We present a system, FIRST, that redesigns structural beams by accessing a case memory of solution plans. FIRST starts by analysing an existing design, using general knowledge about elementary physics. If design constraints are unsatisfied, FIRST searches for similar problem situations in its case memory and retrieves the solution plans associated to those situations. The system performs a transfer by analogy of each plan into the new problem situation and combines the transferred plans and symbolic analysis knowledge into a global redesign plan that is applied to the problem. FIRST is implemented in BB1, a blackboard system that allows the cooperation of problem solving knowledge from different sources. The system, that includes general methods for transferring a plan by analogy and mapping parts of it into a new problem situation is described through the analysis and redesign of a round cantilever beam. 1 Introduction Mechanical design is the process of going from a set of specifications to a physical artifact meeting those specifications. It is a very underconstrained process whose complexity has been pointed out in previous research [Howe et a/., 1987], [Mittal et a/., 1986]. Mechanical design rarely starts from first principles. Rather, an existing artifact is often modified until it meets the design specifications. Problems in which the structure of the redesigned artifact remains fixed throughout the modification process belong to the category of routine design problems. The knowledge to modify a particular design is traditionally encoded as rules or even plans that are instantiated during the solution process. More recent developments such as PROMPT [Murthy and Addanki, 1987] or l st PRINCE [Cagan and Agogino, 1988] have focused on the ability to derive routine and non-routine modifications by performing a symbolic analysis of the behavioral equations describing the design. Another possibility is to derive the modifications from similar cases. In a designer's memory, these cases represent the knowledge acquired on similar projects and make the difference between the expert and the novice designer. This redesign knowledge is naturally expressed by plans. Plans are attractive because they potentially embody all the knowledge required to successfully modify an artifact. Our work investigates the potential of a case-based approach to mechanical design. To serve that purpose, we have built a system that uses a case memory of past problems and solution plans to redesign structural beams. Beam analysis is interesting for various reasons. First, its theory is well established and …
[1]
David C. Brown,et al.
Knowledge and Control for a Mechanical Design Expert System
,
1986,
Computer.
[2]
Jaime G. Carbonell,et al.
Derivational Analogy and Its Role in Problem Solving
,
1983,
AAAI.
[3]
Russ B. Altman,et al.
PROTEAN: Deriving Protein Structure from Constraints
,
1986,
AAAI.
[4]
Alice M. Agogino,et al.
Innovative design of mechanical structures from first principles
,
1987,
Artif. Intell. Eng. Des. Anal. Manuf..
[5]
Paul R. Cohen,et al.
Dominic: A Domain-Independent Program for Mechanical Engineering Design
,
1986,
Artif. Intell. Eng..
[6]
Sanjaya Addanki,et al.
PROMPT: An Innovative Design Tool
,
1987,
AAAI.
[7]
Clive L. Dym,et al.
PRIDE: An Expert System for the Design of Paper Handling Systems
,
1986,
Computer.
[8]
Sanjay Mittal,et al.
A Knowledge-Based Framework for Design
,
1986,
AAAI.