請用此 Handle URI 來引用此文件: http://hdl.handle.net/11455/25625
標題: 雞場管理決策支援系統
Management decision support system for chicken farms
作者: 王斌永
Wang, Bin-Yeong
關鍵字: 雞場
Chicken farms
決策支援系統
管理
Decision support system
Management
出版社: 動物科學系所
引用: 中央畜產會。2011。台灣地區雞蛋及產蛋雞統計資料。(未發表資料) 尹相志。2009。SQL Server 2008 Data Mining資料採礦─類神經網路。知悅文化。台北。pp.145-150。 王斌永、阮喜文。2005。台灣地區蛋雞場經營診斷資訊系統之建立。畜產研究 38(2): 117-124。 王斌永、阮喜文。2009。台灣蛋雞場經營管理知識庫之建立。畜產研究 42(1): 13-18。 王斌永、阮喜文、許振忠。1995。肉雞之採食量預測模式。畜產研究 28(4): 269-283。 王斌永、阮喜文、蕭庭訓、劉曉龍、胡見龍。2002。台灣地區蛋雞場經營成本與利益分析之試算程式。中畜會誌 31(3): 209-220。 行政院農業委員會。2010。農業年報99年版。農業統計年報。台北。 阮喜文、王斌永。1996。肉雞生長之電腦模擬模式。中畜會誌 25(1): 75-96。 阮喜文、胡見龍。1996。飼糧能量對白肉雞屠體性狀之影響。農林學報 45(4): 49-62。 吳美美。2004。數位學習現況與未來發展。圖書館學與資訊科學 30(2): 92-106。 呂禮佳。2009。推動CAS優良農產品證明標章-蛋品類執行成果。農政與農情 206: 47-49。 李淵百、江碧玲、黃暉煌。1997。台灣土雞最適上市週齡之研究。中畜會誌 26(3):285-296。 胡見龍。1995。肉雞屠體生長曲線之研究。碩士論文。國立中興大學。台中。 徐文杰、林沛傑。2003。數位學習標準與SCORM的發展。圖書館學與資訊科學 29(1): 15-28。 葉怡成。1997。應用類神經網路。儒林圖書公司。台北。 劉玉文。2010。99年我國糧食供需統計結果。農政與農情 232: 97-103。 賴碧瑩。2005。人工智慧電腦技術在不動產估價之應用—以類神經網路為例。財團法人台灣不動產資訊中心網站 http://www.ippi.org.tw/realestate/policy/ PolicyIndex_2.aspx?ID=2671。 魏健宏、楊雨青。1999。類神經網路選擇輸入變數之研究-以高雄港轉口貨櫃運量預測為例。運輸學刊 11(3):1-20。 Ahmad, H.A. 2009. Poultry growth modeling using neural networks and stimulated data. J. Appl. Poult. Res. 18: 440-446. Ahmadi, H. and A. Golian. 2008. Non-Linear Hyperbolastic Growth Models for Describing Growth Curve in Classical Strain of Broiler Chicken. Res. J. Biological Sci. 3(11):1300-1304. Ahmadi, H., M. Mottaghitalab, N. Nariman-Zadeh and A. Golian. 2008. Predicting performance of broiler chickens from dietary nutrients using group method of data handling-type neural networks. Brit. Poult. Sci. 49(3): 315-320. Behzadi, M. R. B. and A. A. Aslaminejad. 2010. A Comparison of Neural Network and Nonlinear Regression Predictions of Sheep Growth. J. Anim. and Veter. Adv. 9(16): 2128-2131. Burlacu, G., R. Burlacu, I. Columbeanu and G. Alexandru. 1990. Mathematical model for energy and protein balance simulation in broilers. Archi. of Anim. Nutr. 40: 409-422. Chamsaz, M., A. H. Perai, S. Asadpour and R. H. Shahidi. 2011. Comparison of the 3-phase segmented linear regression and artificial neural network models to predict broiler hatchability. J. Appl. Poult. Res. 20: 447-453. Chamsaz, M., S. Asadpour, A. H. Perai and A. Khatibjoo. 2009. Neural Network Model as a New Powerful Tool to Describe and Predict Hatchability of Broiler Breeder Eggs. J. Anim. and Veter. Adv. 8(7): 1349-1352. Emmans, G.C. 1981. A model of the growth and feed intake of ad-libitum fed animals, particularly poultry. Computers in Animal Production. Occasional Publication of the British Society of Animal Production. No. 5: 103-110. Emmans, G.C. 1984. Computer simulation in poultry production. In: 3rd European Symposium on Poultry Nutrition, Peebles, Scotland, UK. Emmans, G.C. 1995. Problems in modelling the growth of poultry. World’s Poult. Sci. J. 51: 77-89. Farooq, M., M.A. Mian and A. Asghar. 2001. Factors affecting cost of production and net profit per broiler in the subtropics [homepage on the Internet]. Livestock Research for Rural Development, 13(1) [cited 13 December 2002]. Available from URL: http://www.cipav.org.co/lrrd/lrrd13/1/faro131.htm. Garcia Pestana, A. 1972. Study of some of the technical factors influencing the production costs for one kilo of poultry meat. III. Influence of the weight and age of the chicken at slaughter and the contract system used. Revista de Nutr. Anim. 10(3): 177-198. Greig, I.D., J.B. Hardaker, D.J. Farrell and R.B. Cumming. 1977. Towards the determination of optimal systems of broiler production. Agricul. Syst. 2(1): 47-65. Grosskopf, K. and E. Matthaus. 1990. Mathematical simulation of the broiler production process. Systems Analysis Modelling Simulation 7(6): 473-491. Kaewtapee, C., C. Khetchaturat and C. Bunchasak. 2011. Comparison of growth models between artificial neural networks and nonlinear regression analysis in Cherry Valley ducks. J. Appl. Poult. Res. 20:421-428. Leeson, S. and J.D. Summers. 1980. Production and carcass characteristics of the broiler chicken. Poult. Sci. 59: 786-798. Minitab Inc. 2010. Minitab for Windows. Edit 16. Muramatsu, T. and S. Isariyodom. 1989. Computer-simulated growth prediction of replacement pullets with special reference to seasonal changes in feed intake. Poult. Sci. 68: 771-780. Nakamura, K., S. Tokumitsu and H. Tokunaga. 1988. Effect of diets with different protein and metabolizable energy values on the optimum market age of the broiler chicken. Japanese J. Poult. Sci. 25(5): 319-329. National Research Council (NRC). 1994. Nutrient Requirements of Poultry. 9th ed. National Academic Press, Washington, DC, USA. pp. 19-34. Park, D.B., Y.B. Choa and M. Lee. 2007. The use of an e-Learning system for agricultural extension: A case study of the rural development administration, Korea. J Agric. Educa. and Exten. 13(4): 273-785. Peebles, E.D., C.D. Zumwalt, P.D. Gerard, M.A. Latour and T.W. Smith. 2002. Market age live weight, carcass yield, and liver characteristics of broiler offspring from breeder hens fed diets differing in fat and energy contents. Poult. Sci. 81: 23-29. Roan, S.W. 1991. Bio-economic Models for the Simulation of the Production and Management of the Growing Pigs and Sows. PhD Thesis, University of Edinburgh, UK. Roush, W.B., W.A. Dozier III and S.L. Branton. 2006. Comparison of Gompertz and neural network models of broiler growth. Poult. Sci. 85: 794-797. Sarma, N.F., K.K. Baruah and N.N. Bora. 1986. Optimum market age of Khaki Campbell for meat production. Indian J. Poult. Sci. 21(4): 296-299. Savegnago R.P., B.N. Nunes, S.L. Caetano, A.S. Ferraudo, G.S. Schmidt, M.C. Ledur and D.P. Munari. 2011. Comparison of logistic and neural network models to fit to the egg production curve of White Leghorn hens. Poult Sci. 90(3): 705-11. Sossidou, E.N., D. Stamatis, R. Geers and E. Szucs. 2007. Development of quality indicators in poultry welfare, environment and food quality interaction studies through upgraded e-learning. World Poul. Sci. J. 63: 413-419. Supaporn, I., I. Tasaki and J. Okumura. 1988. Construction of a mathematical model for predicting broiler performance. Japanese J. Poult. Sci. 25: 191-200. Talpaz, H. and S. Hurwitz. 1988. Economic optimization of a growth trajectory for broilers. Am. J. Agr. Econ. 70 (2): 382-390. Talpaz, H., J.R. de la Torre, P.J.H. Sharpe and S. Hurwitz. 1986. Dynamic optimization model for feeding of broilers. Agri. Syst. 20: 121-132. Velu, J.G, D.H. Baker and H.M. Scott. 1972. Regression equations for determining body composition of young chicks. Poult. Sci. 51: 698-699. Ward Systems Group. 2010. NeuralShell predictor. Wolynetz, M.S. and I.R. Sibbald. 1986a. Prediction of major body components of broiler chicks from a small subset. Poult. Sci. 65: 2167-2172. Wolynetz, M.S. and I.R. Sibbald. 1986b. Prediction of major body components of broiler chicks. Poult. Sci. 65: 2173-2185. Wolynetz, M.S. and I.R. Sibbald. 1986c. Relationships among major body components of broiler chicks. Poult. Sci. 65: 2324-2329. Wolynetz, M.S. and I.R. Sibbald. 1990. Estimates of body components in broiler chickens from body weight and dry matter. Poult. Sci. 69: 1318-1324. Zoons, J., J. Buyse and E. Decuypere. 1991. Mathematical models in broiler raising. World''s Poult. Sci. J. 47: 243-255.
摘要: 隨著社會經濟的發展、科學技術的進步、貿易的自由化以及人民生活水準不斷提高,國人以雞肉及雞蛋作為日常飲食的選擇,成為一種普遍的現象。就飼養及經營管理技術面而言,台灣的畜牧生產業者已擁有充分的資訊提供技術支援。然而,隨著資訊科技的快速進步,現在有愈來愈多可以輔助應用在各種生產技術的調節及決策支援上的系統程式被開發,不僅可大幅減少諮詢專家的時間,亦可做到隨時隨地試算及上網學習的目的,值得進一步推廣應用。 I.以迴歸模式及類神經網路預估台灣雞蛋產量 本研究之目的在於透過傳統迴歸模式與類神經網路(ANN)比較台灣地區新母雞、產蛋雞、淘汰雞及換羽雞隻數量與產蛋箱數間之關係,用以預估雞蛋產量。透過這些變數的組合,模擬出雞蛋產量的變化趨勢,此結果可以提供給政府單位作為決策與提升產業競爭力之參考。根據中央畜產會之提供資料 (2011),收集自2001年1月至2011年3月份逐月之產蛋箱數 (萬箱),產蛋隻數 (萬隻)及總蛋雛數 (萬隻),共計123筆資料。最終所得迴歸方程式為:(1) 傳統迴歸模式: case = 2.84 + 0.0115 pmonth - 0.00783 pmonth2 - 0.00199 chick + 0.00241 laying, R2 = 0.826; (2) 類神經網路模式: case = 2.83 + 0.115 pmonth - 0.00782 pmonth2 - 0.00198 chick + 0.00241 laying, R2 = 0.999其中pmonth代表雛雞月份,即欲預估月份之前5個月;chick代表該雛雞月份之總蛋雛數;laying代表總產蛋隻數;case代表總產蛋箱數。此結果顯示類神經網路用於預估台灣地區雞蛋產量上較傳統迴歸模式來得精確。 II.POMA-BROILER: 預估肉雞最適上市日齡之電腦模擬模式 本研究以電腦模擬建立之POMA-肉雞模式評估肉雞的最適上市日齡。本模式係以Visual Basic程式語言編寫及使用Windows作業系統完成。本模式採用敏感性分析的方法,基於邊際成本不得超過邊際收益的概念。使用不同的輸入數據包括:飼料成分 (粗蛋白 (CP),代謝能 (ME)和飼養階段)、飼料轉肉率方程式、成本條件 (雛雞、飼料、勞務費、水電費、醫療、折舊及雜項費用等)及生長迴歸方程式。進一步比較本模式的計算結果與市場可接受的重量範圍內之邊際成本和邊際收益,進而逐日計算以確定最大利潤的決策點。這些結果可使用在肉雞生產和管理策略之調整,應具有參考價值。 III.蛋雞場經營數位學習系統之建立 本研究係結合行政院農業委員會畜產試驗所已開發的蛋雞場經營管理相關電腦系統,包括成本效益分析試算、管理診斷諮詢及經營管理知識庫等,另增加其他專業知識,建立線上數位學習系統。本系統之課程內容包含品種介紹、飼養管理、營養、疾病防疫、環保及效益評估等單元。課程分別以影片簡報及動畫互動兩種方式呈現,其中蒼蠅防除及自衛防疫兩項課程為互動式FLASH系統,可提高學習者之興趣,並提供不受人數及時間、空間限制的環境,充實並增進經營管理相關知識,提升產業發展潛能。本系統已透過家禽飼養管理訓練班介紹給農友,擴大推廣運用。
With the social and economic development, scientific and technological progress, trade liberalization and rising living standards, people with chicken and eggs as the choice of diet has become a common phenomenon. Breeding and management techniques on the surface, Taiwan''s livestock industry has sufficient information to provide technical support. However, with the rapid advances in information technology, there are more and more applications can assist in the regulation of various production techniques and decision support systems on the program were developed, not only can significantly reduce the time consultants, but also can be done at anytime and anywhere to use spreadsheet to reach the purpose of online learning. It is worthy of further application. I.Comparison of regression and artificial neural network models of egg production This study compared the relationship between egg production and the number of pullets, laying hens, culling birds and molting birds in Taiwan through traditional regression methods and ANN (Artificial neural network) models. Egg production data and the number of laying hens associated with each data set were gathered from the National Animal Industry Foundation for dates between January 2001 and March 2011, totalling 123 data sets. The final regression equations were: (1) Traditional regression model: case = 2.84 + 0.0115 pmonth - 0.00783 pmonth2 - 0.00199 chick + 0.00241 laying, R2 = 0.826; (2) ANN model: case = 2.83 + 0.115 pmonth - 0.00782 pmonth2 - 0.00198 chick + 0.00241 laying, R2 = 0.999. These results show that the ANN method is more accurate than traditional regression models for predicting egg production in Taiwan. II.POMA-BROILER: The Software of computer simulation model to predict optimal market age of broilers This study presents the POMA-BROILER model, a computer simulation developed to evaluate the optimal market age of broilers. This model was written in the Visual BASIC programming language and uses the Windows operating system. The model was developed from a sensitivity analysis method and is based on the concept that marginal cost must not exceed marginal return. It uses various input data, including feed information (crude protein (CP), metabolizable energy (ME), and feeding stages), an equation for the feed conversion ratio, cost conditions (chicks, feed, labour, water and power, medical treatment, depreciation and miscellaneous costs) and a growth regression equation. The model then compares the calculated results with the range for acceptable market weight. The marginal cost and marginal return are thus calculated every day to determine the decision-making point for maximum profit. These results could represent a valuable reference for use in adjusting the strategy for broiler production and management. III.Establishment of the digital learning system on management of layer farm The purpose of this study was to combine several computer systems on layer farm developed by Livestock Research Institute, Council of Agriculture, Executive Yuan. They include cost-efficiency analysis spreadsheets, management diagnosis and management knowledge base, and introduced other professional knowledge to build the online digital learning system on management of layer farm. The courses of this system comprise breed, feeding and management, nutrition, epidemic prevention, environmental protection and efficiency evaluation. These courses presented by briefing video and interacting animation, individually. Especially, the flies prevention and self-defense courses are FLASH system that could attract the learner''s interest and provide a learning environment not limited by participants, time and location. It assists participants with full of knowledge about management of layer farm, and promotes poultry industry to reach developing potential. The system has been introduced through the management training program to poultry farmers and therefore expands its extension and application.
URI: http://hdl.handle.net/11455/25625
其他識別: U0005-1607201214272600
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-1607201214272600
顯示於類別:動物科學系

文件中的檔案:
檔案 大小格式 
nchu-101-8098037002-1.pdf6.26 MBAdobe PDF檢視/開啟


在 DSpace 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。