Prediction of the production of oil palms (Elaeis guineensis Jacq)

Knowledge of the expected production for the next six months allows the managers of agro-industrial plantations of oil palms to better organize their technical and financial management. However, the estimation methods must be easy to apply, yet sufficiently accurate. This study was initiated in order to contribute to the development of a model for the estimation of oil palm production on the one hand and to evaluate the sensitivity of the method face the data necessary for this forecast. These data are of two types: the actual production data and the inflorescence and bunch count data. The experiment was carried out on the experimental station Robert-Michaux of the CNRA of Dabou, located in the South-East of the Côte d�Ivoire. The proposed methodology is based on the duration of plan formation and maturation. It takes 5.5 to 6 months between the exit of the female inflorescence and the harvest of the corresponding ripe bunch. Counting of bunches and female inflorescences in the crown provides information on the number of rips bunches to be harvested in the next six months. The evolution of the average weight of bunches harvested previously makes it possible to predict the average weight of bunches that will be harvested during the same period. The model makes it possible to estimate the tonnage of production for the next six months provided that, for a plantation unit, it can be applied to representative samples of the whole. The yield is translated into production at different scales taking into account planting density. The estimated production variations and those of the previous production make it possible to estimate production in the months to come. The results are very satisfactory, with error accuracy of 9 %. They demonstrate the economic and technical interest of such a method in the case of production sites with little information on the conditions of oil palm cultivation.