Advertisement
Guest User

Untitled

a guest
Aug 18th, 2019
168
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 4.83 KB | None | 0 0
  1. ## 3. Models
  2. ### 3.1 Motivation
  3. * Purpose is to reduce complexity
  4. * built by abstraction of existing objects and defining relevant relation between elements
  5. * equivalence classes can be formed
  6.  
  7. **Abstracting** a problem can lead to find a solution more easily.
  8.  
  9. The **product concretization model** uses three layers of abstraction
  10.  
  11. ### 3.2 Function Models
  12.  
  13. * A **function** states the action of a system upon an element through a **input-output relationship**, like $$y=f(x)$$
  14.  
  15. * A **function** can also be a **desired input-output relation**, like the steering wheel which must transmit steering commands
  16.  
  17. * can be modeled in a **function model**
  18.  
  19. Can be visualized by:
  20.  
  21. * *Words*: for interdisciplinary team exchange
  22. * *Graphics*: Within a team in common language
  23. * *Graphical description*: Within experts
  24.  
  25. * Can be **flow-oriented** or **relation-oriented**
  26.  
  27. ### 3.2.1 Flow-oriented Function Modeling
  28.  
  29. > Fact sheet - Flow-oriented Function Modeling
  30. > 1. Input:
  31. > * Design goal
  32. > * Requirements
  33. > 2. Output:
  34. > * complete model of important functions
  35. > * design scheme
  36. > 3. Procedure:
  37. > 1. Define the starting and end state of your processed object
  38. > 2. Find functions which are necessary to get from the starting point to the desired end state:
  39. > * After defining a function write down the new state of your object and define the next function using this state as an input
  40. > * Try to formulate your functions as solution neutral as possible
  41. > 3. When you have reached your final state make sure that your process is complete.
  42. > 4. Are there harmful functions which need to be addressed?
  43. > 5. Think about energy and information flows and how are those functions connected.
  44. > 6. Make a legend and explain all symbols you have used.
  45. > Situation:
  46. > * Early stages of PD process
  47. > lack of understanding the problem
  48.  
  49. ### 3.2.2 Relation-oriented Function Modeling
  50. * Functions are differenced by useful and harmful functions
  51. * This is visualized by a rectangle with a black edge (for harmful functions) or no black edge.
  52. * relations between the functions can either be:
  53. * is required for: arrow with no dotted line
  54. * causes: arrow with dotted line
  55. * was introduced to avoid: arrow with circle head, not dotted
  56. * can be developed by systematically questioning the considered technical system
  57. * existing solutions can be used as a starting point
  58.  
  59. Goal:
  60. * Understanding the technical problem
  61. * Derivation of problem formulations from the function model
  62.  
  63. > Fact Sheet Relation Oriented Function Modeling
  64. > Input:
  65. > * Design goal
  66. > * Requirements
  67. > Output:
  68. > * Complete model of important functions
  69. > * Solution-neutral
  70. > * Basis for finding solutions
  71. > Situation:
  72. > * Early Stages of product development process
  73. > * Improvement of existing products
  74. > * Lack of understanding the problem
  75. > Procedure:
  76. > * Question system and derive functions
  77. > * Ask if functions itself cause or require useful or harmful functions
  78. > * Use suitable functions to avoid harmful functions
  79.  
  80. ### 3.3 Graphs and Models
  81. #### 3.3.1 Graphs
  82.  
  83. * A graph is a set of **elements** (also vertices) with a set of **relations** (also edges, arcs) between pairs of these elements.
  84. * Possible element types are e.g. Functions, requirements, CAD sketches, ...
  85. * Possible relation types can be undirected: e.g. Spatial or related to communication or directed, like logical relations, e.g. "requires", "contains"
  86. * graphs can be directed or undirected
  87. * direction often expressed by position
  88. * A **tree** is a graph in which any two vertices are connected by exactly one path
  89. * A **polyhierarchy** can be represented by a directed graph without cycles along directed edges
  90. * A **matrix** is a rectangular array of expressions arranged in roes and columns
  91. * Every graph can be represented by a matrix: input in rows output in columns
  92.  
  93. #### 3.3.2 Types of matrices
  94.  
  95. 1. Intra-Domain: relations between elements belonging to the same type
  96. 2. Inter-domain: -"- of different types
  97. 3. Combinations: inter and intra domain
  98.  
  99. ### 3.4 Design Structure Matrices
  100.  
  101. > Fact Sheet Design Structure Matrices
  102. > Input:
  103. > * Elements of the same type, e.g. organization/ product architecture
  104. > * Relation between those
  105. > Output:
  106. > Relation matrix
  107. > Situation:
  108. > * Analysis of existing product
  109. > * Need to identify important elements
  110. > Procedure:
  111. > * Decompose: Identify relevant elements
  112. > * Document relations
  113. > * Analyze: Identify structural patterns
  114. > * Improve system if needed
  115.  
  116. #### 3.4.1 DSM Classification
  117.  
  118. * To detect subsets with internal dependencies
  119. * Basis for modularization strategies
  120. Targets:
  121. * To minimize external interactions
  122. * To Maximize internal interactions
  123.  
  124. * High connectivity systems need algorithmic support
  125. * partially applicable matrices if clusters are overlapping
  126.  
  127. → Ball pen: Button and Push element form back push element
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement