On the Study of the Heuristic Approach for the Dividable Volume Packing Problem

• Main Authors: Chia-Cheng Hsu, 許家禎
• Other Authors: Chiun-Chieh Hsu
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/7w75h3

碩士 === 國立臺灣科技大學 === 資訊管理系 === 106 === In recent years, with the development of information technology and the growth of e-commerce, many companies have invested in efficient and accurate intellective logistics management to improve the efficiency of their production and service. These made unmanned warehousing, unmanned selecting and unmanned drone important and popular for people to throw themselves into related research. This paper proposes a new packing problem called dividable volume packing problem (DVPP) which can be used in the application of storage and display of goods. The DVPP example used in this study is to lay out books into bookshelf for real situation. Heuristic construction is the basis for many other algorithms, and its property of being able to closely mimic actual circumstances and makes it suitable for real-life problems. We first define properties of the problem and use them to construct two heuristic methods. We also define properties of the problem and use them to construct two heuristic methods to make the constructed process more clearly. Further, the book encoding mechanism is added into the heuristic process to record the position of books, giving the heuristic additional functions when operating warehouse work such as sorting and distributing. Two heuristic methods have been constructed to solve the DVPP example problems. These methods are called Shelf First Locating (SFL), which is based on First-Fit strategy, and Layer First Locating (LFL), which are an extension of SFL. These methods consider the weights, heights and widths of books to improve the quality of solutions. To verify the usage performance of problem solving, a random number non-repetitive search method has been used to compare with the two heuristic methods. The result shows that the LFL heuristic method is the best among these methods for all experiments. The experimental results also show that the heuristic method can process and analyze a large amount of data than the search method.