The researchers developed a method to classify different types of rice plants (Oryza sativa) into two main groups, indica and japonica, using physical characteristics. They used a statistical model called logistic regression and identified seven key physical traits that accurately separated the two groups. By applying this model to a sample of 280 rice plants, they achieved a classification accuracy of 100%. This means that the model can effectively distinguish between indica and japonica rice plants based on their physical features.