Get e-book Innovative Conceptual Design: Theory and Application of Parameter Analysis

Free download. Book file PDF easily for everyone and every device. You can download and read online Innovative Conceptual Design: Theory and Application of Parameter Analysis file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with Innovative Conceptual Design: Theory and Application of Parameter Analysis book. Happy reading Innovative Conceptual Design: Theory and Application of Parameter Analysis Bookeveryone. Download file Free Book PDF Innovative Conceptual Design: Theory and Application of Parameter Analysis at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF Innovative Conceptual Design: Theory and Application of Parameter Analysis Pocket Guide.
About This Item

  1. Sridhar Condoor - Google Scholar Citations
  2. Integrating CAD, TRIZ, and Customer Needs
  4. Upcoming Events

Email address for updates. My profile My library Metrics Alerts. Sign in. Get my own profile Cited by View all All Since Citations h-index 17 10 iindex 23 Articles Cited by. Robotics and Computer-Integrated Manufacturing 15 3 , , Research in Engineering Design 24 2 , , Research in Engineering Design 25 4 , , Articles 1—20 Show more. The requirements may be modi- ed in the later stages of the design process; however, this would require approval from all the parties involved in the product devel- opment and may pose complications.

An effort should therefore be made at the front end, that is, during need analysis, to consider all the downstream issues and compile the best possible set of require- ments. On the other hand, requirements should be challenged by everyone involved in the design process, as more knowledge and understanding of the problem domain is gained. For example, con- sider the following scenario regarding the Mars sample return mis- sion described earlier.

The system integration group may analyze the initial need by studying old photographs of the Martian surface taken by several spacecrafts in the s, and may generate a Innovative Conceptual Design 36 requirement for a Martian ground vehicle that is capable of travers- ing obstacles as big as one meter across. This requirement is dictated to the vehicle design team, which concludes that the resulting vehi- cle would be prohibitively large. The vehicle designers, therefore, challenge the original requirement by negotiating with the integra- tion team, which subsequently modies the requirement to state that the vehicle should be able to traverse obstacles that are 0.

I What particular characteristics is the new product required to have? I How will I know that Im done? Need identication is the step where a possibly nontechnical and sometimes congurational design task is converted to a real need. This need is extensively studied during need analysis to iden- tify both functions and constraints in ve general categories: per- formance, value, size, safety, and special.

Throughout need analysis, the designer should be objective and careful to avoid creating solu- tions or developing a bias through false constraints. This will ensure maximizing the size of the solution space. Analyzing the need also deepens the designers understanding of the task and its domain. This will be extremely helpful later in the design process and is facilitated by being very quantitative and spe- cic. The knowledge gained during need analysis, together with its summary as design requirements, is carried over to the conceptual design stage.

Here, satisfying all the requirements serves as the crite- rion for completion of conceptual design. Moreover, a good set of design requirements will provide measures for the goodness of the design, thus facilitating conception of an optimal solution, not just a solution. Need Identication and Analysis 37 The approach presented in this chapter is a methodology for need identification and analysis.

It is not a step-by-step procedure as are the following two alternative approaches. Quality Function Deployment QFD has been touted in recent years as a struc- tured method of developing design requirements.

Sridhar Condoor - Google Scholar Citations

It leads the designer through several steps, including listing of customers requirements and benchmarking the competition. Alternatively, some textbooks provide comprehensive checklists of generic areas and issues that should be considered, and they ask the designer to choose the relevant points and fill-in the blanks. Both of these approaches may be used to supplement the methodology of this chapter. Several design congurations may be used to bridge across a body of water, such as the English Channel or the Hudson River.

Write a general need statement for this task, and examine how the need is met by bridges, causeways, tunnels, and ferries. What is the real need satised by videocassette recorders VCRs? Identify competing technologies and describe the advantages and disadvantages of each one. Write a need statement and list the inputs and outputs for the black-box model of the following products: a Clothes washer b Computer mouse c Voltage transformer d Gearbox e Automobile brakes f Automobile tires 4. The evolution of tires involved several signicant changes: a From bare metal wheels to solid rubber tires b From solid rubber to pneumatic tires with inner tubes Innovative Conceptual Design 38 c From tubed tires to tubeless bias-ply tires d From bias-ply tires to bias-ply belted and then to radial-ply tires Investigate the evolution of tires to determine the needs satised by the different congurations and the performance trade-off issues in their design.

What are the performance requirements for an ideal automobile tire? Identify the trends in the laptop computer market with the goal of developing a new laptop computer in the next year. Using these trends, determine a target weight, size, screen size, and res- olution for the new laptop. Identify the important safety considerations in the design of a food processor. A popular paper is: Hauser, J. The House of Quality. Harvard Business Review, 66, No.

Total Quality Development. The following two textbooks use QFD as a tool for understanding the design task and developing the requirements dubbed engi- neering specications : Dixon, J. Ullman, D. The Mechanical Design Process. New York: McGraw-Hill, Engineering Design: A Systematic Approach. London: The Design Council, Innovative Conceptual Design 40 In sand-lled electrical fuses, the packing factor of sand controls the overall fuse performance. This chapter discusses an ill-dened, per- ceived need involving the packing factor. The methodology outlined in the previous chapter is applied to dene the real need by establish- ing the scope of the design task at the proper level.

The need is then analyzed in order to fully understand and quantify the key issues. The results of the need analysis are summarized as a set of design require- ments. The case study demonstrates a denitive link between good need identication and analysis and the potential for innovation. They are charac- terized by their interrupt rating, which is the maximum current that a fuse can stop while maintaining its mechanical integrity. Fuses that have high interrupt rating are lled with sand.

A schematic of a sand-lled fuse is shown in Fig. The components are: I Fuse element: made of silver with one or more weak spots depending on the interrupt rating. I End-caps: connect the fuse in an electrical circuit. A hole is provided in one of the end-caps to allow sand lling. The external protrusions are the electrical terminals of the fuse. I Sand: lls the space in the casing around the fuse element. I Fiberglass casing: houses the fuse element and the sand. I Plug: closes the hole in the end-cap after sand lling.

  1. Enlightened empiricism: an examination of W.V. Quines theory of knowledge;
  2. Emergent Curriculum in Early Childhood Settings: From Theory to Practice.
  3. Innovative Conceptual Design | Books | Conceptual design, Design theory, Design.
  5. The Culled (The Afterblight Chronicles: American Sequence, Book 1).

I Pins: attach the end-caps to the casing. When the current in the circuit exceeds the prescribed limit, the weak spots in the fuse element are designed to melt and create a dis- continuity. An arc forms across the discontinuity and continues to conduct current. During this arcing process, the sand serves two key purposes: 1. It melts due to the intense heat of the arc and absorbs energy equivalent to the heat of fusion.

As a result, it diminishes the amount of energy available for heating the gases in the cas- ing and thereby reduces the possibility that the casing will explode due to expanding gases. The porosity of the sand lling allows the molten metal to escape from the arcing interface. This reduces the time to interrupt the circuit. Thus, the sand lling helps to maintain the mechanical integrity of the fuse while reducing the interrupt time.

A short interrupt time is desirable because it decreases the load on various components during the short-circuit condition. The PF is dened at the macroscopic level as the fraction of space occupied by sand grains in the fuse cavity. As the PF decreases, the interrupt time increases and becomes unacceptable at 0. A fuse manufacturer asked that a quality control device be designed and incorporated in the production line to identify faulty fuses. The stated initial need was to check the PF of each fuse before shipment. For this particular task, the desired output is the PF. The fuse is both the input and the output for the black box as shown in Fig.

Based on this model, the need can be formulated as determine the PF of the fuse. Now we must ask ourselves whether the need statement is not too constraining; that is, whether it will permit us to conceive innovative solutions. This particular need statement suggests that the PF be determined after the fuse is completely assembled.

This limits the range of possible solutions to just a few nondestructive evaluation techniques. On the other hand, if the implied constraint the PF must be determined only after assembly is removed, then the designer is free to conceive more innovative solutions. For instance, the PF can be determined based on measurements taken during the manufacturing process. Thus, the same need can be formulated at a more generic level as determine the PF during the manufacturing process.

If we can determine the PF before closing the end-cap with the plug, then it may be possible to add more sand to achieve the desired PF. This raises the interesting question of whether we want merely to nd the PF or whether we actually want to control it. The latter Need Identication and Analysis Case Study 43 choice expands the scope of the design task, so the new need can be stated as ensure the correct PF during the manufacturing process. A black box for this particular need is shown in Fig.

I What are the sources of variation of the PF? I What is the ideal range for the PF? I What are the production rate and volume of the fuses? I What is the cost of a fuse? I How much would the fuse manufacturer be willing to pay for the designed equipment? I How large can the equipment be? Next, we attempt to develop a thorough understanding of the func- tions and constraints involved with this design task, under the ve categories of performance, value, size, safety, and special.

Performance The primary function of the design is to ensure the correct PF. For fullling this goal, it is important to identify: I The maximum possible value for the PF. I The design variables that determine the PF. I The primary sources of variation in the magnitude of the PF. To determine the PF, we can assume the sand grains to be incompressible spheres.

From material science we know that the PF is dependent on the spatial arrangement and independent of the grain size; that is, two different grain sizes with the same spatial arrangement will give the same packing factor. For any grain size, the PF can range from 0. However, including interstitial grains can increase the PF.

Now let us understand the design variables that determine the PF. As discussed earlier, the PF is the ratio of the volume of the sand to the volume of the cavity. The volume of the sand is proportional to its mass. The volume of the cavity depends on the inside diame- ter of the casing and the length of the cavity. The mass of the sand can be determined by subtracting the mass of all the components from the total mass of the fuse.

However, due to manufacturing tolerances, the mass of the fuse components can vary signicantly, as shown in Table 3. Similarly, dimensional tol- erances on the casing and the assembly process can cause the cavity volume to change. These variations are summarized in Table 3. The two tables reveal that if the total weight of the fuse is used for determining the PF, then the variation in the mass of the end- caps and in the length of the cavity can result in signicant error.

The mass variation of other components and the casing diameter variation result in much smaller errors. Value The value of the new system would be the increased quality of fuse production. Quality Loss Function QLF is a systematic and rational methodology for estimating loss of quality due to off-target performance in monetary terms. This methodology will be used here to analyze and quantify the value of the new design.

We know that ensuring a good PF results in high-quality fuses. As the PF deviates from the desired value, the quality of the product decreases, and this, in general, manifests itself as a loss to society. These losses can be due to the product failing to deliver on-target performance, to harmful side effects of the product, and to down- time of the equipment. Component Mean g Standard Deviation g Fuse element 0.

For this problem, we can assume the desired PF to be 0. The quality loss of a fuse is zero i. As the PF deviates from the desired value, the quality loss increases. Customers take eco- nomic action when the quality loss becomes unacceptable to them. The customer tolerance 0 is defined as the value of y at which half of the customers would take economic action. Based on an infor- mal survey of fuse customers, half of the customers who have fuses with a PF of 0.

Therefore, the customer tolerance was set at 0. To quantify the loss, we need to estimate the value of the qual- ity loss coefcient k. This can be done by identifying the loss A 0 at the customer tolerance of the PF. In reality, the estimation of loss is quite complex since the cost should include the downtime and the replacement of damaged equipment for a typical application.

The average qual- ity loss can be computed as:. The last equation shows that the quality loss can be decomposed into two parts: 1. Quality loss due to the mean being off target. Quality loss due to the variance. The quality loss can be reduced by bringing the mean PF close to the target 0. Figure 3. Even though this is not a cash ow to the manufacturer, it is realized in terms of greater customer retention and increased customer satis- faction. Note that the relatively low production volume relates to the particular size of the high-performance fuse.

Size Because of the tight space constraints in the plant, the maxi- mum oor space available for the equipment is 8 5. The maxi- mum permissible height is 8. These regulations often make some solutions less practical. For instance, this requirement may rule out the possibility of using an X-ray machine on the shop oor to determine the PF. The UL regulations set limits on sev- eral factors such as I 2 t proportional to the energy let through the system.

Need Identication and Analysis Case Study 49 0 0. Such expansion of the scope increases the solution space and the probability of nd- ing an innovative solution, as discussed in this section. The first need statement, determine the PF of the fuse, may result in solutions that nondestructively evaluate the PF. One such solution uses acoustic emissions. When a fuse is shaken, the Innovative Conceptual Design 50 Table 3.

Category Design Requirement Performance 1. The device must ensure a minimal packing factor of 0. Target packing factor: 0. Sand: grade 6. Production time: 1. Size 9. Available oor space: 8 5 Maximum height: 8 Safety Must meet OSHA regulations for safety. Special Any modication to the fuse design should meet UL regulations. In the process, they produce sound.

Lec 01 Introduction to Product Design and Development

The intensity of acoustic emission is inversely proportional to the PF. In other words, a loosely filled fuse emits more noise. This energy measurement can be corre- lated with the PF. The modied need statement, ensure the correct packing factor during the manufacturing process, expands the solution space to include various measurements during manufacturing, such as the length of the cavity during assembly and the mass of the sand dur- ing the lling process. Based on the insight that there is very little variation in the cavity diameter, these measurements can be used to determine the PF.

Even more innovative designs are facilitated by shifting the emphasis from quality control to process modification. Identifying the sources of variation in packing factor may lead to a simple, yet innovative, solution involving a fixture to control the length of the cavity.

Since the volume now becomes almost constant, the focus of the design task is shifted to filling a predetermined quantity of sand. Since we learned that changing the uniform grain size has very little effect on the PF, we can focus on introducing a second grain size to fill the interstitial voids.

This can increase the PF without significantly affecting the manufacturing process. In summary, a good need identification coupled with a good need analysis results in crucial insights that can translate into innova- tive design solutions. Quantify the sensitivity of the packing factor to various design variables. Determine the accuracy of computing the PF by weighing the fuse. Identify three different ideas that may be used in a design to ensure the PF.

Need Identication and Analysis Case Study 51 3. Electric Fuses. London: Institution of Electrical Engineers, Quantifying the loss of quality due to off-target performance is dis- cussed in: Fowlkes, W. New York: Addison Wesley, Innovative Conceptual Design 52 This chapter presents the systematic methodology for conceptual design, namely, parameter analysis. First, we look at an invention and try to reconstruct the designers thoughts that led to it. We explore the nature of the conceptual design process and explain it as continuous movement between concept space and conguration space.

This theoretical model is used to demonstrate the invention described earlier and to outline the parameter analysis methodology for conceptual design. When used for sensing earth tide, the distortion of the Earths crust due to the gravitational pull of the moon, the tiltmeter needs to measure changes in angle of the order of 10 6 radians. Another use is in predicting earthquakes, where the required resolution is of the order of 10 7 to 10 9 radians.

Figure 4. The lines repre- sent stiff members, the circles are the pendulum weights, and the solid dots are hinges. A small tilt, , given to the base disturbs the equilibrium of Fig. However, the size of the device presents a prob- lem. The lateral displacement of the weight at the bottom of a sim- ple pendulum is large enough to be measured at small angles of tilt only when the pendulum is extremely long.

For example, a m long pendulum would produce a displacement of 5 m when the angle is 10 7 radians. This is a relatively obvious concept.

Integrating CAD, TRIZ, and Customer Needs

However, the inventor also realized that one could represent this displacement relationship as a simple spring. For the pendulum of Fig. Continuing this logic, the inventor then recognized that there is another way to obtain a small spring constant in addition to extending the length of a simple pendulum. The difference between two large spring constants short pendulums can yield a small spring constant effectively long pendulum. A negative spring would pull in the direction of the initial displacement with a force proportional to the displacement.

Many statically unstable devices exhibit a similar behavior; for example, some light switches jump to a different state when pushed slightly. The inventor of the tiltmeter noted that an inverted pendulum is a negative spring. Thus, all that remained was the coupling between the two pendulums at a point at which the resultant spring constant k 1 k 2 is sufciently small but positive to yield very high sensitivity large lateral displacement for small angles of tilt. This very brief description represents only a small portion of the process that took place in creating the new tiltmeter.

However, it Introduction to Parameter Analysis 55 l f mg x. We shall return to the tiltmeter later, but let us rst exam- ine a few fundamental issues relating to conceptual design and attempt to establish some principles with regard to the process. I Where do new ideas come from? I Why didnt I think of that? These questions seem almost trivial or trite, but they are fundamen- tal to conceptual design, and an attempt to answer them helps to elucidate the subject.

Let us rst consider why conceptual design or the generation of new ideas is such a difcult task. Implicit in this question, of course, is the task of generating very good ideas as opposed to just generat- ing ideas. It is quite a simple matter to generate many, many ideas when the demands on you do not include the quality or goodness of the ideas. As we all know, many more ideas are generated than are developed; many more are developed to some level of completion than are made into products; and many more products are made than are successful.

The real issue comes down to generating ideas of high quality and of good value to others. The result of conceptual design is, hopefully, a new concept or configuration. It is this newness that makes the process so elusive. A new configuration that heretofore did not exist has been created, and the process of reaching that point is very different from the activity that takes place during the refine- ment of the concept downstream in the process. The horizontal axis in Fig.

The vertical axis is a one-dimensional representation of the quality or goodness of a concept. Such a measure may be imagined to be the ratio of performance to cost since, fundamentally, we would all like to pay less for more performance. In any event, such a measure is a convenient way to describe the desirability of con- cepts or configurations. Consider now the curve on the left and imagine that this curve is actually a multidimensional surface in space and that the points on this surface represent different combinations of the design vari- ables that can be reached.

All points on the surface are various real- izations of a single concept.


At this point, two simple examples will help. Imagine that the curve on the left describes a fountain pen. That is, the core technical concept is described by the fact that the pen contains a nib through which ink ows by capillary action from some sort of a reservoir to the paper. Points on this surface may describe pens that range from a quill pen to a sophisticated rellable fountain pen to a fountain pen that has a very wide tip used for calligraphy.

Movements from one point to others on the same surface may come relatively easily by making minor changes to a design variable. These changes might include modifying the type or size of the reservoir of ink, changing the physical dimensions of the nib, or making small changes to the properties of the ink. All of the realizations are still on the surface because they all share the same core technical concept.

Each of the points represents one realization of the common attribute of being a fountain pen. These move- ments may be described as design optimizationa selection of cer- tain existing design variables in order to meet a performance specication. In contrast, conceptual design is represented by a leap from one surface in the design space, the left curve in Fig. The new curve may not even be described by all of the same design variables. Before the leap was made, the new core technical concept did not exist.

To extend the example, imagine that the new curve represents a ballpoint pen. In fact, some of the design variables that are pertinent to the foun- tain pen are also important to ballpoint pens. However, these two congurations do not share a number of design variables, and there is a fundamental difference between the core technical concepts of a fountain pen and a ballpoint pen.

The fountain pen uses capillary action to feed the ink while the ballpoint pen does it through the mechanical rolling motion of the ball. All points on the curve to the left are fountain pens, all points on the curve to the right are ball- point pens, and the generation of the ballpoint pen concept was not obvious in the consideration of the fountain pen designs. It is the fact that this process is a leap to a nonexistent curve that makes con- ceptual design so difcult.

As a second example, suppose the left curve in Fig.

The reciprocating engine curve would consist of points representing various congurations V-shaped, in-line, horizontally opposed, radial , number of cylin- ders, methods of valve actuation side valves, overhead cams , and so on. However, all these congurations would be based on the con- Innovative Conceptual Design 58 cept of reciprocating pistons in cylinders, connecting rods, and crankshafts Fig. The rotary engine curve may represent dif- ferent geometries for example, a triangular rotor in a figure-eight housing as in the Wankel engine, Fig.

Innovative Conceptual Design: Theory and Application of Parameter Analysis

Again, movement along the same curve stands for changing some of the design variables, such as the number of spark plugs used or the shape of the combustion chamber. In contrast, the creation in the s of the concept of rotary engines is denoted by the estab- lishment of the new surface on the right in Fig. This discussion of the generation of new ideas has been only philosophical, but it should help to clarify why the creation of truly new, high-quality congurations or ideas is so difcult and therefore an event too rare in engineering practice today.

It also shows that good ideas do not usually emerge from doing conceptual design in the form of a search over existing solutions. Rather, conceptual design is a discovery process. Let us now try to answer the other two questions posed at the beginning of this section. Study of the process of conceptual design and a look at many inventions allow us to make the follow- ing observations: I Human creativity is more successful or productive when one is attempting to solve simple problems rather than complex ones.

I The best ideas are usually quite simple conceptually. These notions need some explanation. The rst notion deals prima- rily with the number of issues or aspects that must be considered during the solution process.

Upcoming Events

Thus, low-order problems, those that require consideration of very few parameters or factors, tend to be much more easily solved than higher order problems, those with a large number of variables to be considered. That is why all of us try to break down problems into smaller pieces in an effort to make them more tractable. This simple notion comes as no surprise but has profound implications for the nature of successful creative processes.

Introduction to Parameter Analysis 59 The question, Where do good ideas come from? It is not that the result- ing products based on these good ideas core technical concepts are simple, rather, the concepts themselves are most often based on sim- ple but creatively new insights. Often, these simple ideas can be described as resulting from transformations of the key issues in a problem into new relationships or sets of relationships uniquely identied by the creative designer.

As a result, the answer to the frequently asked question, Why didnt I think of that? Innovative Conceptual Design 60 a b Figure 4. The output of the process is generally a physical object or group of objects that here are called congurations. Thus, the various outputs as the process unfolds can be thought of as elements in conguration space. However, consider- ation of the preceding observations concerning problems and solu- tions within conceptual design indicates that movement from one point in conguration space to another point in conguration space is generally made at the level of concepts.

Therefore, the model also contains another space, called concept space, which contains the ideas or concepts that provide the basis for the elements of conguration space. The conceptual design process can be viewed as an iterative process of moving from conguration space to concept space, mak- ing changes or movements within concept space, and then moving back to new points within conguration space corresponding to fur- ther creative generation of congurations based on these new ideas. The process of moving from conguration space to concept space can be thought of as abstraction or generalizationthat is, gen- eralizing from the particulars of a physical conguration to a con- ceptual understanding in order to change it or improve upon it.

Movement from concept space to conguration space can be regarded as realization or particularizationbringing to reality, in particular physical form, the technical concepts arrived at within concept space. Readers should note that in this process description, the elements of conguration space are not actually physical objects but only diagrams, sketches, or other representations of physical objects motivated by the conceptual elements of concept space, which bring some real form to the thoughts created in concept space. This model should be viewed as much more than a philosophical statement.

Readers are encouraged to examine their own experiences in conceptual design to be further convinced of its signicance. As ideas come to life, there are many points along the way which are either intermediate congurations elements in conguration space or con- ceptual statements or relationships elements in concept space. Rarely Introduction to Parameter Analysis 61 within this process is a new conguration created from the previous one without an excursion to concept space, with new conceptual insight being the driving force for the new congurational result.

Let us return now to the tiltmeter example. Of particular importance is the nature of the events that took place within con- cept space. First, there was simplification of the problem by con- sidering only the pendulum conguration and ignoring, temporarily, many other issues that obviously are important. Second, there was a transformation from the normal way of looking at a pendulum to viewing a pendulum as a spring.

Third, there was the creative step of recognizing the relevance of the difference between two large numbers to the situation at hand. Finally, we should not ignore the additional creative steps to generate the double pendu- lum configuration.


On the right-hand side, within congura- tion space, are represented two realizations of pendulum devices: a simple pendulum and the new coupled-pendulum tiltmeter. On the left-hand side, within concept space, are two concepts: the simple spring relationship and the spring relationship represented by the difference between two stiff springs. The transition from the simple pendulum to the complex pendulum within conguration space is not likely to take place by itself. On the other hand, the transition from the concept of a simple spring, correlating with the properties of a pendulum, to the second concept is motivated by recognizing Innovative Conceptual Design 62 Configuration space Concept space Realization particularization Abstraction generalization Figure 4.

Thus, the creative motion within conguration space is actually a movement driven by motion within concept space. In simple terms, the process unfolds with the movements as labeled numerically in the gure. Notice also that the movement labeled 1 was in itself a creative step. It represents a simplication and transformation of the typical understanding of pendulums to a form that, downstream, led to a very interesting result. The reader should be aware that this diagram of movements is extremely simplied. In this example of creative conceptual design, many more motions were taking place back and forth between con- cept and conguration which have not been included in the dia- gram.

Furthermore, the process at this stage of the description is far from over. Many more conceptual issues remain, such as how to obtain hinges with the properties necessary to allow the new device to perform as required and how to measure the lateral displacement of the pendulum weight. This paper presents parameter analysis as a tool to create configurations incorporating design principles in general, and the principle of direct and short load transmission path in particular. View PDF. Save to Library. Create Alert. Share This Paper. Figures from this paper.

References Publications referenced by this paper. Engineering design: a systematic approach: G.