Solutions

Question 1

concept - definition

Answer 1

A concept is an abstract definition of a solution candidate.

Question 2

What are the characteristics of a qualitative solution approach?

Answer 2

▪ Output of qualitative approaches are solution elements defining (a part of) a concept.
▪ The choice of solutions elements determines the availability of specific design variables.
▪ In this sense, qualitative approaches are applicable to an unstructured design space.

Question 3

What are the characteristics of a quantitative solution approach?

Answer 3

▪ The output of quantitative approaches are desired values or value ranges for the design variables.
▪ Knowledge of the design variables is required, i.e., knowledge of their definition – not their values.
▪ These approaches operate on a structured design space.

Question 4

What are examples for a qualitative solution approach? Name 2.

Answer 4

-Creativity methods
-TRIZ
-Morphological chart
-Prototyping

Question 5

What are examples for a quantitative solution approach? Name 2.

Answer 5

-Target cascading & optimization
-Solution spaces
-Parametric study

Question 6

Morphological Chart - Definition

Answer 6

A Morphological Chart is an ordering scheme for functions and partial solutions.

Question 7

What can a partial solutiion in a morphological chart be?

Answer 7

Partial solutions may be:
− Functions
− Physical principles
− Components

Question 8

What is the input for a morphological chart?

Answer 8

▪List of functions and/or requirements
▪Optional: initial ideas for partial solutions

Question 9

What is the output for a morphological chart?

Answer 9

▪Chart of partial solutions ordered by associated functions and relevance
▪One or more concepts

Question 10

What is the right situation to use a morphological chart?

Answer 10

Need to define a concept and systematically explore many options.

Question 11

What is the procedure for a morphological chart?

Answer 11

1. List functions of the system in the left column. Order the functions by importance from top to bottom.
2. Assign partial solutions to functions
3. Order the partial solutions by importance from left to right. If there are many solutions, they can be grouped to solution classes, e.g., by the required form of energy, the required manufacturing technology etc.
4. Combine single partial solutions to concepts.The theoretically possible number of combinations can be reduced to a reasonable level by (1) postponing less relevant functions and (2) less suitable partial solution ideas and by (3) combining only partial solutions of the same solution class.

Question 12

What is the goal of target cascading?

Answer 12

Break down top-level requirements into detailed design requirements. Requirements on sub-systems are then again broken down into sub-sub-system or component requirements.

Question 13

What are the characteristics of numerical optimization?

Answer 13

▪ Numerical optimization is typically performed by algorithms.
▪ Input: definition of objective function 𝑓(𝒙) and constraint functions 𝑔𝒙.
▪ Typical objective function for target cascading: 𝑓𝒙 =|𝑦1𝒙−𝑦1,𝑡𝑎𝑟𝑔𝑒𝑡 | (deviation from target value)
▪ Output: optimal design 𝒙∗ (values for x)

Question 14

Analytical Target Cascading - definition

Answer 14

In Analytical Target Cascading, target values (=point-based) for requirements are computed by numerical optimization.
▪ If attribute dependencies resemble trees → easy
▪ If not → coordination and negotiation necessary

Question 15

Solution space - definition

Answer 15

A solution space is a set of good designs, i.e., designs that satisfy all requirements. Formally, for a design problem with:
(1) design variables 𝒙 = (𝑥𝑖) ,
(2) performance 𝒚 = (𝑦𝑗) and 𝑦𝑗 = 𝑓𝑗(𝒙) and
(3) requirements𝑦𝑗𝑙≤ 𝑦𝑗≤ 𝑦𝑗𝑢,
a solution space Ω satisfies: 𝑦𝑗𝑙 ≤ 𝑓𝑗 𝒙 ≤ 𝑦𝑗𝑢, ∀𝒙 ∈ Ω.

Question 16

complete solution space - definition

Answer 16

The complete solution space is the set of all good designs.

Question 17

box-shaped solution space - definition

Answer 17

A box-shaped solution space is a solution space expressed as product of permissible intervals Ω = [𝑥1𝑙,𝑥1𝑢] × ⋯ × [𝑥𝑑𝑙,𝑥𝑑𝑢]

Question 18

design space - definition

Answer 18

The design space is the set of all the designs considered in the design problem.

Question 19

What is the price of strong decoupling?

Answer 19

loss of solution space

Question 20

Does the definition of solution space provide information on wether a design can be realized / is physically feasible?

Answer 20

No, it does not provide information on whether a design can be realized / is physically feasible. It ensures only that design goals are reached → top-down view.

Question 21

Why can the boundaries for good designs change in Selective Design Space Projection when variating between a variable?

Answer 21

Because several projections are projected on top of each other, making up a mix of good and bad designs.

Question 22

What are benefits of Selective Design Space Projection?

Answer 22

▪ Interactive tool to construct a box-shaped solution space
▪ Trains intuition for high-dimensional problems, in particular space shapes and sensitivities
▪ Designer can include non-formalized knowledge about constraints

Question 23

What are limitations of Selective Design Space Projection?

Answer 23

▪ Numerically expensive: ~1000 function evaluations per 2d diagram
▪ Difficult to find good designs in high dimensions
▪ Only works with quantitative models

Question 24

What is the input for Selective Design Space Projection?

Answer 24

▪ Quantitative model 𝑦𝑗 = 𝑓(𝑥𝑖)
▪ Interval-based or one-sided requirements on 𝑦𝑗

Question 25

What is the output for Selective Design Space Projection?

Answer 25

▪ Interval-based or one-sided requirementson 𝑥𝑖

Question 26

What is the right situation to use Selective Design Space Projection?

Answer 26

▪ In early stages of development toexplore design space ▪ When design goals orsystem requirements are to be broken down into component requirements to distribute or integrate development work ▪ Need for precise requirements at stakeholder interfaces

Question 27

What is the procedure for Selective Design Space Projection?

Answer 27

Goal is to generate sufficient conditions for 𝑥𝑖 for satisfying the requirements on 𝑦𝑗. Each of the following is a possible method: ▪ Analytical derivation by manipulation of inequalities (when model is given as white-box formula) ▪ Selective design space projection ▪ Numerical optimization by solution space optimization

Question 28

What are the benefits of solution spaces?

Answer 28

▪ Quantitative requirements → objective and transparent ▪ Provides design freedom to integrate requirements from different disciplines. ▪ Supports robust design. ▪ Supports decision making. ▪ Supports a solution-neutral way of thinking.

Question 29

What are the limitations of solution spaces?

Answer 29

▪ Only works when design variables and quantities of interest are defined → may not be available in concept phase. ▪ Quantitative models required → often high modeling effort. ▪ Often expensive. ▪ Remember Role-playing game: “point-based is easier”

Question 30

product family - definition

Answer 30

A product family is a set of product variants, sharing component variants or module variants to achieve variety efficiently.

Question 31

module - definition

Answer 31

A module is a distinct self-contained unit that serves a specific function that can be used in more than one system.

Question 32

What is module-based product family design?

Answer 32

Variants are generated by adding, removing or substituting modules, possibly varying the type and changing the property variables.

Question 33

What is scale-based product family design?

Answer 33

Variants are generated by scaling components, i.e., changing the values of their properties.

Question 34

What are reasons to use product family design? Name 3.

Answer 34

▪ Economies of scale: Buying and producing common parts in larger quantities reduces unit cost, because it enabels more efficient use of manufacturing resources. ▪ Streamlined processes: reduced complexity cost = additional expenses that arise when a company’s products, processes, or organization become more varied, intricate, or difficult to manage. ▪ Streamlined maintenance and service: Common parts simplify repairs and reduce inventory for service centers. ▪ Consistent Quality: Reuse of proven components improves reliability and reduces the risk of defects. ▪ Reduced Development Cost and Time: Shared platforms and components may reduce design work needed for each new product variant – if complexity is managed properly.

Question 35

component type - definition

Answer 35

A component type is the most abstract class to distinguish between components, e.g., transmissions vs. motors.

Question 36

component variant - definition

Answer 36

A component variant is used to distinguish between components of the same type that are shared among product variants (e.g., spur gear variant 2, worm gear variant 13, …)

Question 37

component in hardware - definition

Answer 37

A component in hardware is the physically realized component = the instantiation of a component variant

Question 38

assignment scheme - definition

Answer 38

An assignment scheme assigns component variants to product variants.

Question 39

commonality pattern - definition

Answer 39

A commonality pattern states which product variants share component variants.

Question 40

What is a typical Product Family Design Problem?

Answer 40

Typical design problem: Minimizethe number of component variants fora given set of requirements by (1) designing component variants and (2) assigning them to product variants.

Question 41

Bell number - definition

Answer 41

The Bell number is the number of possible partitions of a set.

Question 42

What is the input for a parametric study?

Answer 42

▪ Initial list of design design variables (possibly values) ▪ Initial list of quantities of interests

Question 43

What is the output for a parametric study?

Answer 43

▪ List of all considered design variables ▪ List of all considered QoIs ▪ Design variants including the best design

Question 44

What is the procedure for a parametric study?

Answer 44

▪ Use unique identifier ▪ When you generate a variant: (1) document the parent as reference (2) copy values (3) change design variable value(s) and (4) highlight changes ▪ Add a new DV or QoI by adding a column

Question 44

What is the right situation to use a parametric study?

Answer 44

▪ Concept is defined, i.e., some design variables (DVs) and quantities of interest (QoI) are defined (however not their values) ▪ Seeking or improving a solution

Question 45

What is the traditional view of prototyping?

Answer 45

▪ Provides information for important design decisions, e.g., before committing to significant investements ▪ Enables evaluation of technical feasibilty ▪ Helps to avoid failure and associated cost and development time ▪ Prototyping is established in the late development phase as tool for risk reduction

Question 46

What is the extended view of prototyping?

Answer 46

▪ Helps to explore the design space by generating and testing many ideas. ▪ Promotes failure at reduced cost → fail fast and fail early! ▪ Lets you communicate with stakeholders. ▪ Helps to assess desirability and concretize or confirm requirements

Question 47

What can be possible concerns about prototyping? Name 3.

Answer 47

▪ Dislike of failure ▪ Striving for a perfect and final design ▪ Worries about judgement by other stakeholders ▪ Lack of knowledge about prototyping tools/mindsets ▪ Lack of prototyping culture

Question 48

Prototype - definition

Answer 48

A prototype is a physical or digital embodiment of critical elements in the design, and an iterative tool to enhance communication, enable learning, and inform decision making at any point in the design process.

Question 49

What different types of prototypes exist?

Answer 49

You can prototype products, processes, and services.

Question 50

What can be the purpose of a prototype?

Answer 50

Exploration Exploration Prototyping focuses on learning and producing new insights to generate new ideas. →Look unimportant, bare functionality, works somehow Evaluation Evaluation Prototyping focuses on specific design ideas, and tests explicit hypotheses and assumptions. →Should convince you that it works Communication Communication Prototyping focuses on involving other stakeholders, like users or decision makers. →Should look nice, higher demands on result

Question 51

What are the three lenses of innovation for prototyping?

Answer 51

Desirability Desirability Prototyping should focus on solving a customer need and creating a positive customer experience. Feasibility Feasibility Prototyping should focus on solving a key issue on functionality or technical feasibility. Viability Viability Prototyping should focus on creating an economically viable solution that people are willing to pay for.

Question 52

What are possible disciplines for prototyping? Name 3.

Answer 52

-Interaction Design -Industrial Design -Electrical Engineering -Product Design -Architecture -Experience Design -Computer Science -Mechanical Engineering

Question 53

What is the input for prototyping?

Answer 53

▪ Ideas, hypotheses ▪ Possibly stakeholder output

Question 54

What is the output for prototyping?

Answer 54

▪Confirmation / rejection ▪Learnings ▪New ideas

Question 55

What is the right situation to use prototyping?

Answer 55

▪Need to explore, evaluate, communicate a solution idea ▪Throughout the development process

Question 56

What is the procedure for prototyping?

Answer 56

▪Define a hypothesis or question (possibly also a purpose and a lens of innovation you want to address) ▪Define a definition of donefor the test ▪Design a test and the context how you want to test your prototype ▪Create and test your prototype related to your hypothesis ▪Reflect your prototyping activity and document your key insights