Title: Predictive models for rail-wagon detention in food grain logistics: a technological intervention
Authors: Nitish Vinod Sawant; Vinay V. Panicker; Anoop Kezhe Perumpadappu
Addresses: Department of Mechanical Engineering, National Institute of Technology Calicut, Kozhikode, Kerala, India ' Department of Mechanical Engineering, National Institute of Technology Calicut, Kozhikode, Kerala, India ' Department of Mechanical Engineering, National Institute of Technology Calicut, Kozhikode, Kerala, India
Abstract: This work deals with the movement of food grains in India undertaken by a food grain procurement and storage organisation. The movement is primarily achieved through the railway network, followed by the road network. The scope of the work is confined to the movement of food grains in Kerala region through railway network. This work applies machine learning algorithms to predict the occurrence of rail-wagon detention in the warehouses. Classification models are developed to predict the occurrence of detention at warehouses, and regression models are developed to predict the detention hours, based on the historical data. Popular algorithms used in this work are logistic regression, k-Nearest Neighbour, Naïve Bayes, decision tree, random forest, support vector machine and multiple linear regressions. Various performance parameters are used to evaluate the different models, and the best model is chosen for further prediction.
Keywords: food grain supply chain; machine learning algorithms; predictive analytics; random forest; logistic regression; business analytics.
DOI: 10.1504/IJVCM.2020.109238
International Journal of Value Chain Management, 2020 Vol.11 No.3, pp.250 - 272
Received: 07 Nov 2019
Accepted: 08 Dec 2019
Published online: 02 Sep 2020 *