Models for predicting the mass of apricot fruits by geometrical attributes (cv. Shams, Nakhjavan, and Jahangiri)

Fruits with the similar weight and uniform shape are in high demand in terms of marketing value. Therefore, an awareness of grading fruit based on weight is crucial. A part of this research was aimed to present some physical properties of three Iranian apricot cultivars; (Shams, Nakhjavan, and Jahangiri). In addition, apricot mass was predicted by different physical characteristics with linear and nonlinear models as three different classifications: (1), single or multiple variable regressions of apricot dimensional characteristics. (2), single or multiple variable regressions of apricot projected areas and (3), estimating apricot mass based on its volume. All properties considered in the current study were found to be statistically significant at the 1% probability level. The highest and the lowest dimensional characteristics were found for Jahangiri and Nakhjavan cultivars, respectively. Sphericity values had significant difference among the tested cultivars. The latter property values were 0.971, 0.917, and 0.973 for Shams, Nakhjavan, and Jahangiri cultivars, respectively. Based on the results, the surface area of the Jahangiri cultivar was found to be 6458.35 mm(2), followed by the Shams and Nakhjavan cultivars, which had a mean of 6115.36, and 4395.25 mm(2), respectively. In this paper, the coefficients of static friction of the cultivars on three different surfaces are also reported. The results showed that mass modeling of apricot based on minor diameter and three projected areas are the most appropriate models in the first and the second classifications, respectively. In third classification, the best model was obtained on the basis of the actual volume as M = 0.997V(m) + 0301, R-2 = 0.98, R.S.E. = 1.711 with R-2 = 0.98. It was concluded that the suitable grading system of apricot mass is based on minor diameter as nonlinear relation: M = 0.0019C(2.693), R-2 = 0.96, R.S.E. = 2.2. (C) 2008 Elsevier B.V. All rights reserved,