AbstractIn this research, the effects of different dryers including refractance‐window (RW) with 200 or 300 μm Mylar membrane, or Pyrex glass and oven drying, and various temperatures of 25, 60, 70, and 80 °C on saffron quality were surveyed. Levels of safranal, crocin, and picrocrocin for saffron stigma and anthocyanin content for saffron petal were considered as the major quality indices and modeling of their drying kinetics was carried out. While for saffron stigma, RW dryer through Pyrex glass surface and higher temperatures (70 and 80 °C) led to the highest contents of picrocrocin, safranal, crocin, for saffron petal, RW dryer through 200 μm Mylar membrane resulted in the maximum anthocyanin content (106.79 mg/L). Meanwhile, our comparison revealed that RW is a more reliable approach to dry saffron stigma while keeping its functional ingredients, that is, coloring, aromatic, and bitterness strength, than previously introduced novel approaches. The best Artificial Neural Networks (ANNs) topologies to predict RW (200 or 300 μm Mylar membrane, or Pyrex glass) or oven drying data for saffron stigma were those networks having feed‐forward backpropagation network type, hyperbolic tangent sigmoid transfer function and one hidden layer with 3, 7, 5, or 4 neurons in it, respectively.Practical applicationsSaffron is the dried stigmas of flowers called as “Crocus sativus L.” scientifically and “red gold” in the market because of its high value. Our aim of this research was to introduce a suitable alternative for common traditional drying of saffron as refractance‐Window (RW) and suggest appropriate modeling patterns for industrial application of this drying type in real situations. The second aim of this research was to determine the best drying conditions of saffron petal since the major weight of saffron flower is related to its petal, and anthocyanin of petal extracts could be used as a natural colorant in the food industry, but temperature has a substantial impact on anthocyanin stability.