Increasing load factor in logistics and evaluating shipment performance with machine learning methods: A case from...

Area of Science:

• Logistics and Supply Chain Management
• Operations Research
• Data Science

Background:

• Low logistics load factors are a persistent challenge in the transportation industry.
• Insufficient vehicle loading reduces overall efficiency and increases costs.
• Actual shipment data from an automotive company was used to address this problem.

Purpose of the Study:

• To propose an effective method for improving logistics efficiency through a scenario-based approach.
• To enhance vehicle loading performance using machine learning algorithms.
• To analyze the impact of optimized load factors on transport efficiency.

Main Methods:

• A scenario-based approach with two real-world scenarios was developed.
• Machine learning algorithms, including unsupervised clustering and supervised classification, were applied to unlabeled shipment data.
• Shipment performance was evaluated using average cost for clustering and classification metrics for supervised learning.

Main Results:

• The scenario-based approach clearly demonstrated improvements in load factor.
• High-performing shipments increased from 25.7% in the actual case to 98.4% in the developed scenarios.
• Machine learning effectively grouped and classified shipment performance.

Conclusions:

• Optimizing the logistics load factor leads to more efficient and balanced transports.
• The proposed scenario-based method significantly enhances vehicle loading performance.
• Data-driven approaches using machine learning are effective for solving real-world logistics problems.