A stability study design where a subset of samples is tested at each time point, reducing testing burden while maintaining scientific justification.