Development of the 3-Dimensional Vehicle Loading Algorithm Considering Complex Logistics Environments

다국어 초록 (Multilingual Abstract)

Currently, the logistics industry is actively pursuing a transition to unmanned systems and automation based on the advancement of cutting-edge technologies such as AI and robotics. Amid the changing demographics and industrial landscape, this technological progress emphasizes the need for logistics automation. Automated cargo loading is expected to alleviate labor intensity and enhance productivity in the logistics and transportation sectors. This, in turn, is anticipated to yield benefits in terms of efficient space utilization, time savings, and economic gains. The focus of this study is to develop a 3-dimensional loading algorithm that can be applied to real-world logistics systems. To achieve this, we considered practical factors such as delivery sequences, height restrictions, packaging types, item vulnerability, and the spatial occupancy status of vehicles. The performance of the proposed algorithm was tested using practical numerical examples and assessed through the quality evaluation metrics introduced in this study.

목차 (Table of Contents)

• 1. Introduction . 1
• 1.1. Background . 1
• 1.2. Paper structure 3
• 2. Literature review 4
• 2.1. Research on container loading problem 4
• 2.2. Research on determining factors for loading quality 5
• 3. Problem description 9
• 4. Exact algorithm for 3-dimensional automated loading system 10
• 4.1. Preprocessing 10
• 4.2. Loading 12
• 4.2.1. Initial candidate points generation mechanism within the loading vehicle 13
• 4.2.2. Applying constraints . 17
• 4.2.3. Determination of the best loading coordinates for each cargo 18
• 4.2.4. Generate and adjust candidate points 19
• 4.3. Evaluation 23
• 4.3.1. Volume Utilization . 23
• 4.3.2. Dynamic Stability Metric (M2) 25
• 5. Experiment . 26
• 5.1. Data description 26
• 5.2. Scenario 26
• 5.3. 3-dimensional vehicle loading algorithm experiment results 29
• 5.3.1. Scenario simulation visualization results 29
• 5.3.2. Scenario simulation computation time results 30
• 5.4. Loading quality evaluation 31
• 5.4.1. Volume Utilization . 31
• 5.4.2. Dynamic Stability Metric (M2) 31
• 6. Concluding remarks . 35
• Reference 37