The model driven architecture approach : a framework for developing complex agricultural systems

Papajorgji Petraq J., Clark Ryan, Jallas Eric. 2009. The model driven architecture approach : a framework for developing complex agricultural systems. In : Advances in modeling agricultural systems. Papajorgji Petraq J. (ed.), Pardalos Panos M. (ed.). New York : Springer [Etats-Unis], 1-18. (Springer optimization and its applications, 25) ISBN 978-0-387-75180-1

Résumé : Development and application of crop models is increasingly constrained by the difficulty of implementing scientific information into an efficient simulation environment. Traditionally, researchers wrote their own models and tools, but as software has become much more complex, few researchers have the means to continue using this approach. New modeling paradigms provided by the software engineering industry can be successfully used to facilitate the process of software development for crop simulation systems. This chapter outlines a model driven architecture (MDA)-based approach to construct a crop simulation model. This new modeling paradigm is a Unified Modeling Language (UML)-based approach. A conceptual model of the problem is first constructed to depict concepts from the domain of the crop simulation and their relationships. The conceptual model is then provided with details about the role each of the concepts plays in the simulation. The multiplicity of the associations between concepts is determined, and the behavior of each of the objects representing concepts of the domain is defined. Mostly, an object¿s behavior in the crop simulation domain is expressed using equations. For this type of behavior, this new modeling paradigm offers a declarative way to write equations using attributes of objects participating in the conceptual diagram. For behavior that cannot be expressed through equations, a formal language is used to model behavior without the ambiguities that can be introduced by the use of natural language. Models can be validated and logical flows can be discovered before code generation. An Extensible Markup Language (XML) representation of the conceptual model is used by an engine that generates automatically executable code in several programming environments such as Java , Enterprise Java Beans, Visual Basic, and .NET. Results obtained from this new approach are presented, and they coincide with results obtained with other approaches.

Mots-clés Agrovoc : modèle de simulation

Classification Agris : U10 - Informatique, mathématiques et statistiques

Champ stratégique Cirad : Axe 1 (2005-2013) - Intensification écologique

Auteurs et affiliations

• Papajorgji Petraq J., University of Florida (USA)
• Clark Ryan
• Jallas Eric, CIRAD-DRS-VALO (FRA)

Source : Cirad - Agritrop (https://agritrop.cirad.fr/547725/)

Métriques

[ Page générée et mise en cache le 2024-03-30 ]