Koussoube S., Ndoutoume J.L., Noussi R. (1996). On an interactive system to support the management of industrial rubber tree (hevea) plantations. In :
Moukeli P. (ed.). CARI'96 : actes du 3ème colloque africain sur la recherche en informatique = CARI'96 : proceedings of the 3rd African conference on research in computer science. Paris : ORSTOM, p. 426-435. (Colloques et Séminaires). CARI'96 : Colloque Africain sur la Recherche en Informatique = CARI'96 : African Conference on Research in Computer Science, 3., Libreville (GAB), 1996/10/09-16. ISBN 2-7099-1333-X. ISSN 0767-2896.
Titre du document
On an interactive system to support the management of industrial rubber tree (hevea) plantations
Année de publication
1996
Type de document
Partie d'ouvrage
Auteurs
Koussoube S., Ndoutoume J.L., Noussi R.
In
Moukeli P. (ed.), CARI'96 : actes du 3ème colloque africain sur la recherche en informatique = CARI'96 : proceedings of the 3rd African conference on research in computer science
Source
Paris : ORSTOM, 1996,
p. 426-435 (Colloques et Séminaires). ISBN 2-7099-1333-X ISSN 0767-2896
Colloque
CARI'96 : Colloque Africain sur la Recherche en Informatique = CARI'96 : African Conference on Research in Computer Science, 3., Libreville (GAB), 1996/10/09-16
The use of remote sensing and Geographic Information System (GIS) for the needs of industrial plantations management, has occupied more and more important place during the last years. In the case of the hevea culture, this can be explained by the need expressed by agronomists to achieve development, inventory, maintenance and farming operations, on the plantations, by means of aerial images, topographic and descriptive data. Most of these operations integrate decisional procedures using models from Statistics, Operational Research and Artificial Intelligence. The DSS presented here has been designed in this context. Its main particularity rest on the fact that most of the decisional procedures implemented need a simultaneous use of the three types of data cited above. Our approach uses a modelization of the user's conceptual space in the shape of a state graph. A session in the system may be viewed as a partial exploration of that graph. The control of this exploration is completely left to the user. Furthermore the system provides the user with a set of operators allowing him to generate new states from the current one, and a set of functions based on some decisional models for the evaluation of the generated states. (Résumé d'auteur)
