Design a multi-scale fuzzy sampling model for the quality inspection of massive ocean data

The conventional theory of sampling inspection is not suitable for the quality inspection of ocean data due to its specific characters such as magnanimity, multi-source, multi-dimensions, various types, and obvious spatiality. Thus, in this paper, we proposed a novel multi-scale fuzzy sampling model based on fuzzy mathematics. Given their magnanimity, multi-dimensions, and various types, the ocean data were first divided into lots, and then the inspected item in each lot was defined. Next, the sampling parameters were defined as fuzzy numbers, and the sampling inspection model was designed based on fuzzy mathematics. Finally, the proposed model was implemented to conduct the quality inspection of the ocean data originated from a given sea area in China. The results show that the proposed sampling model is more suitable for the quality inspection of massive ocean data characterized by multi-dimension, uncertain quality characteristics, non-uniform items than the conventional sampling inspection.