We all believe that TRIZ is a powerful tool. But how should it be used?

TRIZ can make a critical difference in the development of new devices. However, there are two problems that commonly crop up:

Let us tackle the second problem first. Yes, TRIZ can be very helpful in solving problems on the production line. But is that the best time to apply it? The application of TRIZ on the production line makes TRIZ a part of the problem-reaction approach. Problem reaction is an outmoded approach that we should minimize. The productive trend of the last 20 years has been to migrate from problem reaction to problem prevention. We want to include TRIZ in the vanguard of the problem-prevention approach. The main role for TRIZ is in the development of new technology.

Now to the first problem. TRIZ is very powerful, but many other activities have to be done very well also. Other tools that we want to combine with TRIZ are QFD, functional analysis, reusability planning, object-oriented approach, Pugh concept evolution and selection, parameter design and tolerance design (Taguchi Methods), design inspections, and robust testing. One-dimensional improvements always fall far short. The complete set of balanced tools is needed for success.

We do Technology Development in four phases: (1) Technology Strategy, (2) Concept Generation and Selection, (3) Robust Design, and (4) Selection and Transfer. TRIZ is a powerful tool for technology strategy and concept generation. Technology Development is the critical application for TRIZ. It is the natural time for the invention of new concepts for performing functions. When TRIZ is combined with other tools, especially robust design, then a flow of successful new technologies can be expected.

In the early 1990s TRIZ started to emerge from the former USSR. Initially there was much hype, but little content. Now thanks to organizations such as the Altshuller Institute the content matches the hype.

So now we all accept that TRIZ has great and unique capabilities. This conference has reports of many great successes. However, TRIZ has had little impact in the countries with advanced economies.

TRIZ obviously has tremendous potential to greatly speed up the development of non-incremental new technologies. What is stopping the widespread use of TRIZ? Two things:

TRIZ has too often been portrayed as a silver bullet that is all that is needed to bring success. There is not any silver bullet, TRIZ or otherwise. TRIZ must be used in its most productive roles, and it must be integrated into a productive process. TRIZ must be surrounded by other productive process elements that will work together with TRIZ to achieve success. This integrative framework will be presented here as Total Technology Development.

TRIZ has implicitly been following a nucleation and growth approach to implementation. Although this approach has a role, by itself it will never succeed.

A separation of technology development from product development and its transfer into a distinct and clearly defined technology stream having clear interfaces to a distinct product stream as illustrated in Figure 1 has been proposed by Clausing [1994] and further developed by Schulz [1998] .

Figure 1. Separate Product and Technology Stream (adapted from Clausing [1994] )

Technology insertion into product programs as illustrated in Figure 1 (‘…fishing out of new technologies…’) is facilitated whenever new technologies are ready, that is to say have reached an appropriate level of superiority, robustness, and maturity. Thus TTD will further enable rapid technology integration into product programs and reduce the quality, cost and schedule uncertainty within those programs. Only technologies that are robust and mature will be transferred into a product program. This greatly improves reliability, and reduces the problem-solving phase of commercialization.

A reduced technology development cycle time associated with early technology robustness and maturity makes superior and mature technologies rapidly eligible for a transfer into product programs. Overcoming the time/maturity gap within technology development (Figure 2) is key to successful systems development, because it enables the company to introduce superior technologies earlier than any competitor. Thus TTD will provide faster and better technologies to markets.

Figure 2. Speeding up the S curve [Schulz and Clausing, 1998]

TTD is called ‘total’ because it aims at capturing and incorporating all critical aspects of technology development by being based on a systems engineering perspective.

New technologies improve both the product and the production of the product. We also sometimes speak a little more broadly of new technologies for product and technology development, such as the content of this paper for example. However, here we restrict ourselves to the more physical product and process technologies.

In general different research and development activities cover a very wide range, from very basic research creating new knowledge or even new areas of research, to typical product development activities taking merely existing technologies and rapidly putting the resulting product into the marketplace. For a better classification of the framework proposed within the context of this paper, four basic areas of research and development are distinguished and illustrated in Figure 3.

Figure 3. Four Types of R&D Activities (adapted from Clausing [1999] )

The first two areas of fundamental and applied research aim at generating new knowledge or new technological capabilities. Technology development aims at achieving fundamental improvements based on the knowledge or technological capabilities provided by the first two types of activities. The fourth type, commercialization, finally aims at transferring the improvements provided by technology development into the marketplace, and is therefore equivalent to the type of activity usually called product development.

The viewpoint of Total Technology Development was presented as Chapter 7 of my book Total Quality Development [Clausing, 1994] , and has subsequently been enlarged in theses that I supervised by Armin Schulz, Michael Frauens, Clemens Bauer, and Siddhartha Sampathkumar. Schulz further developed the complete scope of Total Technology Development, and integrated TRIZ into it. Frauens worked on the technical aspects of technology strategy, with a strong emphasis on TRIZ. Bauer worked on the complete technology strategy, with an emphasis on real option theory. Sampathkumar worked on radical or disruptive technologies. The works of these students will be referred to in this paper simply by their name. The relevant documents are listed in the References section at the end of this paper.

Chapter 7 of Clausing [1994] and the further enhancement of Schulz integrated the following methodologies:

• Various Approaches to Technology Strategy (see [Porter, Roper, et al. 1991] , [Lowe, 1995] , and [Betz 1987, 1998] )
• Theory on Inventive Problem Solving (TRIZ) (see [Fey, 1998] , [Fey and Rivin, 1997] , and [Altshuller 1984] )
• Concept Evolution and Selection (see Pugh [1990, 1996] )
• Enhanced Quality Function Deployment (EQFD) (see [Clausing, 1994] and [Cohen, 1995] )
• Robust Design (see [Taguchi, 1993] and [Phadke, 1989] )
• Function Analysis (see [Akiyama, 1991] and [Sontow, 1993] )

The methods were assigned to the distinct phases of a basic technology development process proposed by Clausing [1994] , which is illustrated in Figure 4.

Figure 4. Four stages of technology development.

The top-level process consists of four major phases, each containing specific sub-processes, which are all discussed in detail in Schulz. The four phases of the top-level process are as follows:

Phase (1) Integrated Technology Strategy
Phase (2) Concept Work
Phase (3) Robustness Development
Phase (4) Technology Selection, Transfer, and Integration

TRIZ is applicable to Phases 1 and 2. Therefore, here we will concentrate on these two phases. (There is research now in progress to also apply TRIZ to robustness work. It would include conceptual robustness during the concept work.)

Technology strategy determines the most important long-term technological needs. New technologies are aimed at improving the performance of some function. Typical types of improvements are (1) functionality, (2) reliability, (3) cost, and (4) size. Functionality means that the function is being done qualitatively better – brighter colors for a printer for example. Reliability, cost, and size are self-explanatory.

Technology strategy decides which functions we will work to improve, and what our primary approach will be. There are a series of decisions to be made.

1. How interesting is the function to customers?
2. Can we improve it very much?
3. Can our competitors improve it very much?

If the answers to these three questions are negative, we probably do not want to work to improve the technology.

If the answer to just the first question is that customers do not much notice the function, then a new technology will not have much financial benefit. As an example let us consider the paper feeder for a printer. A paper feeder cannot add to the customers’ perception of the functionality of the printer. It could detract from it by damaging the paper, but this is not a significant problem. If the feeder’s unreliability, cost, and size are all small fractions of the total system, then improvement will not be much noticed by the customers. Then there is little motivation for new technology.

If the function is interesting to customers – print quality for example, then we address questions 2 and 3. Here we have to be careful not to succumb to the famous tendency to believe that everything has been invented. In the long run there will always be the possibility of some new invention. So we use questions 2 and 3 to prioritize our technology development for the next N years, where N might be five years for example. We try to select functions where improvement seems probable and the improvement will make a difference to our customers. The improvement is especially imperative if our competitors are advantaged over us.

To be more precise we want to do the following:

1. Screen all functions to select those worth further consideration for improvement
2. Develop alternative possible development paths, estimate the improvement that would be achieved for each development path, and estimate the probability of actually achieving the estimated improvement
3. Estimate the market benefit of the improvement. This will be strongly influenced by the competitive situation
4. Apply an investment model to select the most lucrative paths for us to actually work on

The initial screening is done by the application of two graphs (Schulz), Figures 5 and 6.

Figure 5 is used to determine if the product is important. More specifically it is used to determine which parameters of the important products are important.

Then Figure 6 is used to determine which technologies have a critical contribution to the important parameters of the important products.

Study of Figures 5 and 6 will reveal the logical way to use them. For more detail see Schulz.

Often this preliminary screening is all that is needed. Certain technologies will be obvious choices for further development.

Figure 6. Analysis of technological position.

When more analysis is required, we proceed through the remaining three steps.

After we decide that a new technology might be commercially beneficial, then we have to see if any is likely to be available. We have several paths available to us:

1. Develop present S curve

1. New invention (or imported new technology) at lower system level
2. Improve robustness
3. Detailed design improvements

2. Move to new S curve

1. New invention
2. Bring in external technology

We see that invention can move to a new S curve, or contribute to moving up the existing S curve. For example, if the initial screening identified the paper feeder of a copier/printer as a ripe opportunity for further technological development, then we could invent a feedback control system that controls the stack force, which would improve the reliability of the feed function. This was actually done at Xerox in the late 1970s. This can be thought of as a new S curve for the stack-force function, but as moving up the S curve for the complete friction-retard paper feeder.

A representative part of the functional tree for a friction-retard paper feeder is shown in Figure 7.

Figure 7. Portion of functional tree for friction-retard paper feeder.

The decision was to implement a new invention that moved the function provide stack force to a new S curve.

Figure 8. New S curve at lower level moves the higher-level function up its S curve.

Going from S curve 1 to S curve 2 for the lower-level function provide stack force moved up the S curve for the higher-level function feed sheet from the original value O to the improved value L-L I (lower-level invention).

Another alternative that was developed in Xerox in the late 1970s is the vacuum corrugation feeder (VCF). This might be initially regarded as a new S curve, different from the S curve for the friction-retard feeder. However, the VCF had predecessors that were vacuum feeders. Therefore, the VCF can be best thought of as further development of the S curve for vacuum feeders. This too was done by moving to a new S curve at a lower system level (the C in VCF).

Moving to a new S curve for the complete paper feeder might be using electromagnetic forces. This has been used to hold down the paper on a plotter. As far as I know, no one has invented a concept to successfully use it for a paper feeder.

We can reformulate the generic development paths as follows:

1. Invention
2. Robustness improvements
3. Detailed design improvements
4. Bring in external technology

Usually the last three options can be evaluated adequately by known engineering approaches. (More about robustness later.) Here we concentrate on invention.

So what paths are open to us for making inventions? Here we follow after Frauens (2000). He developed a broad approach that features TRIZ tools and the use of the Internet to bring in information. Frauens developed the following table.

These steps are done with the help of TRIZ.

The final result of this process is a set Y of development paths that we can reasonably follow. Each path has an estimated improvement X that following it would achieve, and a probability p that the improvement will be achieved in the time outlook of our strategy. From a technical viewpoint this information (Y,X,p) is all that is needed for completion of our technology strategy.

Once we know (Y,X,p) for all paths that we can identify as promising, we next need to evaluate the likely market impact of each path. The initial screen has already given us considerable insight. In many cases it will be sufficient.

The essential question is: If we succeed in achieving improvement X in the time frame that we used to calculate the probability of success p, how much will our market share increase? In economics this is called a utility function. This approach has been outlined by Bauer.

Here we will outline a simple approach that is used in industry. If we have been in business with this product for some time, we can use our data to develop a customer preference model. This model says that if we improve some function by X, then our market share will be increased by a certain amount. This is essentially a utility function without the trappings of economic theory.

Now we need to apply an investment model to determine if our improvement is worth the investment. We must consider that the market benefits will not occur immediately, but will only start after the new technology is in production, and will then be further delayed by the usual diffusion and substitution time cycles.

Traditionally the investment model was done by using ROI based on some discounted cash flow (DCF). However, this has come under attack.

There is now a trend to use real option theory as the investment model. This is financial option theory applied to real things, such as technology development.

The justification for using real option theory instead of ROI/DCF is simple. The latter assumes that we make one decision, and some time much later learn the result.

Development work does not proceed that way. We make many intermediate decisions. If the development is not progressing well, we can cancel the project. We don’t have to spend the remaining investment.

Putting it in slightly different terms, we invest in an initial bit of work. Then we have a better idea of the probability of success p. We can say that we have bought an option to keep developing the idea if it shows promise. If p keeps getting bigger fast enough at the successive decision points, we keep going. Otherwise we don’t invest any more.

That in a nutshell is the rationale behind the use of real option theory to make decisions about investment in technology development. More details are outlined by Bauer.

Now that the Technology Strategy has focused our work, it is time to do creative work. Creative work is done this way:

1. Generate concepts

1. Understand needs
2. Apply known effects, or combinations of effects, that promise to meet the needs

1. Known engineering devices
2. Known scientific effects

2. Evolve and select the best concepts

First I will discuss how this is done without TRIZ, and then bring in the contributions of TRIZ.

Many, probably most, attempts at generating new concepts go off the tracks by not understanding the needs. Once the need is not understood, it is almost certain that at best the creativity will meet some other need that is not useful in the present context.

The technology strategy presents the needs in broad terms. It is critically important to focus on the specific function that is to be improved. Then it is important to know the vector of improvement that will be most beneficial. For example, if the technology strategy has identified that the main need is better reliability, then concepts that primarily reduce cost will be nice, but not in the direction of greatest benefit. The technology strategy identification of the needs is usually primarily used for funding decisions. When the concept work starts we go into the needs in greater detail.

Beyond understanding the function to be improved, and the customer-observable parameters that will be most beneficial to improve, we need to understand the current situation of the physics. There is some system element that performs the function that we want to improve. How does it work, and what is its role in the higher level system. What are the values of critical functional parameters, and what do these values suggest as fertile directions for improvement? Much weak creativity is based on a “hand waving” understanding of the current physics.

In summary, understanding the needs means:

1. Know what function is to be improved
2. Know which customer-observable characteristics that are associated with the function are most in need of improvement
3. Understand the current physics (more generally the current details of operation)

Once the needs are well understood, then we work to apply known effects that might meet the needs. We might say, “Ah, ha, a feedback control system would satisfy the need.” Or “we could make this part out of two different materials; it doesn’t have to be uniform in its material properties.” Or we might say, “The inelastic strain is only 0.1%; because it is so small we can remove it by a simple process.” Contrary to some assumptions, understanding the current physics quantitatively often leads to an invention.

At Xerox in the late 1970s we had a type of paper feeder called the friction-retard feeder. It worked well, but its reliability was not good enough for high-volume copiers and printers. We decided not to change the basic concept, but to make improvements at lower levels of the system, as shown in Figure 8. One lower-level function was apply stack force. When the stack force was too small, a misfeed would occur. However, when the stack force was too big, a multifeed would occur. So we wanted a small stack force, but we wanted a large stack force. Much experimental data and analysis had revealed that a stack force of only 0.3 pounds would successfully feed the top sheet most of the time – more than 95% of the time. And such a small stack force would essentially never cause a multifeed. So what to do about the small percentage of the feeds that would misfeed when the stack force was only 0.3 pounds?

Armed with this data, a feedback control system became a fairly obvious invention. Put a sensor at a critical point in the paper path. If the top sheet did not reach that point by a critical time, then use an actuator to increase the stack force to 0.7 pounds. I received a patent on this, and it worked very well. You might look into your Xerox copier or printer to see if it is still there.

In summary, note five points:

1. A clear decision was made to not work to improve the function of the total feeder system, feed sheet, but to work to invent at a lower level of the system, apply stack force
2. The customer-observable characteristic that most needed improvement, reliability, had been clearly identified
3. The detailed data of the existing situation were very well known. If the small stack force of 0.3 pounds had only fed successfully 50% of the time, then the feedback control system would not have been feasible.
4. It was recognized that the force of 0.7 pound was only needed at certain times
5. A known engineering effect, feedback control, was applied.

Many times we are happy to develop one promising concept. However, a more complete approach is to develop several or many concepts, and then evolve and compare them until convergence is achieved to one or two concepts to carry forward into new products. The process that has proven highly successful for doing this is the Pugh Concept Evolution and Selection Process.

Figure 9. Pugh Concept Evolution and Selection Matrix for a Gyroscopic Suspension
[Khan, 1989] .

Figure 9 shows a real Pugh matrix [Khan, 1989] for a gyroscope suspension. There were 15 concepts. Later a hybrid concept, 12A, was added in the last column. There were 18 criteria in the row headings, as shown below:

One concept was chosen as the reference or datum concept. For the first run this was concept 1. All of the other concepts were compared to the datum for each criterion, and judged to be +, -, or essentially the same S. The matrix was run two more times, with first concept 8 as the datum, and then concept 12 as the datum. The team evolved the concepts and converged to hybrid concept 12A.

Four of the concepts are shown in Figure 10. Note that they are all drawn to the same level of detail – just enough to represent the basic physics by which the function would be performed.

This process has a very strong track record in industry. Detailed descriptions of the Pugh process are in [Clausing, 1994] and [Pugh, 1990, 1996] .

Stuart Pugh emphasized that creative work had to be done in a creative environment. The point cannot be emphasized too much. If we go in to work in the morning, and have 13 bureaucratic task to complete before we leave, how creative will we be? Or if new ideas are constantly treated with derision, will the inventions flow? The creative environment is essential.

Everything that has been said about Concept Work up to here has been ante TRIZ. How has TRIZ changed Concept Work? The whole point of this conference is that TRIZ has greatly changed the best approach to Concept Work.

However, note that everything that has been said so far is still relevant in the world of TRIZ. We still can benefit from everything that has been described. Now TRIZ helps us to do some steps. We still use the basic process that has been given:

1. Generate concepts

1. Understand needs
2. Apply known effects, or combinations of effects, that promise to meet the needs

1. Known engineering devices
2. Known scientific effects

2. Evolve and select the best concepts

Now I need some help from all of the TRIZ experts at this conference; i.e., everyone at the conference except me. It seems to me that TRIZ basically does three things:

The Sufield represents a model of the minimum system necessary to perform a certain function, consisting of an object, a tool, and a field [Fey, 1998] . The function is comprised of the function object and the function interaction (e.g., feed sheet), represented by the noun and the verb in the function statement. The design parameter to be selected is comprised of a tool (i.e. feed belt) and the field (e.g., stack force + friction). While the tool indicates a hardware concept, the basic physical principle to perform the function is defined by the selected field. This is illustrated in Figure 11.

Figure 11 helps us to see the relationship between the sufield analysis and the functional tree (Figure 7). The functional tree is used with a hardware tree (see my book Total Quality Development). So both the sufield model and the traditional functional tree present the same information.

The functional tree has proven very useful in the development of better concepts. By concentrating attention on the function product-development people have moved away from the devices in the Rube Goldberg cartoons that were popular in the Sunday paper when I was young. We might say that the movement was in the direction of increasing ideality.

There are invention principles that are associated with the sufield model that are useful in guiding invention. For everyone who has made good use of the functional tree, the sufield model is an opportunity for further enhancement of one’s methodology.

Figure 11. The Sufield Model, with a paper feeder as an example.

In the case of the paper feeder we wanted the force to be both big and small. This is a conflict that was alleviated by the feedback control system. In general the identification of the conflict in the requirements provides great help in identifying the need. This is a key feature of TRIZ. An attempt to completely eliminate the primary conflict has been shown to be a major contributor to creativity.

Note that in the case study of the stack force in the Xerox paper feeder I separated the needs in time. I didn’t think to myself, “Try separation of needs in time.” I just did it, because once I understood the needs in detail, that option occurred to me. This isn’t surprising because Altshuller cataloged what inventors were already doing.

So we might say that TRIZ isn’t teaching us to do anything that is completely new. It is helping us to do familiar things in a more systematic way with a much higher hit rate. It helps us to leverage all previous experience.

In the case study I applied the known effect of a feedback control system. From a TRIZ viewpoint feedback control systems are probably an indication of a lack of ideality.

TRIZ concentrates on the application of known scientific effects. This has the potential to create new types of devices that are not currently among the known engineering effects, such as feedback control systems.

Robustness keeps the performance of the system close to the ideal function. After the concept work, the new technology works well under favorable laboratory conditions. The challenge is to make it work well, close to ideal, under all conditions.

Every system is controlled by different parameters during its operation. Phadke [1989] has introduced the P diagram, Figure 12.

Signal Factors are set by users or operators to gain the intended response of the system.

Noise factorsare factors, which decrease the system performance and cannot be controlled. There are three types of noise factors.

1. External (environmental conditions)
2. Internal (variations within the production process)
3. Deterioration (performance degradation during system lifecycle)

Control Factors are the critical functional parameters whose valuesare set by the system designer, and which strongly influence the system sensitivity due to the noises. To achieve the targeted performance of the system, there are two major methods within robust design.

1. Parameter Design, does not affect cost issues, by focusing on determining best control factor values for lowest system sensitivity against noise.
2. Tolerance Design and Specification, does affect cost issues, by focusing on reducing the tolerance range for noise factors through applying higher quality parts.

An adjustment of control factors usually doesn’t influence the cost of the system, therefore they are key to robustness and have to be determined to their best values. Tolerance Factors and Adjustment Factors are specific types of control factors. While the adjustment of tolerance factors will influence the cost of the system, setting of adjustment factors only influences systems performance, but not its variation.

Figure 12. Classification of Parameters influencing a System [Phadke, 1989]

The Response is the intended output of the system, which is expected to be within the intended range by the user. Any deviation from the target value causes quality loss according to the quality loss function.

The ideal function is the ideal relationship between the signal and the response. It is:

Parameter Design optimizes the nominal values of the critical functional parameters that control the performance of the system. The important steps of parameter design are:

The key to success it so introduce the noises early:

The engineers’ natural tendency is to protect their new invention from noises. It is critically important to overcome this, and introduce the noises early.

As an example, the Xerox paper feeder that has been described before was moved much farther up its S curve by a very thorough improvement of its robustness, from L-L I to R, Figure 13. This has produced a very robust paper feeder that was put into production in 1981. It has since been flexibly applied in many copiers and printers over a wide range of speeds without the necessity for any additional development.

Figure 13. Farther up the S curve with robustness.

Robustness achieves the fundamental reliability of the new invention. It greatly facilitates technology transfer, flexibility, and reduced commercialization time.

After the technology is successfully developed there is a natural tendency to believe that the job is finished. Nothing could be farther from the truth. Technology selection and transfer are often difficult or insurmountable hurdles.

There are several criteria, which have to be considered for a new technological concept to be eligible into product programs. In general four aspects have to be fulfilled:

• Superiority
• Robustness
• Flexibility
• Maturity

In selecting the technologies for a new product, the Pugh concept selection is used again. Now the alternatives that are considered are new technologies, current technologies, and outside purchase.

Basically each transfer of technologies has a multi-dimensional perspective, as already pointed out in a study of the International Space University [ISU, 1997] . A successful transfer of technology consists of four major aspects.

• Know-What, that is the knowledge about the technology to be transferred
• Know-Who, that is the ‘knowledge carrier’ having all elements of the knowledge necessary for the transfer
• Know-Why, that is the theoretical base of the technology in terms of its basic and fundamental knowledge
• Know-How, that is the processes, methods, or boundary conditions necessary for a successful application of the transferred technology

Clausing [1994] identified a ‘unfortunate tendency towards rivalry between the product engineering division and the technology organization’. A cultural barrier, ‘not invented here’ and other dysfunctions lower the probability of a successful technology transfer, which is illustrated in Figure 14 (upper part), and therefore endanger a successful technology development. Those problems are mainly based on competitive instincts within the organization and have to be overcome by the right organizational environment.

Thus the most successful approach to technology transfer into product programs is based on the ‘transfer of people’. Clausing [1994] proposes the establishment of a technology transfer team, which is consisting of engineers from product engineering and technology development. This will furthermore enhance the necessary knowledge transfer for incorporating new technologies into product programs and is illustrated in Figure 14 (lower part). The acceptance of new technologies can even be increased, by assembling the appropriate development teams throughout the entire technology development process.

Figure 14. Enhancing Technology Transfer by Transferring People (adapted from [Clausing, 1994] )

The current approach to the implementation of TRIZ – nucleation and growth – will not be successful. TRIZ must be integrated with the total process – this paper has one recommendation – and implemented by a major improvement activity that is led from the top of the organization.

TRIZ is most productive during the development of new technology. It is most successful when it is imbedded in the Total Technology Development process, associated with other powerful approaches such as Robust Design.

Much of this paper is based on the theses of the students who I supervised 1998 – 2000: Armin Schulz, Michael Frauens, Clemens Bauer, and Siddhartha Sampathkumar.

For my small understanding of TRIZ I am greatly indebted to Victor Fey. Victor is responsible for my understanding, while I am responsible for the smallness of it.

C. Bauer:

C. Bauer, D. Clausing, C. Whitcomb, An Integrated Decision Model for Strategic Evaluation of the Viability of New Technologies, Working Paper, CIPD, Massachusetts Institute of Technology, Cambridge, 2000.

C. Bauer, D. Clausing, A. Schulz, Improving the Valuation of Research and Development: An Integrated Decision Framework, EuSEC 2000, Munich, Germany, September 2000.

F. Betz:

F. Betz, Managing Technology – Competing through new Ventures, Innovation, and Corporate Research, Prentice Hall, New York City, 1987.

F. Betz, Managing Technological Innovation, John Wiley & Sons, New York City, 1998.

C. Christensen, The Innovators Dilemma – When New Technologies Cause Great Firms to Fail, Harvard Business School Press, Boston, 1997.

D. Clausing, Total Quality Development – A Step by Step Guide to World Class Concurrent Engineering, ASME Press, New York, 1994.

D. Clausing, Presentation on Technology Development, CIPD, Massachusetts Institute of Technology, Cambridge, 1999.

L. Cohen, Quality Function Deployment – How to make QFD work for you, Addison Wesley Publishing, 1995.

V. Fey, Theory of Inventive Problem Solving (TRIZ), The TRIZ Group, Seminar Handout given at Massachusetts Institute of Technology, 1998.

V. Fey and E. Rivin, The Science of Innovation – A Managerial Overview of the TRIZ Methodology, Book 1 – The Technology of Innovative Engineering Series, The TRIZ Group, 1997.

M. Frauens, Improved Selection of Technically Attractive Projects Using Knowledge Management and Net Interactive Tools, M.S. Thesis, Massachusetts Institute of Technology, Cambridge, 2000.

ISU, Technology Transfer – Bridging Space and Society, International Space University (ISU) Summer Session, Design Project Report, Houston, 1997

M. Khan and D. G. Smith, Overcoming conceptual barriers – by systematic design, Proceedings of the International Conference on Engineering Design, 1989.

P. Lowe, The Management of Technology – Perception and Opportunities, Chapman & Hall, 1995

M. Phadke, Quality Engineering using Robust Design, Prentice Hall, 1989.

A. Porter, T.A. Roper, et al., Forecasting and Management of Technology, John Wiley & Sons, New York, 1991.

S. Pugh, Total Design, Addison Wesley Publishing, 1990.

S. Pugh, Creating Innovative Products using Total Design, Eds. D. Clausing and R. Andrade, Addison Wesley Publishing, 1996.

S. Sampathkumar:

S. Sampathkumar and D. Clausing, Developing Non-Incrementally New Technologies, Working Paper, CIPD, Massachusetts Institute of Technology, Cambridge, 2000.

S. Sampathkumar, D. Clausing, A. Schulz, E. Fricke, New Technology Development – The Joint Role of Product and Technology Development, EuSEC 2000, Munich, Germany, September 2000.

A. Schulz:

A. Schulz, D. Clausing, H. Negele, and E. Fricke, Development And Integration Of Winning Technologies As Key To Competitive Advantage, Systems Engineering (INCOSE Journal), pp. 180 – 211, Vol. 3, Issue 4, 2000.

A. Schulz, D. Clausing, H. Negele, and E. Fricke, Total Technology Development – Shifting the View in System Development, Proceedings of the 9th Annual Symposium of INCOSE, Brighton, UK, 1999a, pp. 1243-1251.

A. Schulz, D. Clausing, E. Fricke, and H. Negele, Shifting the View in Systems Development – Technology Development at the Fuzzy Front End as Key to Success, 1999 ASME DETC, 11th International Conference on Design Theory and Methodology, Las Vegas, 1999b, DETC99/DTM 8773.

A. Schulz and D. Clausing, An Embedding Process Framework for Total Technology Development, Working Paper, CIPD, Massachusetts Institute of Technology, Cambridge, 1998.

K. Sontow and D. Clausing, Integration of Quality Function Deployment with further Methods of Quality Planning, Working Paper, Laboratory for Manufacturing and Productivity, Massachusetts Institute of Technology, Cambridge, 1993.

G. Taguchi, Taguchi on Robust Technology, ASME Press, New York, 1993.

Author’s Biography

Don Clausing joined the MIT faculty in 1986 after working in industry for three decades. He created the first MIT course on product development. It integrated basic concurrent engineering, Taguchi methods, QFD, Pugh concept selection, technology readiness, reusability, and effective management in a comprehensive development process to achieve lower manufacturing cost, higher quality, and shorter development times than are currently standard in the United States. Clausing’s book, Total Quality Development, was published by ASME Press in March 1994. In 1996 he was a cofounder of MIT’s Center for Innovation in Product Development (CIPD). Clausing retired from MIT in the summer of 2000.