Multiagent model of dynamic design

2002

Is it possible to design for creativity? This is one of the most important research questions in AI and Design. This paper raises the question whether automated software agent design can be creative. Not only is the artefact dynamic in the sense that it adapts to its environment, it is also autonomous: an agent decides on its own when to be modified and by whom. An agent’s functionality may evolve beyond the expectations of its designers and/or users resulting in very new, unique artefacts. Is this type of design creative? Is the process creative? The result?

International Journal of Human-Computer Studies, 2002

Computer-based multimedia technologies allow designers to construct interactive and animated graphical presentations to communicate dynamic information. The conventional wisdom is that such presentations are more effective than printed materials. This paper presents research that critically examines this assumption. Design guidelines and principles were derived from a cognitive process model of multimodal comprehension. These guidelines and principles were used to create several expository presentations in two domains}the concrete domain of mechanical systems and the abstract domain of computer algorithms. A series of experiments evaluated the efficacy of these presentations and compared them with other kinds of presentations such as books, CD-ROMs and animations. The experiments also compared computer-based interactive graphical presentations and static printed presentations containing the same information. Experimental results suggest that the communicative efficacy of multimodal presentations is more related to their match with comprehension processes than with the interactivity and dynamism of the presentation media. The results support a model-based approach to the design of multimodal expository presentations of dynamic information. The comprehension model and corresponding design guidance should aid designers in building interactive graphical presentations that are more effective than intuitive designs in communicating dynamic content.

h i g h l i g h t s

Eshaq, A, Khong, C, Neo, K, Neo, M and Ahmad, S l.(eds), CAADRIA, 2002

Abstract. This paper presents the outline of experiments with a behaviourbased computational model of creativity in design called Creative Design Situations. This model aims to allow the study of social creativity through the computational implementation of a community of creative design agents.

2015

Abstract. Concept invention refers to the act of discovering new concepts for the first time. Concept invention may require a new interpretation that motivates new mappings of the situation prior to the discovery of a new concept. This paper is concerned with developing intelligent design agents that would be capable of inventing creative concepts of design forms, shapes and compositions while involved in the design process. A new approach of exploiting the old notion of “displacement of concepts ” with regard to concept invention in designing in the form of mobile design agents is adopted and utilised in developing the structure of intelligent mobile agent-based system. 1. Invention of Design Concepts Design concepts are composed of rich sets of categorised abstractions of the mutual relationships between observations and actions. This research adopts the view that the process of designing is situated (Reffat and Gero

Anais do International Conference on Engineering Design, ICED, 2007

This paper presents a set of concepts that provides the cognitive and computational foundations for answering fundamental questions about designerly behaviour. A design agent should be dynamic and able to handle changes in external representations that describe its designs and changes in how it is guided by its past experiences. Such a view of agency is afforded by situated design computing and can represent and explain much designerly behaviour. It can model how a designer can ...

Proceedings of the 14th International Conference on Engineering Design ICED03, 2003

It is a challenge for designers in traditional organisations to work together in teams to achieve a consensus based solution space optimality [1] and this challenge is further compounded by the dispersion of the design information, design personnel and knowledge across the extended enterprise environment. Extensive work has been carried out on passive information management systems for design, however, very little work has been carried out in the muchneeded area of proactive, holistic knowledge management and presentation systems required by the extended enterprise. In this paper we propose an agent-based system to access, retrieve and present information (both visually and auditorially) to distributed design teams, in such a way that their collective conceptual space is expanded [2], learning strategies are supported [3] and design solutions are optimised [4]. This agent based information system will support the distributed design process whilst considering human and ICT protocols, for the purposes of enhancing design, learning, creativity, communication and productivity.

Gerber, D., Huang, A. and Sanchez, J., eds., ACADIA 2014 Design Agency Proceedings. Proceedings of the 34th Annual Conference of the Association for Computer Aided Design in Architecture, October 23-25, 2014, Los Angeles, California, pp. 479-486., 2014

In this paper we look at interactive architecture from a design theoretical/methodological point of view. In particular, we aim to outline how principles from agent theory can form a theoretical foundation for conception and following the design of interactive architecture. Interactive architecture is a special case of responsive buildings—buildings that can intelligently react to, or even anticipate, change. Interactive buildings have an internal representation of the user on which basis they reach decisions for the manipulation of the outside world. For the design of responsive and interactive buildings it is necessary to take change into account throughout the whole design process. This requires rethinking our notions about architecture and design processes. We define an agent, and show how it matches for objects, with increasing degrees of responsiveness. Agents by themselves are not enough to design interactive buildings—they also need to have a behavior defined. For this purpose we define attitudes as mode of operation for agents.

Multiagent Systems, Artificial Societies, and Simulated Organizations, 2003