以螞蟻演算法探討NDF協商之最佳讓步態度組合

Using Ant Algorithm to Study the Optimal Concessions Under a NDF Negotiation Environment

陳奕豪(Yi-Hao Chen)

指導教授 ： 饒忻

中原大學/工學院/工業工程研究所/碩士(2004年)

https://doi.org/10.6840/cycu200400727

摘要

不同學者所提出的協商結果評估方法眾多，也沒有一個適切的評估手法來顯示出當協商發生時，參與協商者對於不同協商結果之偏好進行取捨來滿足買賣雙方的需求。本研究針對價格、數量、時間與品質以協商決策函式為基礎進行多議題的協商。在協商的過程中，買賣雙方讓步態度的改變將會產生不同的協商結果。在可行的協商範圍內，本研究利用螞蟻演算法探討在不同態度參數的組合下，輔以費落蒙的機制，買賣雙方以柏拉圖最佳化的距離與協商次數的多寡作為協商結果評估準則，進而尋找協商最佳解。依據買賣雙方對於協商結果偏好的不同，本研究以田口方法來決定螞蟻演算法求解問題時最佳參數的設定，藉由這些參數的設計來找出適合買賣雙方的提案值。相信本研究可提供往後研究類似問題時參考的方向。

關鍵字

螞蟻演算法 ； 多議題 ； 協商決策函式 ； 田口方法

並列摘要

Previous researchers proposed various methods to evaluate negotiation results; but, there is no appropriate method to determine whether the negotiation is the best to satisfy the requirements from both parties. This study uses Negotiation Decision Function as the negotiation mechanism for four issues: price , quantity , time and quality. During the negotiation process, different concession attitudes from both parties produce different negotiation results. This research uses the Ant algorithm to find the optimal negotiation result with the considerations of pheromone mechanism and concession attitudes for the performances of Pareto optimization distance and negotiation times. In addition, we use the Taguchi method to determine the best parameters used in the Ant algorithm for the negotiation with the difference preferences. We believe that this research can serve as a reference for future researches dealing with parameter design for the optimal negotiation.

並列關鍵字

Multiple issues ； Ant algorithm ； Negotiation Decision Function ； Taguchi Method

參考文獻

[OP02]Oprea, M., “An Adaptive Negotiation Model for Agent-Based Electronic Commerce,” Studies in Informatics and Control, Vol.11, No.3, pp.271-279, 2002

[BA94]Bazerman, M. H., Judgment in Managerial Decision Making, New York: John Wiley & Sons, Inc,1994

[BS96]Beam, C. and Segev, A., “Electronic Catalogs and Negotiations,” CITM 96-WP-1016, Fisher Center for Information Technology Management, University of California at Berkeley, August 1996

[BV99]Bui, H.H., Venkatesh, S., Kieronska, D.,“Learning other agents’ preferences in multi-agent negotiation using the Bayesian classifier,” International Journal of Cooperative Information System, 8(4), pp.275-294, 1999

[CA02]Chen, J.H., Anane, R., Chao, K.M., Godwin, N.,“Architecture of an Agent-Based Negotiation Mechanism,” Proceedings of the 22 nd International Conference on Distributed Computing System Workshops (ICDSW’02), 2002.