Two Approaches to the Induction of Graph-Rewriting Rules for Function-Based Design Synthesis

Author(s):  
Julian R. Eichhoff ◽  
Felix Baumann ◽  
Dieter Roller

In this paper we demonstrate and compare two complementary approaches to the automatic generation of production rules from a set of given graphs representing sample designs. The first approach generates a complete rule set from scratch by means of frequent subgraph discovery. Whereas the second approach is intended to learn additional rules that fit an existing, yet incomplete, rule set using genetic programming. Both approaches have been developed and tested in the context of an application for automated conceptual engineering design, more specifically functional decomposition. They can be considered feasible, complementary approaches to the automatic inference of graph rewriting rules for conceptual design applications.

Author(s):  
Ang Liu ◽  
Wei Wei ◽  
Stephen C.-Y. Lu

Synthesis is a common activity in engineering design. It is widely recognized to be important to the whole engineering design process in general, and to the early design stages in particular. In the past, however, there remains lack of a set of rigorous metrics to evaluate the effectiveness of performing design synthesis in conceptual design. Based on relevant studies of abductive reasoning in logic, this paper introduces a set of domain-independent design synthesis metric: clarity, feasibility, testability, simplicity, and analogy. For each metrics, the rationale of including it is explained, and a systemic evaluation procedure is prescribed. Individually, each metrics addresses a particular aspect of design synthesis in conceptual design. Collectively, the combined consideration of all metrics as a single vector helps the designer to identify the most promising synthesis outcome, the best design concept, which both satisfies upstream objectives and meets downstream constraints.


Author(s):  
Julian R. Eichhoff ◽  
Dieter Roller

This paper compares methods for identifying determinism within graph-rewriting systems. From the viewpoint of functional decomposition, these methods can be implemented to search efficiently for distinct function structures. An additional requirement is imposed on this comparison that stems from a cooperative design application where different organizations contribute to a distributed graph-rewriting system: Inspecting the definitions of production rules is not allowed for identifying determinism because production rules are considered to be confidential corporate knowledge. Under this assumption, two approaches were selected and empirically compared with respect to random search and guided search scenarios. The results suggest that the herein proposed dynamic rule independence analysis outperforms traditional approaches in light of the above restriction.


Author(s):  
Joshua T. Gyory ◽  
Jonathan Cagan ◽  
Kenneth Kotovsky

A commonly held presumption is that the production of a team is superior to that of individual performance. However, in certain scenarios, such as during brainstorming activities and in configuration engineering design problems, it has been shown that individuals working alone are more effective than teams working together. This research considers whether the same outcomes hold for a more open-ended scenario, in conceptual engineering design. Thus, a behavioral study is run with freshman engineering students solving a conceptual design problem working in teams or individually. Results corroborate previous findings, showing that individuals outperform teams in the quality of their design solutions. One of the primary differences between individuals and group problem solving is the fact that groups need to verbalize to communicate ideas. Consequently, this study also analyzes how verbalization, which may be one disadvantage of team problem solving, affects the performance of individuals in this context of conceptual engineering design. Individuals who verbalize throughout problem solving, however, perform similarly to those who did not. Overall, the results from this study suggest that, individuals are still better performers and teams may not always be the optimal circumstance. Moreover, verbalization does not seem to act as a cognitive barrier to problem solving, and further investigation needs to be done to diagnose the potential impediments which put teams at a disadvantage to individuals during conceptual design.


Author(s):  
Patricia Kristine Sheridan ◽  
Jason A Foster ◽  
Geoffrey S Frost

All Engineering Science students at the University of Toronto take the cornerstone Praxis Sequence of engineering design courses. In the first course in the sequence, Praxis I, students practice three types of engineering design across three distinct design projects. Previously the final design project had the students first frame and then develop conceptual design solutions for a self-identified challenge. While this project succeeded in providing an appropriate foundational design experience, it failed to fully prepare students for the more complex design experience in Praxis II. The project also failed to ingrain the need for clear and concise engineering communication, and the students’ lack of understanding of detail design inhibited their ability to make practical and realistic design decisions. A revised Product Design project in Praxis I was designed with the primary aims of: (a) pushing students beyond the conceptual design phase of the design process, and (b) simulating a real-world work environment by: (i) increasing the interdependence between student teams and (ii) increasing the students’ perceived value of engineering communication.


2021 ◽  
Vol 9 (4B) ◽  
Author(s):  
Bin Chen ◽  
◽  
Jie Hu ◽  
Weixing Chen ◽  
◽  
...  

The trend of inter-disciplinary conceptual design synthesis requires designers to involve more and more distributed multi-disciplinary design resources. Therefore, this paper proposes a graph-based computerized optimal conceptual design synthesis to help designers explore novel design schemes within the distributed multi-disciplinary resource environment. The design resources tightly related to the design goal can be extracted from the huge resource environment by a proposed searching engine. The optimal design scheme can be generated from these related design resources by a proposed graph-based algorithm. A set of computer applications called Automatic Conceptual Design System (ACDS) is established to verify the feasibility of this proposed conceptual design synthesis, and a garbage power system’s conceptual design is completed by this software prototype.


2017 ◽  
Vol 823 ◽  
pp. 012041
Author(s):  
Prabal Biswas ◽  
Prosenjit Santra ◽  
Kirit Vasava ◽  
Snehal Jayswal ◽  
Tejas Parekh ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document