An Insight into the Big Data-driven Logistics Industry using Clustering and Time-Series Analysis
Big data is one of the hottest trends in the supply chain and logistics industry currently. Every organization wants to get involved and reap the benefits of big data analytics. It enables a company to understand consumer trends and resolve challenges at strategic, operational, and tactical levels. This includes enhancing inventory management, reducing operational overhead, streamlining logistics, improving supply chain visibility, and improving service levels.
In general, companies use Excel/Tableau to carry out analyses on a day-to-day basis. This project aims at leveraging data towards improving the existing supply chains, by utilizing machine learning algorithms such as time series analysis and clustering algorithms alongside traditional data analysis techniques such as excel reports, line plots, and geographical maps.
A K-Prototype clustering algorithm has been implemented to identify the most likely shipment attributes at the Distribution Center level. Time-series analysis has been implemented to forecast the total number of shipments going out of each DC for the years 2020-2021. Later these forecasts are compared to the actual number of shipments to determine the impact of Covid-19 on the number of shipments. From a managerial point of view, this paper will provide an operating procedure for firms if they opt to go for a more detailed analysis than what they generally resort to.
Paper Advisor: Dr. Qiushi Chen, Assistant Professor of Industrial Engineering
|Work Title||An Insight into the Big Data-driven Logistics Industry using Clustering and Time-Series Analysis|
|License||In Copyright (Rights Reserved)|
|Work Type||Research Paper|
|Deposited||July 06, 2021|
This resource is currently not in any collection.