AI EDAM: Artificial Intelligence for Engineering Design, Analysis and Manufacturing



An integrated agent-oriented approach to real-time operational design coordination


GRAHAM  COATES  a1 c1 , ALEX H.B.  DUFFY  a2 , IAN  WHITFIELD  a2 and WILLIAM  HILLS  a1
a1 Engineering Design Centre, University of Newcastle, Newcastle upon Tyne, United Kingdom
a2 CAD Centre, University of Strathclyde, Strathclyde, United Kingdom

Article author query
coates g   [Google Scholar] 
duffy ahb   [Google Scholar] 
whitfield i   [Google Scholar] 
hills w   [Google Scholar] 
 

Abstract

Within the engineering design community there is support for further research into the development of improved approaches to design management. Such research has lead to coordination being identified as an important and pervasive characteristic of many existing approaches (e.g., concurrent engineering and work-flow management). In this article, operational design coordination is proposed as the basis for an improved approach. This article also presents a novel integrated approach that incorporates the key elements of operational design coordination: coherence, communication, task management, resource management, schedule management, and real-time support. Through unifying these key elements, this approach provides an integrated means of managing design in a controlled and harmonious fashion. The approach also provides knowledge of the constituent techniques involved in operational design coordination, the interrelationships and dynamic interactions between them, and the knowledge used and maintained within and between them. The approach has been realized within an agent-oriented system called the Design Coordination System, which provides a systematic means of simultaneously coordinating operational management tasks and technical design tasks. To evaluate the approach, the system has been applied to an industrial case study involving the computational process of turbine blade design. This application has been shown to enable the structured undertaking of interrelated tasks by allocating and using resources of varying performance efficiency in an optimized fashion in accordance with dynamically derived schedules in a coherent, appropriate, and timely manner. This is achieved by managing tasks, their dependencies, and the information required to undertake them. In addition, the approach enables and sustains the continuous optimized use of resources by monitoring, forecasting, and disseminating resource performance efficiency. The approach facilitates dynamic scheduling and the subsequent enactment of the resulting schedules. Decision making for rescheduling is also incorporated within the approach such that it is only performed as and when appropriate. If rescheduling is performed, it is done so in parallel with task enactment such that resources continue to be utilized in an optimized manner.

(Received July 17 2002)
(Accepted September 15 2003)


Key Words: Agent-Oriented Approach; Design Management; Real-Time Coordination.

Correspondence:
c1 Reprint requests to: Graham Coates, School of Engineering, University of Durham, Durham, DH1 3LE, United Kingdom. E-mail: graham.coates@durham.ac.uk


0Comments