An artificial intelligence (AI) algorithm to automate symbolic design reformulation is an enduring challenge in design automation. Existing research shows that design tools either require high levels of knowledge engineering or large databases of training cases. To address these limitations, we present a singular value decomposition (SVD) and unsupervised clustering-based method that performs design reformulation by acquiring semantic knowledge from the syntax of design representations. The development of the method was analogically inspired by applications of SVD in statistical natural language processing and digital image processing. We demonstrate our method on an analytically formulated hydraulic cylinder design problem and an aeroengine design problem formulated using a nonanalytic design structure matrix form. Our results show that the method automates various design reformulation tasks on problems of varying sizes from different design domains, stated in analytic and nonanalytic representational forms. The behavior of the method presents observations that cannot be explained by pure symbolic AI approaches, including uncovering patterns of implicit knowledge that are not readily encoded as logical rules, and automating tasks that require the associative transformation of sets of inputs to experiences. As an explanation, we relate the structure and performance of our algorithm with findings in cognitive neuroscience, and present a set of theoretical postulates addressing an alternate perspective on how symbols may interact with each other in experiences to reify semantic knowledge in design representations.

This paper explores the use of heuristics as cognitive strategies invoked during the process of design. Heuristics are reasoning processes that do not guarantee the best solution, but often lead to potential solutions by providing a simple cognitive “shortcut.” We propose that designers use specific design heuristics to explore the problem space of potential designs, leading to the generation of creative solutions. We test whether design heuristics can be taught to novices, and suggest their use will facilitate the design process at multiple levels of instruction. In the present empirical study, we evaluate a set of six instructional heuristics and validate their effectiveness with product concepts generated by novice designers. Six hundred seventy-three drawings were created by 120 first-year college students under four instructional conditions. Drawings were coded according to their content, use of heuristics, creativity, and practicality. The most creative concepts emerged from the experimental conditions where heuristics were introduced. Heuristics appeared to help the participants “jump” to a new problem space, resulting in more varied designs, and a greater frequency of designs judged as more creative. Our findings suggest that simple demonstration of design heuristics may, at times, be sufficient to stimulate divergent thinking, perhaps because these heuristics are readily grasped and contextual application is not required. Based on these findings, a conceptual model for design education emphasizing the importance of using a variety of heuristics is proposed. This model suggests that learning can be enhanced through exposure to a variety of design heuristics, and can supplement formal education and foster personal development in design learning.

The natural world provides numerous cases for inspiration in engineering design. Biological organisms, phenomena, and strategies, which we refer to as biological systems, provide a rich set of analogies. These systems provide insight into sustainable and adaptable design and offer engineers billions of years of valuable experience, which can be used to inspire engineering innovation. This research presents a general method for functionally representing biological systems through systematic design techniques, leading to the conceptualization of biologically inspired engineering designs. Functional representation and abstraction techniques are used to translate biological systems into an engineering context. The goal is to make the biological information accessible to engineering designers who possess varying levels of biological knowledge but have a common understanding of engineering design. Creative or novel engineering designs may then be discovered through connections made between biology and engineering. To assist with making connections between the two domains concept generation techniques that use biological information, engineering knowledge, and automatic concept generation software are employed. Two concept generation approaches are presented that use a biological model to discover corresponding engineering components that mimic the biological system and use a repository of engineering and biological information to discover which biological components inspire functional solutions to fulfill engineering requirements. Discussion includes general guidelines for modeling biological systems at varying levels of fidelity, advantages, limitations, and applications of this research. The modeling methodology and the first approach for concept generation are illustrated by a continuous example of lichen.

The main hypothesis investigated in this paper is based upon the suggestion that the discursive reasoning in architecture supported by an explicit knowledge of spatial configurations can enhance both design productivity and the intelligibility of design solutions. The study consists of an examination of an architect's performance while solving intuitively a well-defined problem followed by an analysis of the spatial structure of their design solutions. One group of architects will attempt to solve the design problem logically, rationalizing their design decisions by implementing their explicit knowledge of spatial configurations. The other group will use an implicit form of such knowledge arising from their architectural education to reason about their design acts. An integrated model of protocol analysis combining linkography and macroscopic coding is used to analyze the design processes. The resulting design outcomes will be evaluated quantitatively in terms of their spatial configurations. The analysis appears to show that an explicit knowledge of the rules of spatial configurations, as possessed by the first group of architects can partially enhance their function-driven judgment producing permeable and well-structured spaces. These findings are particularly significant as they imply that an explicit rather than an implicit knowledge of the fundamental rules that make a layout possible can lead to a considerable improvement in both the design process and product. This suggests that by externalizing the design knowledge and restructuring it in a design model, creative thought can efficiently be evolved and stimulated.

A metric for computing the information content of function models in mechanical engineering design is proposed. Function models are graph-based representations used to describe the functionality of engineered artifacts, where the nodes are function verbs and the edges are the objects of action. The functional basis, a controlled vocabulary of these verbs and nouns organized in a three level hierarchy, is intended to support consistent representation of function models. The Design Repository is a Web-based archive of function models of consumer products described with the functional basis. This paper presents the theoretical underpinnings of a metric for the information content of function models, the assumptions required to support it, the definitions of key terms associated with it, and its practical interpretation. Finally, the metric is used to study the usefulness of the functional basis through a series of experiments on function models within the Design Repository. The results of the experiment indicate that the secondary level of the functional basis is the most beneficial to designers, both in terms of information content and information density.

When developing a better understanding of the design process there are several possible approaches to choose from. Many studies are based on novice designers (e.g., students) or designers of relatively modest talents. By contrast, some studies have queried designers who are considered to have outstanding and exceptional ability in order to gain an understanding of design at the highest level that it is practiced. The study reported here adopts yet another approach by exploring how design processes are perceived by design educators. The approach is motivated by the observation that teaching design requires consciously distilling the essence of the design process for the students, observing students during their design process and guiding them through the process. As a result, design teachers tend to develop a more articulate view of design processes than most other designers. Nineteen design teachers are interviewed using general topics as discussion points. Such an approach is invaluable when exploring more abstract research questions such as the notion of design processes. This approach differs from more controlled approaches (e.g., protocol analysis) in that it accepts that the data obtained are partially driven by negotiation between the researchers and the participants, and that the discussions are largely stories or narratives about design and designing. The resulting data illustrate that design processes are interpreted, articulated, and understood in a variety of ways by different teachers. These data and subsequent results tell us in rich detail about designing and design teaching, and as a result extend our understanding of the design process.

Fixation prevents the associations that are bridges to new designs. The inability to see alternative solutions, or even to see how to map known solutions onto current problems, is a particularly acute problem in the design of software-intensive systems. Here, we explored two related ways of liberating fixated thinking: abstracting and rerepresenting. Although both techniques helped designers generate original ideas, not all the added ideas fit the problem constraints. We discuss ways the results might be used to generate reflective design aids that help designers to first generate original ideas and later prune them.

This paper presents a framework for teaching design engineering in a global context using innovative technologies to enable distributed teams to work together effectively across international and cultural boundaries. The Digital Libraries for Global Distributed Innovative Design, Education, and Teamwork (DIDET) Framework represents the findings of a 5-year project conducted by the University of Strathclyde, Stanford University, and Olin College that enhanced student learning opportunities by enabling them to partake in global, team-based design engineering projects, directly experiencing different cultural contexts and accessing a variety of digital information sources via a range of innovative technology. The use of innovative technology enabled the formalization of design knowledge within international student teams as did the methods that were developed for students to store, share, and reuse information. Coaching methods were used by teaching staff to support distributed teams and evaluation work on relevant classes was carried out regularly to allow ongoing improvement of learning and teaching and show improvements in student learning. Major findings of the 5-year project include the requirement to overcome technological, pedagogical, and cultural issues for successful eLearning implementations. The DIDET Framework encapsulates all the conclusions relating to design engineering in a global context. Each of the principles for effective distributed design learning is shown along with relevant findings and suggested metrics. The findings detailed in the paper were reached through a series of interventions in design engineering education at the collaborating institutions. Evaluation was carried out on an ongoing basis and fed back into project development, both on the pedagogical and the technological approaches.

According to both cybernetics and general system theory, a subject develops and uses an adequate model of a system to widen his/her knowledge about the system. Models are then the interface between a subject and a real-world system to solve a problem and to construct knowledge. Hence, evaluating these models is crucial to ensure the quality of the constructed knowledge. We propose here an evaluation framework to assess complex models based on the intrinsic properties of these models as well as the properties of the derived knowledge. A series of 40 evaluation criteria are proposed under the four systemic axes: ontology, functioning, evolution, and teleology. Through a case study, we show how our evaluation model allows both presenting a given model and assessing it.

This paper presents the use of first-order logic to improve upon currently employed engineering design knowledge management techniques. Specifically, this work uses description logic in unison with Horn logic, to not only guide the knowledge acquisition process but also to offer much needed support in decision making during the engineering design process in a distributed environment. The knowledge management methods introduced are highlighted by the ability to identify modeling knowledge inconsistencies through the recognition of model characteristic limitations, such as those imposed by model idealizations. The adopted implementation languages include the Semantic Web Rule Language, which enables Horn-like rules to be applied to an ontological knowledge base and the Semantic Web's native Web Ontology Language. As part of this work, an ontological tool, OPTEAM, was developed to capture key aspects of the design process through a set of design-related ontologies and to serve as an application platform for facilitating the engineering design process. The design, analysis, and optimization of a classical I-beam problem are presented as a test-bed case study to illustrate the capabilities of these ontologies in OPTEAM. A second, more extensive test-bed example based on an industry-supplied medical device design problem is also introduced. Results indicate that well-defined, networked relationships within an ontological knowledge base can ultimately lead to a refined design process, with guidance provided by the identification of infeasible solutions and the introduction of “best-case” alternatives. These case studies also show how the application of first-order logic to engineering design improves the knowledge acquisition, knowledge management, and knowledge validation processes.

Public perception of engineering recognizes its importance to national and international competitiveness, economy, quality of life, security, and other fundamental areas of impact; but uncertainty about engineering among the general public remains. Federal funding trends for education underscore many of the concerns regarding teaching and learning in science, technology, engineering, and mathematics subjects in primary through grade 12 (P-12) education. Conflicting perspectives on the essential attributes that comprise the engineering design process results in a lack of coherent criteria against which teachers and administrators can measure the validity of a resource, or assess its strengths and weaknesses, or grasp incongruities among competing process models. The literature suggests two basic approaches for representing engineering design: a phase-based, life cycle-oriented approach; and an activity-based, cognitive approach. Although these approaches serve various teaching and functional goals in undergraduate and graduate engineering education, as well as in practice, they tend to exacerbate the gaps in P-12 engineering efforts, where appropriate learning objectives that connect meaningfully to engineering are poorly articulated or understood. In this article, we examine some fundamental problems that must be resolved if preengineering is to enter the P-12 curriculum with meaningful standards and is to be connected through learning outcomes, shared understanding of engineering design, and other vestiges to vertically link P-12 engineering with higher education and the practice of engineering. We also examine historical aspects, various pedagogies, and current issues pertaining to undergraduate and graduate engineering programs. As a case study, we hope to shed light on various kinds of interventions and outreach efforts to inform these efforts or at least provide some insight into major factors that shape and define the environment and cultures of the two institutions (including epistemic perspectives, institutional objectives, and political constraints) that are very different and can compromise collaborative efforts between the institutions of P-12 and higher education.

This is the second paper in the series on shape decompositions and their use as shape approximations. This time we investigate hierarchical and topological shape decompositions or hierarchies and topologies. We showed earlier that bounded decompositions behave the same way as shapes do. The same holds for hierarchies and topologies, which are special kinds of bounded decompositions. They are distinguished by their algebraic structures, which have many important properties to facilitate their application as shape approximations. We provide an account of their properties with an emphasis on their application.

Ontologies are an emerging means of knowledge representation to improve information organization and management, and they are becoming more prevalent in the domain of engineering design. The task of creating new ontologies manually is not only tedious and cumbersome but also time consuming and expensive. Research aimed at addressing these problems in creating ontologies has investigated methods of automating ontology reuse mainly by extracting smaller application ontologies from larger, more general purpose ontologies. Motivated by the wide variety of existing learning algorithms, this paper describes a new approach focused on the reuse of domain-specific ontologies. The approach integrates existing software tools for natural language processing with new algorithms for pruning concepts not relevant to the new domain and extending the pruned ontology by adding relevant concepts. The approach is assessed experimentally by automatically adapting a design rationale ontology for the software engineering domain to a new one for the related domain of engineering design. The experiment produced an ontology that exhibits comparable quality to previous attempts to automate ontology creation as measured by standard content performance metrics such as coverage, accuracy, precision, and recall. However, further analysis of the ontology suggests that the automated approach should be augmented with recommendations presented to a domain expert who monitors the pruning and extending processes in order to improve the structure of the ontology.

This paper presents a formal analysis of design process dynamics. Such a formal analysis is a prerequisite to come to a formal theory of design and for the development of automated support for the dynamics of design processes. The analysis was geared toward the identification of dynamic design properties at different levels of aggregation. This approach is specifically suitable for component-based design processes. A complicating factor for supporting the design process is that not only the generic properties of design must be specified, but also the language chosen should be rich enough to allow specification of complex properties of the system under design. This requires a language rich enough to operate at these different levels. The Temporal Trace Language used in this paper is suitable for that. The paper shows that the analysis at the level of a design process as a whole and at subprocesses thereof is precise enough to allow for automatic simulation. Simulation allows the modeler to manipulate the specifications of the system under design to better understand the interlevel relationships in his design. The approach is illustrated by an example.