本論文主要探討在需求不確定下的百合花利潤最大化問題。與過去研究不同，在本研究的問題中所有百合花分送到不同市場之價格都為當日供需所決定。另外，由於傳統最佳化方法都以單一品種，採用機率分配、風險控管或平均數等方式來處理花農採購、種植安排與市場分送的決策問題。然而，這些方法中當品種不斷增加，常常造成使用者需耗費大量時間與精力去計算不同品種之機率分配，如果都採用平均的方法則不能夠有效產出百合花利潤最大化問題的可靠答案。

本研究改為使用穩健最佳化方法進行求解，將問題建構成兩階段穩健最佳化的模型，並使用限制式及變數產生演算法（column-and-constraint generation algorithm）進行求解。此舉不但解決獲取準確機率的困難度，也比較適合花農作業流程在第一階段負責採購與種植決策，第二階段解決分送市場問題，但此方法在第二階段計算未知價格時，採用過去歷史資料進行運算，所獲得解會趨於保守。因此，我們採用隨機森林(Random Forest)預測當日市場需求與價格，藉此調整市場分送的數量，以獲得更高的利潤。依據實驗結果，此方法可以輔助花農種植決策，並能夠比平均值和兩階段穩健最佳化方法獲得更好的獲利。

In this thesis, we investigate the problem of lily profit maximization under uncertain demand. Traditional optimization methods are based on a single variety and use a probability distribution, risk control, or the average way to deal with farmers' procurement, planting arrangements, and market distribution decision-making issues. However, when variety continually increases, the result is that the probability of our calculation is not very accurate. Moreover, assuming the use of the average method,which is the average purchase volume and the box of market distibution. We cannot produce a reliable solution to the lily profit maximization problem. Hence two-stage robust optimization approach is adopted to solve the problem,and a column-and-constraint generation (C&CG) algorithm is utilized to obtain a solution to the problem. This method does not assume a probability distribution, and is more suitable for the flower farming process. C&CG partitions the problem into a Master problem and a sub-problem. In the Master Problem that requires solving the procurement and planting decision-making, the sub-problem is solving the distribution of market problems. However, the price is unknown in the second stage of this method, as calculating it usually requires historical data. Thus, the solution at which we arrive will typically be rather conservative. Therefore, we use the Random Forest to predict the market demand and price and to adjust the number of market distribution to obtain higher profits.

Ahumada, O., & Villalobos, J. R. (2009). Application of planning models in the agri- food supply chain: A review. European journal of Operational research, 196(1), 1–20.
Ahumada, O., & Villalobos, J. R. (2011). Operational model for planning the harvest and distribution of perishable agricultural products. International Journal of Produc- tion Economics, 133(2), 677–687.
Allen, S. J., & Schuster, E. W. (2004). Controlling the risk for an agricultural harvest. Manufacturing & Service Operations Management, 6(3), 225–236.
Ben-Tal, A., & Nemirovski, A. (1998). Robust convex optimization. Mathematics of operations research, 23(4), 769–805.
Bernard, S., Heutte, L., & Adam, S. (2008). Forest-rk: A new random forest induction method. In International conference on intelligent computing (pp. 430–437).
Bohle, C., Maturana, S., & Vera, J. (2010). A robust optimization approach to wine grape harvesting scheduling. European Journal of Operational Research, 200(1), 245–252.
Breiman, L. (1996). Bagging predictors. Machine learning, 24(2), 123–140.
Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32.
Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A. (1984). Classification and regression trees. CRC press.
Caixeta-Filho, J. V., van Swaay-Neto, J. M., & Wagemaker, A. d. P. (2002). Opti- mization of the production planning and trade of lily flowers at jan de wit company. Interfaces, 32(1), 35–46.
Darby-Dowman, K., Barker, S., Audsley, E., & Parsons, D. (2000). A two-stage stochastic programming with recourse model for determining robust planting plans in horticulture. Journal of the Operational Research Society, 51(1), 83–89.
De Toro, A., Gunnarsson, C., Lundin, G., & Jonsson, N. (2012). Cereal harvesting– strategies and costs under variable weather conditions. Biosystems engineering, 111(4), 429–439.
Dietterich, T. G. (2000). An experimental comparison of three methods for construct- ing ensembles of decision trees: Bagging, boosting, and randomization. Machine learning, 40(2), 139–157.
Ekman, S. (2000). Tillage system selection: a mathematical programming model in- corporating weather variability. Journal of agricultural engineering research, 77(3), 267–276.
Ferna ́ndez-Delgado, M., Cernadas, E., Barro, S., & Amorim, D. (2014). Do we need hundreds of classifiers to solve real world classification problems. J. Mach. Learn. Res, 15(1), 3133–3181.
Ferrer, J.-C., Mac Cawley, A., Maturana, S., Toloza, S., & Vera, J. (2008). An optimiza- tion approach for scheduling wine grape harvest operations. International Journal of Production Economics, 112(2), 985–999.
Gigler, J., Hendrix, E. M., Heesen, R., Van Den Hazelkamp, V., & Meerdink, G. (2002). On optimisation of agri chains by dynamic programming. European Journal of Op- erational Research, 139(3), 613–625.
Jena, S. D., & Poggi, M. (2013). Harvest planning in the brazilian sugar cane industry via mixed integer programming. European Journal of Operational Research, 230(2), 374–384.
Jiao, Z., Higgins, A. J., & Prestwidge, D. B. (2005). An integrated statistical and optimisation approach to increasing sugar production within a mill region. Computers and electronics in agriculture, 48(2), 170–181.
Kusumastuti, R. D., van Donk, D. P., & Teunter, R. (2016). Crop-related harvesting and processing planning: a review. International Journal of Production Economics, 174, 76–92.
Oshiro, T. M., Perez, P. S., & Baranauskas, J. A. (2012). How many trees in a random forest? InInternationalworkshoponmachinelearninganddatamininginpattern recognition (pp. 154–168).
Tan, B., & C ̧o ̈mden, N. (2012). Agricultural planning of annual plants under demand, maturation, harvest, and yield risk. European Journal of Operational Research, 220(2), 539–549.
Villalobos, J. R., Mason, A., Flores, H., Wishon, C., & Ahumada, O. (2014). Planning, consolidation and coordination the keys for small farmers to advance in the fresh produce value chain.
Visagie, S. E., De Kock, H., & Ghebretsadik, A. (2004). Optimising an integrated crop-livestock farm using risk programming. ORiON, 20(1), 29–54.
Zeng, B., & Zhao, L. (2013). Solving two-stage robust optimization problems using a column-and-constraint generation method. Operations Research Letters, 41(5), 457– 461.
Zhang, W., & Wilhelm, W. E. (2011). Or/ms decision support models for the specialty crops industry: a literature review. Annals of Operations Research, 190(1), 131–148.
鍾國成. (2006). 台灣花卉產業發展現況與趨勢. 農業生技產業季刊, 5, 10–19.