Machine Learning Models for Cost-Effective Shipping Line Selection: A Comparative Analysis for Freight Forwarders

نوع المستند : المقالة الأصلية

المؤلفون

1 College of International Transport and Logistics, Arab Academy for Science, Technology and Maritime Transport, Aswan, Egypt,

2 Information System Department Faculty of Computers & Information Menoufia University, Menoufia, Egypt.

3 Business Information Systems Department, College of Management and Technology, Arab Academy for Science, Technology, and Maritime Transport, Alexandria, Egypt,

المستخلص

The effectiveness of machine learning models in making cost-effective shipping line selections is investigated in this study from the perspective of freight forwarders. We had access to a dataset encompassing 983 shipment records from 37 different Egyptian freight forwarding companies. We then tested three different machine learning algorithms to see which one best predicted cost-effective shipping selections. The three algorithms were: Decision Trees, K-Nearest Neighbors (KNN), and Naive Bayes. After thoroughly testing these three algorithms, we determined that the best algorithm for use with our dataset, and the best one for use broadly within the market, was the Decision Tree method.
 

الكلمات الرئيسية

الموضوعات الرئيسية