Journal of Integrated Design and Process Science

This paper describes how the concept of Personal Assistant Agent (PA) can be used to improve CSCW and to support knowledge management. We claim that available groupware and collaborative tools do not exploit domain knowledge efficiently enough. Here we describe a PA using domain knowledge to help a user in a collaborative environment. The PA records user's tasks on the fly, capturing knowledge. It uses domain knowledge to organize them, and offers help and suggestions regarding the user's work. A PA communicates with other PAs to exchange information about users and about how they solve problems. Therefore, one can reuse everyone's experiences. The interaction with a user occurs through window dialogs or by means of natural language. Our PAs are to be used in an educational environment to aid remote groups of students do collaborative mechanical engineering design.

This paper presents a multi-agent architecture and a supporting data structure suited for the verification of constraints in a mechanical engineering design context. The data structure is composed of three levels, namely an environment level, providing a link to existing mechanical engineering software, an expert level, capturing product requirements and trade knowledge through constraints formulation and finally a product level, dealing with specific products under development. Constraints and models are assigned distinctive attributes, which are then exploited by the agents in their verification strategies, which are rightly termed as constraint or model driven. Multi-agent task planning provides an additional level of optimization for the operation of agents in their mission to check the product against the various constraints throughout the design process.

In a complex engineering design, multiple engineering design teams are required to work cooperatively in order to design a single product. However multiple disciplinary teams in different departments and companies often use different systems with different modelling methods and techniques. While the parallelism of this approach can reduce the design life cycle, the complexity of the optimisation problem can lead to difficulties. Agent technology can alleviate some of the difficulties of concurrent design. In this paper, we present an architecture in which agent technology is used to model interactions between design systems, and thereby enable design agents from various design disciplines to explore the design problem and solution space. We also introduce a multiple-evolutionary approach with an automated negotiation mechanism to enable agents to exchange design solutions and to reach an agreed optimised solution for a global design problem.

Engineering design activities in distributed environments like the Web require fault tolerance and concurrent access to shared resources such as databases and Web servers. Such activities are generally dynamic, cooperative, long-lived, interactive and non-prescriptive. We propose a new multi-agent transaction model, which is based on extended transactions and multi-agent technologies. The novelty of this model is that it automatically customises transactions to the requirements of design activities. In addition, this model is believed to improve concurrency, fault tolerance, facilitate interaction between and co-operation among the participating systems involved in design activities. The proposed model is formally specified using CCS (Calculus of Communicating Systems) language. Formalisation is crucial in ensuring the correctness, reliability, and recovery of multi-agents transactions, given the complex and unreliable nature of the distributed design activities.

This paper proposes an open data management facility called Collaboration Oriented Data Agent (CODA) for computer supported collaborative work in design community. CODA provides a proactive service for engineering data management, supporting data consistency, concurrent access, multidisciplinary teamwork, and engineering workflow. It is equipped with a number of advanced technologies and several important behavioral and functional features. This paper is concerned with combining agent based technologies with project oriented data management to develop CODA based framework. The design and implementation issues have been identified and addressed in terms of agent paradigm to facilitate engineering data management. This framework has been incorporated into a prototype of a multidisciplinary design optimization environment.

This paper presents a distributed dynamic object model that is aligned with the concept of design history in the context of design problem solving activities. A distributed object model is proposed as an enabling feature for distributed CAD (Computer-Aided Design) allowing teams to work cooperatively, accessing and exchanging information at run time in a distributed engineering environment. The architecture of the proposed CAD environment allows for the artifact properties to be associated with any relevant aspect of the design process, including those related with the artifact specification, but also with the organization hierarchy, planning and process workflow.

Feature is a fundamental design unit in modern solid modeling systems, and this makes a well defined feature interface a matter of concernment in the component-based distributed modeling environment. This paper introduces a feature-based modeling service framework and proposes an interface model of feature-based modeling service toward such a modeling environment, including the support of remote feature attachment and semantic maintenance. A prototype-based feature interface definition strategy using procedural attaching and declarative validation mechanism is discussed in details and how the interface model solves the problems in distributed environment, such as locality-transparent model reusing, is also explained.

Feature is a fundamental design unit in modern solid modeling systems, and this makes a well defined feature interface a matter of concernment in the component-based distributed modeling environment. This paper introduces a feature-based modeling service framework and proposes an interface model of feature-based modeling service toward such a modeling environment, including the support of remote feature attachment and semantic maintenance. A prototype-based feature interface definition strategy using procedural attaching and declarative validation mechanism is discussed in details and how the interface model solves the problems in distributed environment, such as locality-transparent model reusing, is also explained.

Getting the requirements right is essential for good design. A currently popular technique for capturing requirements is the specification of use cases. However getting a conflict free, complete and representative set of use case descriptions can be difficult. In this paper we look at how the RECOCASE approach provides collaborative requirements RECOnciliation through the use of a computer aided software engineering (CASE) tool. In the approach, use case descriptions are entered asynchronously in natural language from individual stakeholders. The tool automatically produces a visual representation of the requirements to assist the group to identify and resolve conflicts and produce a representative requirements specification.

Getting the requirements right is essential for good design. A currently popular technique for capturing requirements is the specification of use cases. However getting a conflict free, complete and representative set of use case descriptions can be difficult. In this paper we look at how the RECOCASE approach provides collaborative requirements RECOnciliation through the use of a computer aided software engineering (CASE) tool. In the approach, use case descriptions are entered asynchronously in natural language from individual stakeholders. The tool automatically produces a visual representation of the requirements to assist the group to identify and resolve conflicts and produce a representative requirements specification.

In this paper a real example of a globally distributed collaborative design project is analyzed. Some of the problems identified as a result of this analysis include: 1- the lack of a complete design environment prior to the modeling stage, and 2- the need for a flexible and intuitive collaborative environment. To address these problems we propose some preliminary research results of task ontology, common to both cultures, in order to provide globally distributed design teams with a personal assistant agent.

In this paper a real example of a globally distributed collaborative design project is analyzed. Some of the problems identified as a result of this analysis include: 1- the lack of a complete design environment prior to the modeling stage, and 2- the need for a flexible and intuitive collaborative environment. To address these problems we propose some preliminary research results of task ontology, common to both cultures, in order to provide globally distributed design teams with a personal assistant agent.