Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/27835
標題: 模糊多目標規劃方法應用於地區農業生產決策之研究
A Study on The Application of Fuzzy Multi-objective Programming Method to Regional Agricultural Production Planning Decision
作者: 蔣寬和
Chiang, Kuen-Ho
關鍵字: regional agriculture
地區農業
agricultural production decision
fuzzy multi-objective linear programming
fuzzy sets theory
農業生產決策
模糊多目標線性函數
模糊集理論
出版社: 應用經濟學系
摘要: 探討地區農業生產決策之主要目標在於尋求該地區資源有效配置利用,使地區農業永續經營發展。過去文獻大部分採用確定性和單目標模式來研究地區農業之生產規劃,因而忽略了農業生產環境是充滿不精確性及模糊性。本研究係從模糊環境觀點,來探討該地區如何規劃以獲得最適化農業生產決策之問題。 本研究依Zimmermann 和Hannan 理論為基礎,建構出模糊多目標線性規劃模式(FMOLP)及模糊多目標區段線性規劃模式(FMOPLP),此兩模式皆適用於解答地區農業生產決策之規劃問題。模式首先考慮到在同時滿足「利潤極大」與「工作天數極小」兩目標下,求得地區農業現況整體生產之滿意水準,此滿意水準可作為擬訂地區整體生產決策之參考指標;其次是在相互衝突目標式及限制式下,求得最適目標值及決策變數解,此決策變數可作為擬訂生產決策之依據;此外,本研究亦發展出一系統化之求解過程與步驟,供地區農業生產決策者實際運作之需要,決策者只要修改模糊資訊或參數,就可得到最適妥協解,使兩模式更具實用性。同時也可經由模擬分析,選擇出較佳策略,供農業決策者參考,使農業決策者能擬訂出較佳政策,有助於提高地區農業競爭力。因此與其他線性模式比較,本研究所建構之模式,更適合於地區農業生產決策之應用。 最後,按本研究FMOPLP模式求得地區最高滿意水準,係蓮農種植蓮花198公頃,休耕152公頃,此為該地區蓮農最適規劃生產決策。但現況調查蓮農未來五年內願種植面積僅103.7公頃而已,如何鼓勵蓮農增加種植面積來縮短些差距?端賴農政單位或農民團體協助解決蓮農種植蓮花所遭遇之相關問題,而本研究調查蓮農種植蓮花所遭遇之困難項目,依序為:「颱風」、「進口農產品威脅」、「肥料不足或太貴」、「販賣價格太低」、「種植勞力不足」、「缺水」等項目,這些問題中,除了「颱風」及「缺水」是自然因素外,其他問題應可透過農政單位或地區相關農民團體來協助解決?如此將有助於提高蓮農之種植意願。
The primary objective of this study is about the regional agricultural production decision making in order to efficiently allocate regional resources. In the past, scholars often use single criterion and crisp pre-assumptions to study this question. Unfortunately, real-world agricultural situations are much more complex. By using the theory of fuzzy environment, This paper is aimed at improve the efficiency of allocating regional agricultural resources. By using Zimmerman and Hannan's concepts, This work uses a fuzzy multi-objective linear programming (FMOLP) model and a fuzzy multi-objective piecewise linear programming (FMOPLP) model to solve the problem of regional agricultural production decision making. The proposed models attempt to efficiently allocate resources under the maximum profit and the minimum labor-day input. The proposed models can yield overall level of satisfaction of the region with goal value and decision variable value. The former can be used in index of the regional agricultural production planning, and the latter can be regarded as decision variable when making production decision. Furthermore, the proposed model leads to a systematic framework for facilitating the production decision process and enabling a decision maker to modify both the fuzzy data and related model parameters. Compared with other Linear Programming models, these models are more efficient. By using the lotus planters as an example, the major empirical result is that the optimal areas for lotus planting are 198 hectares in the Baihe area. The other 152 hectares should be Set-Aside Land. By field work, This paper also find that in the future five years, farmers are planning to plant 103.7 hectares of lotus. It seems like the agricultural decision makers should encourage farmers to plant more lotus. But,accounting to the field investigation,most farmers are worried about (1) typhoon , (2) the threat of foreign agricultural products , (3) the price and supply of fertilizer , (4) the price of lotus , (5) shortage of man powers , and (6) shortage of water . In order to meet the optimal model, the government should renew their policy for the benefits of the lotus farmers.
URI: http://hdl.handle.net/11455/27835
Appears in Collections:應用經濟學系

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.