Title: | 探討多人共享有限資源問題中個體的策略動態 The Individual Strategies Dynamics for Sharing The Limited Recourses Problem |
Authors: | 黃思綿 Szu-Mien Huang 孫春在 Chuen-Tsai Sun 資訊科學與工程研究所 |
Keywords: | 酒吧模型;個體行為;有限資源;個體策略;策略動態;Bar Model;Individual Behavior;Limited Recourses;Individual Strategy;Strategy Dynamic |
Issue Date: | 2007 |
Abstract: | 只要是符合,個體擁有主控權、週期性、多種選擇、有限資源與個體共享五個特徵的問題都屬於多人共享有限資源問題,而現實生活中有許多的問題都是屬於這種多人共享有限資源問題,例如澳門賭場增設問題、連鎖飲料店關店問題、遊樂場增設遊樂設施問題。而影響這類問題的主要因素就是個體所使用的策略,故本研究藉由觀察個體策略的動態來尋求解決這類問題的方法。
我們保留過去模型的優點,改善過去模型對於個體行為不合理的設定,以酒吧模型為基礎,設計了一個符合真實狀況的多人共享有限資源模型。可以提供給現實生活中遇到類似問題的管理者一個測試平台,各管理者可在本模型中模擬各種解決問題的方案,藉此尋找最適當的辦法在現實生活中進行推動。
最後,本研究就利用發展出來的模型,嘗試的解決現實生活中不同的問題。在不同的情況下,管理者可藉由提出各種方案或是過去5∼10 次的總消費人數來提高自身的效益。另外,也發現管理者在增加或減少分店時,必須掌握最佳時機,而透過本研究之模型也可以找何時為最佳時機。 When a situation has the following five characteristics, the individual has the command, the periodicity, the multiple options, the limited resources and the individual must share these resources, all belongs to the Multi- Agent Sharing Problem. In the realistic society, there are many situations which belong to the Multi-Agent Sharing Problem, for example the problems about Macao government increases or decreases the casinos, the amusement park additionally the facilities…etc. The strategy is the primary factor which affects these problems, therefore we observe these dynamic of strategies to discover some methods which can solve these Multi-Agent Sharing Problem. We build a Multi-Agent Sharing Model based on the Bar Model and it keeps the advantages of the past model. Besides, our new model improves the disadvantages about unadvisable setting of the individual behavior. This model provides a testing platform for some managers who face the above problem. These managers can use this new model to search for the best method to solve their problems and carry it out in real life. Finally, according to the new model, this paper finds some ways to deal with the real-life problems. In the different situations, the managers can provide many plans or the total amounts of consumer in the last 5~10 times to advance their income. Otherwise, managers must catch the best timing when they want to increase or decrease branches. This new model can help them to find the best timing. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT009555562 http://hdl.handle.net/11536/39514 |
Appears in Collections: | Thesis |
Files in This Item:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.