以深度學習為基礎之貨物裝載排程最佳化;A Novel Deep Neural Network-Based Cargo Loading Optimization

NCUIR > School of Management at National Central University > Executive Master of Information Management > Electronic Thesis & Dissertation > Item 987654321/93130

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/93130

Title:	以深度學習為基礎之貨物裝載排程最佳化;A Novel Deep Neural Network-Based Cargo Loading Optimization
Authors:	洪婷瑮;Hung, Ting-Li
Contributors:	資訊管理學系在職專班
Keywords:	Self-Attention;Machine Learning;Deep Neural Network;Bin Packing Problem;Air Cargo Loading;自注意力機制;機器學習;深度神經網路;裝箱問題;航空貨運裝載
Date:	2023-07-03
Issue Date:	2024-09-19 16:43:53 (UTC+8)
Publisher:	國立中央大學
Abstract:	本篇碩士論文主要目的在於探討如何應用深度學習技術有效地解決航空貨運裝載優化問題。航空貨運裝載優化在降低運營成本和提高市場競爭力上扮演著關鍵的角色。然而，由於貨物的多樣性，如尺寸、重量、和特定需求等，使得裝載優化問題展現出高維度和高複雜度的特性，這也正是本研究所需面對的挑戰。為了解決這個問題，本研究提出了一種基於Self-Attention機制的深度神經網絡模型。這種模型具有有效地找出最佳或接近最佳的貨物裝載策略的能力，旨在在有限的機艙空間中最大化裝載貨物的數量，同時確保運輸過程的安全和可靠。本文首先對航空貨運裝載優化問題及其在商業和科學研究中的重要性進行了深入的分析和探討。緊接著，我們對深度學習技術進行了詳盡的介紹，涵蓋了深度神經網絡、Self-Attention機制等主題。然後，我們提出了一種基於Self-Attention機制的深度神經網絡模型，並對其進行了詳細的數學分析和模型設計。我們提供了該模型的理論基礎，並解釋了如何根據問題的具體需求和條件來調整模型參數。最後，我們透過一系列的實驗來驗證所提出模型在航空貨運裝載優化問題上的性能。實驗結果證明了我們的模型不僅有效，而且在不同的裝載情況和條件下都能提供穩定可靠的解決方案。這些結果讓我們相信，基於Self-Attention機制的深度神經網絡模型在處理航空貨運裝載優化問題上具有廣泛的應用前景。 ;The primary aim of this Master′s thesis is to investigate how to apply deep learning techniques effectively to solve the air cargo loading optimization problem. Air cargo loading optimization plays a key role in reducing operating costs and enhancing market competitiveness. However, due to the diversity of the cargo, such as size, weight, and specific requirements, the loading optimization problem demonstrates high-dimensionality and complexity, which are the challenges this research has to face. To solve this problem, this research proposes a deep neural network model based on the Self-Attention mechanism. This model has the capability to effectively identify the best or near-optimal cargo loading strategy, aiming to maximize the number of cargo loaded in the limited cabin space, while ensuring the safety and reliability of the transportation process. This paper first provides an in-depth analysis and discussion of the air cargo loading optimization problem and its significance in both business and scientific research. Following this, we give a comprehensive introduction to deep learning technologies, covering topics such as deep neural networks and the Self-Attention mechanism. Then, we propose a deep neural network model based on the Self-Attention mechanism and provide a detailed mathematical analysis and model design. We offer the theoretical foundation of this model and explain how to adjust the model parameters according to the specific demands and conditions of the problem. Lastly, we verify the performance of the proposed model in the air cargo loading optimization problem through a series of experiments. The experimental results prove that our model is not only effective but also can provide stable and reliable solutions under different loading conditions and circumstances. These outcomes make us believe that the deep neural network model based on the Self-Attention mechanism has broad application prospects in dealing with the air cargo loading optimization problem.
Appears in Collections:	[Executive Master of Information Management] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	16	View/Open

社群 sharing

Loading...