MECIMPLAN: an agent-based methodology for planning. (15th May 2009)

• Record Type: Journal Article
• Authors: Castillo, Jose Miguel; Ossowski, Sascha; Pastor, Luis
• Is Part Of: International journal of intelligent information and database systems. Volume 3:Number 2(2009)
• Page Start: 125
• Page End: 145
• Publication Date: 2009-05-15
• Subjects: short-term planning -- long-term planning -- strategic planning -- tactical planning -- projects -- agents -- multi-agent systems -- MAS -- agent-based systems -- methodology -- artificial intelligence -- information systems -- database systems -- intelligent behaviour -- adaptive behaviour; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

Planning is a human process that we constantly use in our daily life. However, when the planning process exceeds from our mental capacity the use of computers can be a good solution to complement our natural planning capacity. The availability of planning functionalities is essential for information and database systems to show intelligent and adaptive behaviour. In order to build such a functionality in a principled manner, a methodology is required that is flexible enough to be tailored to the specific needs of an information system for a particular application. In this article, we will introduce a methodology that draws upon agent technology so as to incrementally construct models for information systems with planning capabilities. In this methodology, each agent is responsible for carrying out a specific task within the planning process, such as classifying, qualifying, assigning or optimising an intelligent search. The use of agents, as software entities in charge of developing specific reasoning processes, facilitates the dynamic generation and refinement of solutions with desired functionality. We have developed two software prototypes that prove the adequacy of our approach in different real-world domains.