Increasing load factor in logistics and evaluating shipment performance with machine learning methods - A case from the automotive industry

Aug 5, 2026

Upload by Rowan

Insufficient vehicle loading reduces logistics load factor and harms operational efficiency. This study uses real shipment data from an automotive company to improve logistics performance via a scenario-based approach. A dataset generated from the company ERP contains unlabeled data, enabling both unsupervised and supervised learning. Clustering groups shipments by similarity, then supervised models classify within each cluster. Cluster quality is assessed through average cost, while classification uses supervised performance metrics. Results show a rise from 25.7% high-performing shipments to 98.4% under scenarios, improving transport balance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages26

Reading Time1 h 6 min

AccessFree to download

Problem description

Empty transports and cost impacts

Current planning and operational challenges

Load factor definition and measurement issues

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Increasing load factor in logistics and evaluating shipment performance with machine learning methods: A case from the automotive industry Raziye Kılıç Sarıgül1􀀍, Burak Erkayman1,2 & Bilal Usanmaz3 The insufficient loading of vehicles, which leads to a low logistics load factor is a common problem in the logistics industry. This study addresses this issue by utilizing actual shipment data from an automotive company. An effective method has been proposed to improve the company’s logistics efficiency through a scenario-based approach. Two real-world scenarios were developed to enhance vehicle loading performance. Machine learning algorithms were employed to evaluate the shipment performance of these scenarios. For the study, a dataset was generated from the company’s ERP system and real-world shipment data. Since this is a real-world problem, the dataset consisted of unlabeled data. To solve this problem, both supervised and unsupervised learning algorithms were applied. First, unsupervised clustering algorithms were used to group the shipment performance based on similarities. Then, supervised learning algorithms were utilized to classify the data within each group. The average cost was used to evaluate the clusters obtained through the unsupervised methods, while classification performance was measured using supervised machine learning techniques. The scenario-based approach has significantly improved the performance of the shipments as it shows the changes in load factor more clearly. In the actual case, only %25.7 of shipments were high performing, while this percentage gradually increased to %98.4 in the scenarios. The results show that optimizing the load factor makes the transports more efficient and balanced. Keywords Logistics Load Factor, Logistics Vehicle Organization, Shipment Performance, Scenario-based approach, Supervised Learning, Unsupervised Learning In recent years, companies have tended to increase their operational efficiency by focusing on logistics processes in order to gain a competitive advantage. One of the key factors in the success of logistics process management, which brings companies closer to their targeted economic returns and provides opportunities for efficient resource utilization, is the feedback of performance evaluations into the system to keep this ecosystem under control. Performance measurements that support decision-making processes encompass a wide range of different approaches1. In connection with the performance of logistics processes, the load factor proves to bea frequently used criterion. When going deeper, it is evident that it is a critical parameter in the evaluation of freight transportation. Load factor can be simply defined as “the ratio of a vehicle’s carrying capacity to its actual usage”2,3. Along with the diversity of definitions, various terms such as ‘occupancy rate’, ‘vehicle utilization’,‘vehicle loading’, and ‘vehicle filling’ are observed to be used interchangeably with ‘load factor’ in the literature 4. The inadequate loading of vehicles is a difficulty that has a direct impact on operational performance. The measurement of performance emerges as an additional challenge due to the lack of standardization in calculating the load factor5. When calculating the maximum carrying capacity of the same vehicle, weight-based criteria for high-density goods and volume-based criteria for low-density goods can be used as sub-variables of the load factor6. This uncertainty makes the measurement of logistics performance more complex. To mitigate the mentioned challenges, more specific metrics such as “the level of empty running”, “the weight-based loading 1Department of Industrial Engineering, Faculty of Engineering, Ataturk University, Erzurum, Turkey. 2Department of Industrial Engineering and Business Information Systems, Faculty of Behavioral, Management and Social Sciences, Universit

FAQs

What problem does the study address in logistics operations?

The study addresses insufficient vehicle loading, which lowers the logistics load factor and degrades operational performance.

How are machine learning methods used to evaluate shipment performance?

Unsupervised clustering first groups shipment performance by similarity, and supervised learning then classifies data within each group.

What improvement does the scenario-based approach achieve in the automotive case?

High-performing shipments increase from 25.7% in the actual case to 98.4% under the developed scenarios, making transport more efficient and balanced.

Aug 5, 2026

Upload by Rowan