Statistical evaluation of quality parameters of olive stone to predict its heating value

•Olive stone energetic properties and variability of these parameters.•Climate and geographical variability.•Variations between olive stone quality parameters supplied by both olive-oil mills and distribution companies.•A correlation between the ultimate analysis and higher heating values of olive stone. In Andalusia, residual biomass produced in the olive sector results from the large amount of olive groves and olive oil manufacturers that generate byproducts with a potentially high energy content, suitable for thermal and electrical energy production. The main residue, olive stone, is an important solid biofuel and is widely generated and consumed. Consequently, olive stone quality parameters must be studied in order to achieve an optimum energetic efficiency. Therefore, the main objective of this study is to describe olive stone energetic properties and to evaluate variability of these parameters before consumption. For this purpose, mean values, normal distributions, intervals and deviations of these parameters have been obtained and studied. Concerning to statistical results, climate and geographical variability of quality parameters has been described. Furthermore, variations between olive stone physicochemical parameters supplied by both olive oil factories and distribution companies have been calculated. Finally, a correlation between the ultimate analysis and higher heating values (HHV) of olive stone has been determined. Results obtained show that olive stone pretreatments developed by distribution companies have a significant effect on quality parameters such as moisture content and low heating value. Moreover, olive stone properties dependence on factors such as rainfall or soil type has not been confirmed. Lastly, the calculated correlation based on ultimate analysis (i.e. HHV(MJ/kg)=0.401C−0.164H+0.493N+2.381S+0.791) has been developed and validated with olive stone samples with HHV range from 20 to 21MJ/kg (dry weight). Correlation has a mean absolute error (MAE) of 0.43% and a mean bias error (MBE) of −0.12% which indicate that it can be successfully used as a more economical and faster tool to accurately estimate olive stone HHV. The HHV prediction accuracies of 14 other correlations introduced by other researchers are also compared in this study.