遞迴找尋方法於箱子尺寸設計之研究;A Study on the Recursive Search Method for Box Size Design

NCUIR > Engineering College > Graduate Institute of Mechanical Engineering > Electronic Thesis & Dissertation > Item 987654321/98610

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/98610

Title:	遞迴找尋方法於箱子尺寸設計之研究;A Study on the Recursive Search Method for Box Size Design
Authors:	楊詠年;Yang, Yong-Nian
Contributors:	機械工程學系
Keywords:	箱型設計;容器設計;空間利用率;空間最佳化;Box Sizing Problem;Box Size Selection;Space Utilization;Space Optimization
Date:	2025-06-27
Issue Date:	2025-10-17 13:00:05 (UTC+8)
Publisher:	國立中央大學
Abstract:	在包裝作業中，箱子的尺寸是否契合產品，對於空間利用效率具有關鍵性影響。若工廠生產100種產品，並僅使用5種箱子進行收納，則其選用之箱型尺寸需兼顧所有產品之包裝需求，盡可能降低收納的閒置空間。本研究針對此類「少量箱型對應多樣物品」的實務情境，提出一套箱型設計策略──遞迴找尋方法，以求解最佳的箱型尺寸組合。該方法透過迴圈機制逐一設計箱型，並可納入多項實務運作條件，包括重量限制、擺放方向限制、裝填數量規則及尺寸級距等。研究中針對不同收納條件進行實驗分析，歸納各項條件對利用率的影響，以使用5種箱型收納100種物品為例，模擬結果顯示整體空間利用率可達78.6%，若增加重量限制或擺放方向限制等條件，則導致空間利用率結果下降。研究亦納入簡化之經濟效益評估策略，模擬工廠在不同收納條件下所需面對的成本與效能權衡，作為方案選擇依據。此外，本研究亦導入Optuna參數優化模組，進一步強化遞迴方法之運算表現。在收納情境相同且計算時間相近之下，與傳統基因演算法相比，本方法可提升約6%的空間利用率。 ;In packaging operations, the compatibility between box dimensions and product sizes has a critical influence on the efficiency of space utilization. Taking a manufacturing plant as an example, if the factory produces 100 different types of products but utilizes only 5 types of boxes for storage and shipping, the dimensions of these boxes must be carefully designed to accommodate the full range of product sizes. The goal is to minimize unused space within each box while still meeting packaging requirements. This study focuses on the practical problem of using a small number of box types to accommodate a wide variety of items. To address this, we propose a box design strategy──the “recursive search method”──which aims to identifying the optimal combination of box sizes. The method designs each box type iteratively through a looping mechanism and is capable of incorporating various operational constraints, including weight limits, orientation restrictions, packing quantity rules, and dimensional intervals. A series of experiments was conducted under different packing conditions to analyze how these constraints affect utilization efficiency. In one case, using 5 box types to pack 100 products, the simulation showed that an overall space utilization rate of up to 78.6% could be achieved. However, the addition of constraints such as weight limits or orientation restrictions led to a noticeable decrease in utilization efficiency. To support practical decision-making, a simplified cost-benefit evaluation strategy was also developed, simulating the trade-offs between cost and efficiency under various packing conditions faced by real-world factories. In addition, this study integrates Optuna, a parameter optimization framework, to further enhance the computational performance of the proposed recursive search method. Under equivalent packing scenarios and comparable computational time, this method achieved approximately a 6% improvement in space utilization over conventional genetic algorithms.
Appears in Collections:	[Graduate Institute of Mechanical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	13	View/Open

社群 sharing

Loading...