Lernmaterialien für Methoden der Produktentwicklung an der Fachhochschule Lübeck

A model is a simplified description of reality. Thus, is a form of abstraction of reality. It helps to understand
complexity and to deal with the limitations of the human mind.

a function model models the functions of a technical system and their relationships

Purpose

• analysis and understanding of the desired technical system
• basis for the fuctional concept an the search for solutions
• black-box approach, means a solution neutral formulation of requirements

Elements

• design variable, attributes
• values, artefacts, CAD sketches
• functions
• requirements
• parts of the product, components
• states
• tasks / process steps
• individuals, groups, stakeholders

Relations

• unidirectional
  • spatial
  • related to communication
• directed
  • logical
    • necessary/sufficient condition
    • "requires"
    • INUS
    • "contains" (for decompostition)
  • flow of material / energy
  • flow of information

organize many elements with pair-wise relationships / dependencies into groups with a small number of relationships / dependencies between seperate groups

used for modularization of complex products eg product family design

• reordering of rows and columns of a DSM with the objective of arranging all existing dependecies at one side or at least asclose as possible to the diagonal
• produces "smaller" feedback lops
• triangular matrix - process without any iterations steps, no feedback loops

• may change during design or use of the product and are
• necessary to formulate INUS conditions > building plan for requirement derivation.

ADG - multi-level model

• resolves dependencies in detail
• higher modelling effort

DMM - 2-level model, maps two domains onto each other

QFD - 2 1/2 model

• does not fully resolve dependencies between design variables
• easier to create, visualize and communicate

• Search Space is not the Solution Space
• interpretation of the problem may depend on the designer
• more than one solution of different value

• simplified models
  • Bsp. restrain system in car
  • simplify by focusing on most relevant effects
  • reverse abstraction
  • possibly along ADG
• solution spaces
  • Bsp. front crahs vehicle design
  • use a physical surrogate model (Ersatzmodell)
  • compute solution spaces here for force-deformation characteristics
• Experience-based Iteration
  • Bsp. Front crahs vehicle design, component design
  • trial and error solution generation to sub-problems
  • supported by engineering judgement
  • if requirements not independet, risk of iteration on system level
• Numerical Optimization
  • Bsp. Front Crash vehicle design, component design
  • Input: iojective function and constraint functions (optimization problem min f subject to g)
  • Output: optimal design
• Product Family Design
  • Bsp. Vehicle Chassis Design
  • common components can be identified by properies in overlapping solution spaces
• Top-down vs Bottom-up Development
  • Top-down: design by reverse abstaction
  • buttom-up: evaluation
  • combination possible (good design with proven ability to build it)

Quantitative models are compact representations where a (single) (differential or difference) equation may describe the performance of the system for a large set of input functions and initial states

Collection of products and components that may be shared

Driven by

• expreience
• intuition
• feeling
  > subjective process

use methodes to

• rationalize and organize ingredients of decision process
• making reasons or preferences transparent to others

choosing methods depends on

• depth of analysis needed
• available informatio and knowledge
• effort willing to spend