文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > V. Agent-Based Framework for Simulation and Support of Dynamic Engineering
侵权投诉

V. Agent-Based Framework for Simulation and Support of Dynamic Engineering

Abstract. The paper reports on the first results of the Productivity Simulation Initiative (PSI) project of Cadence Design Systems GmbH. The project addresses the problem of fine-grained modeling and simulation of dynamic engineering design processes in or

6 Discussion and Related Work

The constellation of projects pioneered R&D in agent-based engineering design support and automation began to appear about a decade ago e.g. ([3], [4], [1]). These findings initially motivated PSI project. Some projects of the “second wave” ([18],

[5]) helped to specify the focus of PSI in automating the near-optimal arrangement of DEDPs in terms of their productivity. Besides that PSI aims to provide the industrial strength solution in the mentioned niche.

DEDP modeling framework and simulator prototype implementation is based on research in: dynamic planning process modeling, methodologies and frameworks for the design and implementation of multi-agent systems.

DEDP modeling framework in the part of organizational and actor-related knowledge representation bases itself on the frameworks ([12], [20], [6], [7]). PSI contribution in this part is the incorporation of roles, actors with its specific subclasses, teams of actors, negotiation context in one coherent ontology and its binding to the engineering design domain by, e.g., introducing the sub-ontologies of Design Artifacts and Software Tools [9]. The main emphasis of PSI DEDP ontology is the model of a dynamic team of designers which are formed through contracting negotiations and perform dynamically orchestrated processes. In the part of process modeling PSI borrows the ideas and the approach from ([2], [6], [11]). In DEDP ontology engineering design processes are modeled as tasks composed of subtasks and atomic activities. Similarly to [17] subtasks and activities may have weak and strong dependencies. However the knowledge on these dependencies is in PSI local and differs from actor to actor as specified in their partial local plans. Similarly to

[11] activities have pre-conditions, post-conditions and post-effects. However PSI ontology constrains the semantics of pre-/post- conditions and effects by making them sub-classes of an event concept. Material inputs and outputs semantically and structurally belong to PSI Design Artifacts ontology. Some inspirations for the development of agent reasoning mechanisms in PSI were provided by RAPPID set based reasoning framework [18] and RACING negotiation framework [6]. PSI extends these initial percepts to the family of negotiation mechanisms: task allocation, design re-use, choice of a software tool to perform a design activity [10]. 7 Conclusions and Future Work

As reported in the paper the first outcomes of PSI are (i) the formal framework for modeling DEDPs, (ii) implemented simulation prototype of DEDP simulator software tool, (iii) initial PSI testbed comprising DEDP execution logs built for two mentioned application scenarios, (iv) results of the initial experiments which prove the feasibility of concept implementation.

The contribution of the modeling framework is the provision of the DEDP model in the form of the set of DEDP-full ontologies. These ontologies were simplified to DEDP-light version and used in the design of the meta-model of DEDP-MAS, in the implementation of PSI DEDP execution log questionnaire and editor as well as in the

上一页第8页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email