A New Regression Model Based on an Extended Inverse Gaussian Distribution with Application to Soybean Processing Plants in Brazil

Abstract

Grain producers in Brazil often depend on third-party services for the transportation, processing and storage of their production, as, for the most part, they do not have silos on their properties. In this context, efficient logistics is essential to optimize processes and increase reliability between customers and service providers. This study focuses on the logistical analysis of truck traffic at two grain processing plants, examining different receiving protocols to evaluate internal vehicle flow during peak production conditions. The data is analyzed using a multiple regression model with two systematic components based on the proposed New Weibull inverse Gaussian distribution. The research is conducted in grain processing and storage units in the southwest region of São Paulo-SP, belonging to an agro-industrial cooperative. The study monitors all stages of soybean receipt during the peak harvest month, in March 2020. The results indicate the dependence of service times on the sector's logistical variables. This research addresses the pressing need for efficient logistics in the grain industry, especially in soybean processing. By focusing on truck traffic and receiving protocols, the study aims to provide a better understanding to optimize internal logistics processes, thus contributing to improving operational efficiency and customer service in grain processing units.