Kiln-dried fruit drying time is readily predicted from initial moisture content since the environment is tightly controlled. For uncontrolled environments, such as a greenhouse solar dryer, a product's drying time varies depending on ambient conditions and is thus more difficult to predict. Prediction of the drying time is needed to better schedule dryer use. Data was obtained from a set of wireless scales that weigh the waste during solar drying after initial moisture content measurement of a sample. A set of linear and quadratic models for drying rate are tested with the best yielding a 39% reduction in RMSE over traditional models. The results indicate that the modelling approach is likely to be useful for open solar dryers where the temperature, and thus the drying rate, is not controlled.