Baidu uses the deep-learning platform PaddlePaddle and its click-through rate (CTR) model to predict the purchased quantity of fresh goods in convenience stores. This technology can effectively reduce food waste.
The click-through rate (CTR) model could predict the purchased quantity based on the sales data that includes weather, location, festivals, et al. from hundreds of convenience stores in the past one to two years. According to Robin Li, Baidu's founder and CEO said, "In the testing stores, their profits have increased by 20%, and food waste has dropped 30%. That means we can avoid wasting food thanks to AI technology."
The convenience stores could determine the purchased quantity based on the click-through rate (CTR) model, and the appropriate purchase quantity will increase the profits as well as reducing the food waste.