Artificial Intelligence in Engineering Design is a three volume edited collection of key papers from the field of artificial intelligence and design, aimed at providing a state-of-the-art description of the field, and focusing on how ideas and methods from artifical intelligence can help engineers in the design of physical artifacts and processes. The book surveys a wide variety of applications in the areas of civil, mechanical, chemical, VLSI, electrical, and computer engineering. The contributors are from leading academic computer-aided design centers as well as from industry.
Artificial Intelligence techniques for design problemsThis book is all about how ideas and methods from Artificial Intelligence (AI) can help engineering designers. The edited collection of papers herein by leaders in the field of "AI and Engineering Design" is a state-of-the-art look at that discipline.
MULTI-LEVEL METHODOLOGY. These three volumes are organized around the viewpoint that AI provides a methodology for engineering knowledge-based problem-solving systems, that is organized around three levels: knowledge level, algorithm level, and program level.
A knowledge-level specification of the system (and the class of design problems it must solve) is mapped into an algorithm level description of an efficient search algorithm for efficiently and acceptably "solving" that class of problems. These algorithms draw upon extensive bodies of design knowledge effectively encoded in algorithm components that perform refinement, hillclimbing, constraint propagation, backtracking, analogy, knowledge acquisition, estimation, budgeting, simulation, spatial, temporal, or "common sense" reasoning, etc.
That simulatable design algorithm description is then mapped into an optimized piece of code at the program level, using one or more programming paradigms (e.g., procedural programming, rule-based programming, object-oriented programming), design "shells" (e.g., VP-EXPERT), or commercially available subsystems (e.g., an ATMS in KEE, a parametric modelling system, etc.).
DESIGN TASK CATEGORIZATION. The second key principle for organizing this 3-volume series distinguishes design tasks by amounts and types of available knowledge (and the form in which the knowledge is available), identifying three categories: routine design, innovative design, and creative design.
The more design knowledge available in the right form, the more routine (or "direct") a design process model can be used (involving a top-down refinement and/or hillclimbing process that converges on an acceptable design with little or no search).
Any missing knowledge or knowledge in the wrong form or incorrect knowledge must be compensated for. Such innovative design tasks can be addressed by various "indirect" techniques such as case-based reasoning, structural mutation, combining multiple knowledge sources, and explicit planning of the design process.
WHAT'S IN EACH VOLUME. Volume I focuses on "Design Representation" and Models of Routine Design". Volume II focuses on "Models of Innovative Design", "Reasoning about Physical Systems", and "reasoning about Geometry". Finally, this volume (Volume III) focuses on "Knowledge Acquisition", "Commercial Systems", and "Integrated Environments".
