Real-Time Cold Chain Forecasting Model for Agricultural Logistic Transportation using Feature Centric Deep Neural Classification for Seasonal Recommendation

Abstract

In recent days, cold chain logistic progression has been affected due to covid quarantine because real-time human resources being affected. Agriculture transportation and food safety are essential for human lives to avoid wastage of product. Analyzing the stock hold management needs more prediction accuracy in the seasonal recommendations for producing Agri-products. Increasing information and collaborative approaches in big data leads to more dimensions to analyze the prediction leads to inaccuracy for a recommendation. To improve the cold chain process, intend a Real-time Cold chain forecasting model for agricultural logistic transportation using feature centric deep neural classification for a seasonal recommendation. Initially, the preprocess was carried out to reduce the noise present in the seasonal collective and cold chain logistic dataset, which contains information about agriculture in stock detail, production, seasonal, and daily requirement ratio. The cold chain recommendation big data analytics estimate the seasonal productive margin factor (SPMF) and Stock hold production hit rate (SPHR) for feature logistic margins. Then selects the features using Intensive Agro feature successive rate (IAFSR) be grouped into clusters. Then the selected features are trained with Multi-objective Deep sub spectral neural network (MODS2NN) to categorize the needs of classes for recommendation. This cold chain process improves the prediction accuracy as well than other methods to recommendation the logistic stock hold management by right seasonal recommendation.