.

Fig. 1 Kano’s Model

  • Expected requirements are ‘self-evident’ and unspoken. 100% fulfillment of these aspects will never satisfy a customer. On contrary, when these requirements are absent or exhibit a failure, customers will be dissatisfied.

  • Revealed requirements are expressed in one-way or the other. They are not hidden. The degree of fulfillment is highly correlated with customer satisfaction.

  • Exciting requirements are not expected and are not asked for. However, should they be available by the product, customer can become very exited.

It is the most important task to identify customer requirements. There are two patterns to obtain the customer requirements, which are ‘pull’ and ‘push’. A general practice for pattern ‘pull’ is as following:

  • Directly use the product, in which designers are ‘to be the customers’.

  • Circulate questionnaires.

  • Hold focus group discussions with multiple customers.

  • Conduct interviews.

  • Go to the locations or places where the products are actually used by customers for designers to obtain information directly. This is called ‘Gemba’.

  • Look at competitors’ products.

The requirements obtained above are the raw materials and could not be used by R&D department or designers directly. After analysis of them a weighted customer requirement list is developed for use in designs. The characteristic for ‘pull’ pattern is that the wishes, wants, and needs of customers for products pull the R&D process and designs for products.

The second pattern is ‘push’. Akia Morita, founder of Sony Corporation, boasts ‘our plan is to lead the public to new products rather than ask them what they want, The public does not know what is possible, we do.’ [13] . The result is products such as the Betamax. This is a typical pattern of ‘push’. Behind this, a new idea with possible implementation of technology produced in a manufacturing firm’s internal environment pushes the design and offers the market something new for which the customers are prepared to pay. Generally, the sources of knowledge for new ideas arise outside the company and reside in experts and publications. Twiss[1] proposes a procedure to capture them as following:

  • Identification of sources of knowledge.

  • Using these sources.

  • Evaluation of the significance of the knowledge.

  • Acquisition and transfer into the company.

Savransky[6] mentions that knowledge about evolution of technique can be used in marking of innovation. If the trends and paths of technical evolution in TRIZ are used one of the method for ‘push’ pattern is formed.

3 TRIZ Technology Forecasting

If technological process consisted of a succession of random events any attempt to forecast would be impossible. Fortunately, analysis of historical data from a considerable number of phenomena shows that process is not random but follows regular patterns[1] .

The TRIZ technology forecasting developed by Altshuller[4] is a set of patterns and paths, which show the trends of technological systems evolution in structures. These patterns and paths are revealed by analysis of hundreds of thousands of invention descriptions available in the world patent databases. The most important finding is that the patterns and paths revealed in one engineering field can be transferred to other kinds of artificial systems[6] .

There are different visions of description and contents for TRIZ technology forecasting, which are all development of traditional technology forecasting of TRIZ developed by Altshuller[4] . Savransky[6] introduces Evolution of Technique (ET) as the technology forecasting in TRIZ. Fey and Rivin [7] presents Guided Technology Evolution of TRIZ. Zusman et al[8] of Ideation International Inc. describe Directed Evolution (DE) and its application. There is a little difference among these versions. This paper only applies the knowledge having been developed and DE is selected. The DE is introduced as following.

There are eight patterns of evolution according to DE, which are as following.

  • Stages of evolutions.

  • Uneven development of subsystems.

  • Increasing ideality.

  • Increasing dynamism and controllability.

  • Increasing complexity, followed by simplicity.

  • Matching and mismatching of parts.

  • Transition toward micro-level and increasing use of fields.

  • Decreasing human involvement.

There are several paths for every pattern. According to Zusman et al[8] , over 350 paths of evolution have been identified to date in DE.

DE can be shown using a tree as a sorting method in Fig.2.

Fig.2 Directed evolution tree

TRIZ-based technological forecasting has following advantages (Savransky, 2000)[6] :

  • Point to subsystems that should be improved.

  • Avoid development of subsystems that passed maturity stage of evolution or that are in decline stage.

  • Show the set of possible paths for technical development.

  • Indicate the way to build a patent fence around promising techniques at the stages of childhood and growth and destroy the patent fences of competitors at the growth and maturity stages of evolution.

The first step for product design is VOC. If the ideas produced from the knowledge of DE for a selected product are a kind of information feedback to the marketing and R&D department of a firm the decision-making will be strengthen.

4 Voice of Customers Driven by Directed Evolution

The exciting requirements of VOC are usually hidden and customers do not know them. Designers should find them and implement them in the products to be designed. The DE of TRIZ can be used to support designers to find exciting requirements. The Fig. 3 is a model to integrate Kano’s model and DE to produce exciting requirements. Here, the model is called VOC pushed by DE and includes four steps.

Step 1: Select a specific product. This is a start point for redesign. It should be selected according to long or short-term goal of the firm. If it is not necessary to carry out analysis of life cycle, for example for the redesign of a component, turn to Step 3.

Step 2: Decide its location in the S-curve. The data of the selected product is analyzed and the location in the S-curve is determined. If it is in maturity or decline stages new core technology should be found following Step 3. If it is in childhood or growth, optimization should be carried out to make improvements. The data for a product include profit with time, inventions number with time, inventions level with time and performance with time.

Step 3: Chose an advanced state. There are eight patterns and many paths of evolution for products. A state from patterns and paths is chosen as the object of design for the selected product. In this step, customers may invite to make some evaluations when choosing one from several possible states.

Step 4: Abstract customer requirements. Three categories of customer requirements are abstracted from the selected advanced state for the product together with the questionnaires and gemba. The exciting requirements are mainly abstracted from the advanced state of the product. Also, customers should be asked to give their ideas. The output of this step is a customer requirement list for the following design stages.

Fig. 3 VOC pushed by DE

5 Engineering Case

A simple example is selected as an engineering case here.

One type of fixture for machine manufacturing is produced in a firm of China, which is shown in Fig.4. A clasping mechanism, which is a subsystem of a fixture used in machine center, is a screw mechanism. The mechanism is operated by an operator’s hand. The speed for clasping and releasing a workpiece is slow and not suitable for the mass production of mechanical elements. It should be improved to suit fast operations. The following are the four steps to obtain new customer requirements for its redesign.

Fig. 4 A fixture

Step 1: Select a specific product

In the firm, different types of fixtures are produced. In one type as shown in Fig.4, the middle staged one in volume has a great market potential. It is selected as the specific product and start point for the redesign.

Step 3: Chose an advanced state

Because it is needed to redesign the clasping mechanism and the fixture body is not needed to be changed, turn to the Step 3.

There are eight patterns and many paths in AD. Designers should try different patterns and paths and determine one pattern and one path and one or two states in the path as the final selected general concept of the solution to be searched. This process is done and pattern 4,one path in that pattern are selected.

Pattern 4 of the DE is ‘increasing dynamism and controllability’. One of the evolution paths under this pattern is shown in Fig. 5.

Fig. 5 Path of ‘increasing dynamism and controllability’

Now the product is in state of ‘one joint’ in the path. And advanced state should be in ‘liquid’, which is decided by detailed analysis. ‘Liquid’ here means oil hydraulic system, in which a hydraulic cylinder as a ‘tool’ should be used for the reason of reducing dimension.

Step 4: Abstract customer requirements

The hidden exciting requirement is ‘fast clasping and releasing’. In designer’s mind it should be implemented using a hydraulic system.

For this simple product, other two kinds of requirements can be obtained from marketing department of the firm or from the designers themselves directly. Table 1 shows the three kinds of requirements.

Table 1 Three kinds of requirements

Exciting requirement Fast clasping and releasing workpiece
Revealed requirements Low cost High reliability Small volume Easy operation
Expected requirements Produce and keep clasping force Ability for pre-adjustment Easy maintenance

The list of requirements can be as the input information of the following design steps, such as, as the input of the HOQ (house of quality) of this product.

6 Conclusions

Two patterns ‘pull’ or ‘push’ for product designs are defined. In ‘push’ pattern, the new idea with possible implementation of technology for product design is produced in a firm’s internal environment.

TRIZ technology forecasting presents the structure evolution of product and the knowledge of evolution is potential resources of ‘push’ pattern for a firm.

A new procedure to obtain voice of the customers driven by directed evolution, a version of TTRIZ technology forecasting, is developed, in which Kano’s model and directed evolution are integrated. It is an implementation tool to obtain the list of customer requirements, including the exciting requirements for the product to be designed.

A simple engineering case shows how the procedure is applied in practice.


The author is grateful to the Chinese National Natural Science Foundation (No. 50175025), Natural Science Foundation of Tianjin (No. 003804611), for funding this work.


  1. Twiss B., Managing technological innovation, Fourth edition, Pitman Publishing, London,1992

  2. Twiss B., Forecasting for technical decisions, Peter Peregrinus, London, 1992

  3. Frauens M W, Improved selection of technically attractive projects using knowledge management and net interactive tools. MS thesis, MIT, 2000

  4. Altshuller G, The Innovation Algorithm, TRIZ, systematic innovation and technical creativity, Technical Innovation Center, INC., Worcester, 1999

  5. Tan Runhua, ‘Design for innovation – TRIZ: theory of inventive problem solving,'(in Chinese), China Machine Press, 2002

  6. Savransky S D, Engineering of creativity, CRC Press, New York, 2000

  7. Fey V.R. and Rivin E.I., Guided technology evolution (TRIZ Technology Forecasting),
  8. Zusman A., Zlotin B., Zainiev G., An application of directed evolution,
  9. Terninko J, The QFD, TRIZ and Taguchi connection: customer-driven robust innovation, TRIZ Journal, Jan., 1998,
  10. Noel Leon-Rovira, A new model of the conceptual design process using QFD/FA/TRIZ, TRIZ Journal, July, 1998,
  11. Schlueter M, QFD by TRIZ, TRIZ Journal, June, 2001,
  12. Cohen Lou, Quality function deployment. Addsion Wesley Longman, New York,1995

  13. Barabba, V. and G. Zaltman, Hearing the voice of the market comp. advantage through the creative use of market info., Harvard B.S. Press, Boston, MA., 1991

Tan Runhua, Professor, Ph.D (Zhejiang University, China). He used to study in Brunel University of UK as a visiting scholar and Munich University of Applied Science of Germany as a guest professor. His research interests are design engineering, CAD, RP/RT, fluid power transmission and control, software engineering. He has published more than 100 technical papers and holds several patents now. He has studied and applied TRIZ for four years and solved several problems for manufacturing firms. He has published the first book to introduce TRIZ using Chinese in China and has made several reports at both conferences and for some firms to introduce TRIZ in China.