SoD-TEAM: Problem-Solving Methodology in Collaborative Design

计算机和信息科学与工程局（CISE）计算机和网络系统（CNS）设计科学（SoD）计划提案编号：0613550 P/I： Steven Tanimoto PI的部门：计算机科学与工程机构：华盛顿大学获奖： $449，642标题： “SoD-TEAM：协作设计中的问题解决方法“该项目侧重于促进设计团队成员之间协作的软件工具。该项目允许设计团队管理替代设计空间的探索部分。此外，它还支持团队成员之间关于设计方案的评估和沟通。该项目的一个目标是为软件设计师开发一种方法和框架，该方法和框架基于研究，将导致理解如何支持最终用户对解决方案空间的探索。此外，该项目还促进了具有不同兴趣和专业知识的设计师之间的合作，并需要拥有私人以及共享基于群体的共同设计。构成该项目基础的问题解决理论涉及“状态空间搜索”。状态空间搜索方法既适用于正式的问题解决（在没有人为干预的自动化是可能的），也适用于复杂的设计问题，其中某些方面可能是“模糊的”。然而，该项目提出的方法是将两种类型的软件合并，并创建一个新的软件：允许自动设计过程和人类设计师之间协作的软件，以及一个不仅可以看到特定设计的细节，而且可以看到可能的设计空间的实质性部分的软件。为了支持这个项目，提案人开发了一个初步研究原型软件系统的核心，以支持设计与设计空间的透明显示。这个系统，T-Star，实现了一个人机界面，支持使用人工智能的方法来解决问题，包括到目前为止已经探索的解决方案空间的一部分的视图。对探索的空间背景进行视觉再现有助于地面设计团队的协作。项目经理：Anita J. La Salle日期：2006年6月26日

Directorate for Computer and Information Science and Engineering (CISE)Division Computer and Network Systems (CNS)Science of Design (SoD) ProgramProposal Number: 0613550 P/I: Steven Tanimoto PI's Department: Computer Science and Engineering Institution: University of WashingtonAward: $ 449,642Title: "SoD-TEAM: Problem-Solving Methodology in Collaborative Design"This project focuses on software tools that facilitate collaboration among members of a design team. The project permits a design team to manage the explored portions of a space of alternative designs. In addition, it supports evaluation and communication among team members about design alternatives. One goal of the project is to develop a methodology and a framework for software-designers that is based on studies that will lead to understanding how to support end-user exploration of solution-spaces. In addition the project facilitates collaboration among designers with different interests, expertise and the need to have both private as well as share common group-based designs. The theory of problem solving that forms the basis of the project involves "state-space search." The state-space search methodology applies to both formal problem solving (where automation without human intervention is possible) and to complex design problems in which some aspects can be "fuzzy". However, the approach proposed with this project is to merge the two genres of software and create a new one: software that allows a collaboration among automatic design processes and human designers, and one that makes visible not only the details of particular designs but also a substantial piece of the space of possible designs. In support of this project, the proposers developed the core of an initial research prototype software system to support design with transparent display of design spaces. This system, T-Star, implements a human interface that supports problem-solving using an approach from artificial intelligence that includes a view of the portion of the solution space that has been explored so far. Having a visual rendition of the spatial context of exploration helps ground design team collaboration.Program Manager: Anita J. La SalleDate: June 26, 2006