Forecasting of Transportation cost for Logistics data

Aim of the paper is to analyze the issues and challenges faced by the transportation sector under supply chain. Optimization of the distance and reduction of transportation cost is one of the major challenge for the Retail Industries. The experiment aims at analyzing the minimum distance required for transporting the products from one or more warehouse to number of costumers. Experiment is based on the execution of Travelling salesman problem (TSP) using branch and bound and dynamic algorithms. Cost and time complexity are analyzed for varied number of costumers. This paper also predicts the Transportation cost using Regression Algorithm, which helps Retail sector to predict the cost incurred for logistics. Also Hypothesis for multiple regression coefficients are conducted to determine statistical evidence is sufficient for given parameters.

I. Introduction

Due to customer expectations, globalization and shorter lifecycle of products, supply chain risks are accelerating in terms of frequency and severity [1]. Transportation plays a major role in supply chain for distribution of goods in Business to Business (B2B) or Business to Customers (B2C). Transportation is the process of managing the movement of products from source to the end users. It requires technological innovative ideas to optimize the cost and transformation in the industry. Retail industry is facing challenge in handling the return goods. Order picking and delivery is one of the major part in supply chain [3], where labour force, cost of transportation and time need to be considered. Nowadays, many strategies are used in the warehouse beginning from picking, arranging in store wise and delivery to the consumer. In this paper, comparison of the TSP using Dynamic and Branch and bound method is done along with time complexity. Based on the output, regression analysis is applied to varied number of customers.