Juan C. Garcia-Ojeda

O-MaSE: a customizable approach to developing multiagent development processes

This paper describes the Organization-based MultiagentSystem Engineering (O-MaSE) Process Framework, which helps processengineers define custom multiagent systems development processes.O-MaSE builds off the MaSE methodology and is adapted from theOPEN Process Framework (OPF). OPF implements a Method Engineeringapproach to process construction. The goal of O-MaSE is to allowdesigners to create customized agent-oriented software development processes.O-MaSE consists of three basic structures: (1) a metamodel, (2)a set of methods fragments, and (3) a set of guidelines. The O-MaSE metamodel defines the key concepts needed to design and implementmultiagent systems. The method fragments are operations or tasks thatare executed to produce a set of work products, which may include models,documents, or code. The guidelines define how the method fragmentsare related to one another. The paper also shows two O-MaSE processexamples.

O-MaSE: a customisable approach to designing and building complex, adaptive multi-agent systems

The complexity and scope of software systems continues to grow. One approach to dealing with this growing complexity is the use of intelligent, multi-agent systems. However, due in part to its relative infancy when compared to other software paradigms, the use of multi-agent systems has yet to be used extensively in industry. One reason is the lack of industrial strength methods and tools to support multi-agent development. This paper presents the organisation-based multi-agent software engineering (O-MaSE) methodology framework, which integrates a set of concrete technologies aimed at facilitating industrial acceptance. Specifically, O-MaSE is a customisable agent-oriented methodology based on consistent, well-defined concepts supported by plug-ins to an industrial strength development environment, agentTool III.

The O-MASE Methodology

Today’s software industry is tasked with building evermore complex software applications, and multiagent system technology is a promising approach capable of meeting these new demands. Unfortunately, multiagent systems have not been widely adopted in industry for reasons that include lack of industrial strength methods and tools to support multiagent development. Method engineering, an approach to constructing processes from a set of existing method fragments, has been suggested as a solution to this problem. This chapter presents the Organization-based Multiagent Software Engineering (O-MaSE) methodology framework, which integrates a set of concrete technologies aimed at facilitating industrial acceptance. Specifically, O-MaSE is a customizable agent-oriented methodology based on consistent, well-defined concepts supported by plug-ins to an industrial strength development environment, agentTool III. O-MaSE is defined, and demonstrations of customizing O-MaSE for the CMS problem as well applying the customized process to the CMS design are presented.

agentTool process editor: supporting the design of tailored agent-based processes

This paper describes the agentTool Process Editor (APE), an Eclipse plug-in based on the Eclipse Process Framework. The aim of APE is to facilitate the design, verification, and management of custom agent-oriented software development processes. APE provides five basic structures. The Library is a repository of agent-oriented method fragments. A Process Editor allows the management of tailored processes. Task Constraints help process engineers specify guidelines to constrain how tasks can be assembled, while a Process Consistency mechanism verifies the consistency of tailored processes against those constraints. Finally, the Process Management integrates APE with the agentTool III development environment and provides a way to measure project progress using Earned Value Analysis.

Planning evacuation routes with the P-graph framework

The P-graph framework is proven to be highly effective in solving Process Network Synthesis. In the present work, the P-graph framework has been adopted for solving the routing and scheduling of evacuees, facing a life-threatening situation. First the building evacuation problem is represented by means of a P-graph model, which is then transformed into a time-expanded process network synthesis (PNST) problem that can be algorithmically handled by the P-graph framework. In the proposed method, each location in the building and their passages are given by a set of attributes to be taken in the evacuation route planning. In addition to the globally optimal solution of the building evacuation problem, the P-graph framework provides the n-best suboptimal solutions, when computational possible. The viability of the proposed model is illustrated by an example.

Paving the way for implementing multiagent systems: integrating gaia with Agent-UML

This paper describes how to refine a Gaia design by applying agent-oriented extensions of UML. First, we show how the Gaia Interaction model can be improved by applying the first two layers of the Agent Interaction Protocol (AIP) of AUML. Second, Gaia Agent and Service models are refined by applying the AIPs third layer combined with Extended UML Class Diagrams. Third, Gaia Organisational Structure is enriched by applying the Aalaadin model. The final aim of the whole process is to obtain a more concrete design closer to implementation. We demonstrate how the refinement process can be applied to the development of an agent-based system for conference management.

Extending the Gaia Methodology with Agent-UML

This paper describes how to integrate a design produced by Gaia with the AIPs layers of AUML, taking into consideration Bauerýs extensions to UML Class diagrams in the context of agent-oriented development. We take as a case study the development of an agent-based system for conference management.

Process Network Synthesis for Benzaldehyde Production: P-graph Approach

a Department of Chemical and Environmental Engineering, Universidad Nacional de Colombia, Bogotá, Colombia b Department of Computer Science and Systems Technology, University of Pannonia, Veszprém, Egyetem u.10 H-8200, Hungary c Faculty of Information Technology, Pázmány Péter Catholic University, Práter u. 50/A, 1083 Budapest, Hungary d Department of Systems Engineering, Universitaria de Investigación y Desarrollo, Calle 9 No. 23-55, Bucaramanga Colombia e Department of Chemical Engineering, Kansas State University, 1005 Durland Hall, Manhattan, Kansas 66506, USA argoti@k-state.edu