Design is distinguished in many respects from other intellectual endeavors.
Simon explains that, “The natural sciences are concerned with how things are. … Design, on the other hand, is concerned with how things ought to be, with devising artifacts to attain goals.” ¹ Designers interpret design problems and formulate design objectives. They explore opportunities and represent possibilities, reflect on prior designs and communicate design proposals to others. Designs are contingent upon the contexts in which they arise. ² When a designer creates a design, they do so in light of thier current understanding of what is to be accomplished and the opportunities available to them at that time.

Contingency presents a twofold challenge to designers. First, change is a prominent and growing force in our time. Peter Drucker argues, “No century in recorded history has experienced so many social transformations and such radical ones as the twentieth century.” ³ Traditional design approaches focus on the specification of the object to be realized. When the context changes, the contingent relations upon which the design are based break and the design must be revised to regain its persuasiveness. ⁴ Because designers are constrained by limitations of resources and time, change presents a significant impediment to exploration and development of design work in a traditional approach. Second, the context of design and the objects of design themselves are more complex now than ever before. While industrialization has significantly expanded our capacity to produce the same or similar objects millions of times over and with ever increasing economy, it has not led to a similar increase in our ability to address the complex conditions of our contemporary world. Complexity challenges designers to be cognizant of both the interrelatedness and particularity of phenomena that characterize design problems. However, human cognition is limited in its ability to deal with complexity, and constraints typical to professional practice impede the exploration and thorough study of contexts that inform design work.

Understanding and devising means for dealing with change and complexity in design is a beneficial and timely area of research. Simon argues that the challenge of research in design, “is to show how empirical propositions can be made at all about systems that, given different circumstances, might be quite other than they are.'' The proper means of studying design, he suggests, is to observe the way in which means are adapted to environments, “and central to that is the process of design itself.” ⁵

Our research has benefited greatly in this regard from an exceptional example of design process. The Morphosis Architects/George Yu IFCCA competition project provides a concrete and intelligible model of how change and complexity can be addressed through design. Our contention is that the approach exhibited therein renders the design as a system of interacting parts. The systemic nature of the process and resulting artifact presents clear opportunities for description and analysis in terms of systems, accords with accounts of designer cognitive behavior, and creates apparent opportunities for computational support. Drawing on systems models of organizational management, we hypothesize that the capacity of designers to deal with change and complexity can be increased by focusing on the management of interactions that realize the design, rather than on the specification of the design itself. To this end, we have conducted exploratory research aimed at the development of a parametric modeling application intended to facilitate change in design through the management of interactions.

Our work is presented here in four parts. In Chapter 1, we describe the Morphosis/Yu project and the properties that make it relevant for our research. We situate our observations in the domains of General Systems Theory, organizational management, Cognitivist research in designer behavior, parametric modeling, and a class of design support systems known as Design Space Explorers. We outline major concepts found in each of these research areas and discuss their relevance to the issues at hand. In Chapter 2, we recast the Morphosis/Yu project as an exemplary research problem in parametric modeling and describe particular objectives for research in this area. We propose a novel propagation-based, parametric modeling application intended to facilitate change and development in design. The proposed application is organized around the interactions of three computational entities: Assembly, Behavior and Scenario. Underpinning these entities is a parallel computing architecture known as the Systolic Array. We describe how Behaviors manage interactions and facilitate change in the model over time through an implicit relational modeling approach that takes the form of the Dynamic Proxy pattern. Chapter 3 provides a detailed description of our implemented system. We outline in detail the high level organization, particular concepts and interactions that characterize the system. We describe our user interface studies and vectors of inquiry, both past and ongoing, in the development of the application. Having developed a demonstrable system, we presented the system to experts in the domain of architectural design and parametric modeling. In Chapter 4, we give an account of our discussions with these individuals. They provide an initial evaluation of the application, our working hypothesis and approach. We conclude with a summary of the current status of system development and recommendations for further research.