Probabilistic model of fruit removal during vibratory morello harvesting

The quantity of the fruit removed under specified harvest conditions and duration of the vibratory impact on a fruit-bearing tree are analysed in the present study, considering the harvest duration as a random variable. Defined are basic probabilistic-statistical characteristics of the removal of a fruit set belonging to a fruit-bearing tree subjected to vibrations, thus establishing a probabilistic model describing mathematically the vibratory fruit removal under different specified harvest conditions. By means of both conducted and previously published morello harvested experiments, it is established that under different harvest conditions and characteristics of the vibratory impact on a fruit-bearing tree, the relative quantity of the non-detached fruit from the tree and the probability for non-sufficient harvest duration are coincident and both decrease exponentially when the vibratory impact duration for the tree increases. Tests of the concordance between the experimental data and the statistical hypothesis for exponential probability distribution of the harvest duration show that at the 0.01 level of significance, the experimental data does not contradict the raised hypothesis. The value of the correlation coefficient between the experimental and predicted results is approximate to 0.99. (C) 2004 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd.