請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/77915
標題: | 旅遊排程規劃問題研究:以台北地區旅遊排程規劃為例 Trip Scheduling Research: A Case Study of Taipei Trip Scheduling |
作者: | Chia-Fu Liu 劉家福 |
指導教授: | 陳正剛(Argon Chen) |
關鍵字: | 旅遊排程規劃,動態規劃,啟發式演算法, Trip scheduling problem,Dynamic programming,Heuristic algorithm, |
出版年 : | 2017 |
學位: | 碩士 |
摘要: | 身處於一個資訊爆炸的年代,現代人每天都能輕易接觸大量的資訊。尤其近年來,智慧型裝置日漸普及,舉凡智慧型手機、平板電腦和穿帶式裝置等,在許多現代化城市中,幾乎已達到『人手一機』的普及率。藉由這些智慧型裝置,使用者能非常容易地取得網路上各樣資訊、線上新聞、媒體廣告等等。但是面對這些大量資訊,使用者該如何判斷哪些是對自己有用或正確的資訊?並且從個樣資訊中,整合出真正對自己有『價值』的資訊?
以『旅遊規劃』為例:當規劃一趟旅程時,需要考量許多旅遊相關的資訊,例如:景點間的交通方式和其估算的移動時間、各個景點的花費時間等。使用者固然可以自行蒐集、整理各項所需資料,但此方法是一項高時間成本的工作、並且不見得有效率。因此,本研究欲建立一套『旅遊排程規劃系統』,此系統可建構於智慧型裝置上,藉由無線網路取得各樣旅遊所需之資訊,並考量使用者對各景點的喜好程度、使用者目前所在的位置、旅程開始時間和結束時間,自動規劃出一套完整的旅遊方案,其內容包括景點拜訪的順序和停留時間、每一段旅程建議搭乘的交通工具和其搭乘時間點。此旅遊排程問題首先被制定為一最佳化問題模型,首先要在旅程中盡可能排進更多使用者希望去的景點,接著是盡可能在比較偏好的景點停留較長的時間,本研究根據所提出排程規劃模型的特性,以動態規劃、派工法則、動態規劃啟發式演算法進行規劃求解,並將所提出的演算法利用 Python 語言寫成程式。 研究最後以台北地區一日遊—旅遊排程規劃為例,藉由收集在台北旅遊相關的資料,包括交通移動方式、各景點的開放時間和建議停留時間等,以實際的資料驗證所提出的排程規劃演算法及 Python 程式;並在給定一旅遊情境下,以上述所提出不同的規劃求解法進行排程規劃,並對各演算法結果進行分析和比較。 Living in the age of information explosion, people can access to massive information in their daily life. Recently, mobile smart devices, such as mobile phone, tablet PC and wearable devices have become round-the-clock companion to almost everyone. By using these smart devices, users can obtain easily all sorts of real-time information, updated news, even commercials. Although there is a large amount of available information, users still have to figure out how to find the correct and useful information that they really need. Take “trip planning” as an example. When planning a trip, users need to take a lot of travel information into account, e.g. means of transports between different attractions and their traffic time, recommended stay time in each attraction, etc. Though users could always collect and organize the data themselves, it would be time-consuming and difficult to attend to every aspect simultaneously. Based on above reason, this research aims to build a “trip scheduling system” to be embedded in smart devices. This system uses accessible tourism information from communication network to generate a trip plan by considering user’s information including user’s preference for attractions, user’s current location, and the start time and the end time of the journey. More specifically, this generated trip plan will recommend the visited order and stay time of the attractions, means of transports between each attractions and their scheduled times. The problem is formulated as an optimization model to optimize two sequential objectives: first to schedule as many preferred attractions as possible in the trip and second to stay in the most preferred attractions as long as possible. This research then develops dynamic programming methods, dispatching rules and heuristic dynamic programming methods to solve the problem. The developed algorithms are then coded and implemented as a computer program using Python. Finally, we take the problem of “one-day trip in Taipei” as a case study to validate the proposed algorithms and the Python computer program. We collect the information about tourism in Taipei, such as available transportation, the opening hours and recommended stay time of the attractions to construct the optimization model. Under given scenarios, we solve the problem using proposed algorithms and then conduct analyses to compare different algorithms under different scenario settings. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/77915 |
DOI: | 10.6342/NTU201702991 |
全文授權: | 有償授權 |
顯示於系所單位: | 工業工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-106-R04546026-1.pdf 目前未授權公開取用 | 11.06 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。