營建材料價格變動屬於營建專案管理範疇中的重要風險，若能掌握材料價格的波動將有助於業者有效安排採購策略，而鋼筋為運用最廣泛的材料之一，且價格變動甚劇，然而其受多行業領域牽動，導致承包商難以準確預測價格變化情形，雖然目前有許多資料庫或趨勢分析能協助預估鋼筋價格走向，但是多有資料樣本少且更新不及時之問題，較不符合實際應用狀況。
近年因電腦運算能力進步及各式演算法的廣泛應用，加上分析處理大數據(Big Data)的技術進步，人工智慧技術成為各領域的熱門主題；其中，深度學習(Deep Learning)裡的卷積神經網路(Convolutional Neural Network)演算法能對於多維度的資料進行局部特徵提取，同時保留其空間性及連續性，許多研究亦證明其於圖形辨識上的表現更勝人為判斷，常運用於從歷史資料趨勢預測未來的風險；唯模型訓練需要大量資料樣本，且必須經過篩選整理，避免過多不相干資訊降低預測準確度。
綜合上述，鋼筋價格影響因素所牽涉範圍廣泛又充滿不確定性，而且營建業內外環境的事件發生亦可能引起價格波動，為了協助承包商執行更高準確度的預估，本研究應用卷積神經網路模型進行鋼筋價格漲跌幅之預測。首先解析價格影響因子至數值資料項目，再進一步考慮可能造成影響的國內外突發事件，並以相關關鍵字於Google的搜尋熱度代表無法量化之事件影響，將所蒐集之原始資料整理後轉為圖片，作為模型訓練用之資料集。訓練完成之模型可有效利用新輸入之影響因子及關鍵字熱度資料執行鋼筋價格漲跌之預測，最終實驗結果比較輸入不同事件關鍵字資料之預測模型準確率。

Steel rebar is one of the common materials in construction project, and also its price fluctuates a lot. The causes of the price fluctuation are from every stage of its production, transaction, delivering and so on, which is too complicated to monitor all of them by the general contractor. It is the reason why the contractor could not forecast rebar price accurately only depend on their experience.
The Artificial Intelligence technology has developed rapidly these years. Especially the Deep learning is popular in recognizing images. And the Convolutional neural network, which is one of the batches in Deep learning, is good at extracting features from multi-dimension data. It has been applied in risk prediction by several research.
The CNN prediction model for rebar price fluctuation is developed in this research. The impact factors and events of rebar price are analyzed. The factors are resolved as numerical data and the unpredictable events are connected to related keywords. All the chosen numerical data and the search trends data of the keywords from January 2007 to January 2020 are collected and entered to the prediction model for training. The prediction way is to forecast the percentage of rebar price fluctuation after six months with the data of the past whole year. Finally, the result shows the model training score of each model with 13 events separately and the new predictions with new data are executed.

英文文獻
[1]. A. G. Malanichev and P. V. Vorobyev, “Forecast of Global Steel Prices”, Studies on Russian Economic Development, Vol22: 304–311, 2011.
[2]. Alexander Pustov, Alexander Malanichev, Ilya Khobotilov, “Long-term iron ore price modeling: Marginal costs vs. incentive price”, Resources Policy Volume 38, Issue 4: Pages 558-567, 2013.
[3]. Ashwin Bhandare, Maithili Bhide, Pranav Gokhale, and Rohan Chandavarkar, “Applications of Convolutional Neural Networks”, IJCSIT, vol7: 2206-2215, 2016.
[4]. Chi-Wei Su, Kai-Hua Wang, Hsu-Ling Chang, Adelina Dumitrescu–Peculea, “Do iron ore price bubbles occur?”, Resources Policy Volume 53: Pages 340-346, 2017.
[5]. FIP, http://www.idef.com, 1993.
[6]. Hao Wu, and Jinsong Zhao, “Deep convolutional neural network model based chemical process fault diagnosis”, Computers and Chemical Engineering, vol115: 185-197, 2018.
[7]. Jian-Feng Guo, Qiang Ji, “How does market concern derived from the Internet affect oil prices?”, Applied Energy, Volume 112: 1536-1543, 2013.
[8]. Kaplan Andreas and Michael Haenlein, “Siri, Siri in my Hand, who's the Fairest in the Land? On the Interpretations, Illustrations and Implications of Artificial Intelligence”, Business Horizons, vol62:15-25, 2018.
[9]. Mohammad Reza Moghaddam, Masood Manfezi, Amir Hossein Mehr Danesh, and Gholamhassan Kakha, “Prediction of monthly price of iron ore by using artificial neural network”, Indian Journal of Scientific Research, vol1:1200-1204, 2014.
[10]. Neural Networks and Deep Learning, http://neuralnetworksanddeeplearning.com/ , Nielsen, Michael A , 2015.
[11]. Preeti Kulkarni1, Shreenivas Londhe, and Makarand Deo “Artificial Neural Networks for Construction Management: A Review”, Soft Computing in civil engineering, vol1:70-88, 2017.
[12]. Soo-Yong Shin, Dong-Woo Seo, Jisun An, Haewoon Kwak, Sung-Han Kim, Jin Gwack, “High correlation of Middle East respiratory syndrome spread with Google search and Twitter trends in Korea”, Scientific Reports volume 6, 2016.
[13]. Sumana Sharma, “An Integrated Knowledge Discovery and Data Mining Process Model”, Virginia Commonwealth University, 2008.
[14]. Sumana Sharma, Kweku-Muata Osei-Bryson, and George M. Kasper, “Evaluation of an integrated Knowledge Discovery and Data Mining process model”, Expert Systems with Applications, vol39: 11335-11348, 2012.
[15]. Trefor P.Williams and Jie Gong, “Predicting construction cost overruns using text mining, numerical data and ensemble classifiers”, Automation in Construction Vol43: Pages 23-29, 2014.
[16]. What’s the Difference Between Artificial Intelligence, Machine Learning, and Deep Learning?, https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/ , NVIDIA, 2016.
[17]. Yiqun Ma, Junhao Wang, “Co-movement between oil, gas, coal, and iron ore prices, the Australian dollar, and the Chinese RMB exchange rates: A copula approach”, Resources Policy Volume 63: 101471, 2019
[18]. Yun-Cheng Tsai, Jun-Hao Chen, and Jun-Jie Wang. “Predict Forex Trend via Convolutional Neural Networks”, Cornell University, 2018.
[19]. Yuyao Feng, Guowen Li, Xiaolei Sun, Jianping Li. “Forecasting the number of inbound tourists with Google Trends”, Procedia Computer Science Volume 162: Pages 628-633, 2019.
[20]. Zhishuo Liu, Yongcong Wang, Shuang Zhu, Baopeng Zhang and Lingyun Wei, “Steel Prices Index Prediction in China Based on BP Neural Network”, LISS 2014, vol1: 603-608, 2015.
 
中文文獻
[1]. 王勝榮，應用卷積神經網路於預測台灣鋼筋價格漲幅之研究，碩士論文，國立成功大學土木工程學研究所，2019。
[2]. 呂思葦，應用知識本體論及BIM於施工架之佈設和風險模擬，碩士論文，國立成功大學土木工程學研究所，2017。
[3]. 李倩瑜，應用基因演算法結合時間序列於台灣地區鋼鐵價格漲跌幅之預測，碩士論文，國立臺北科技大學工業工程與管理系碩士班，2014。
[4]. 林大貴，深度學習人工智慧實務應用，博碩文化，新北市，2017。
[5]. 林秀貞，國際油價波動對重要營建材料成本影響之研究－以鋼筋、水泥、砂石、瀝青為例，碩士論文，國立中央大學土木工程研究所，2007。
[6]. 財團法人經濟研究院.2019.’重要原物料國內外市場情勢分析及研究’
[7]. 許鈞甯，經濟指標與國際鋼鐵價格之關聯性－以中國、美國和日本為例，碩士論文，國立成功大學經營管理碩士學位學程， 2009。
[8]. 陳玫英，製程工廠統包工程專案管理與物料管理之探討，碩士論文，國立台灣科技大學營建工程研究所，2002。
[9]. 陳建任，台灣鋼鐵產業發展趨勢與鋼價展望，金屬工業研究發展中心，2011。
[10]. 陳建任，全球化競爭趨勢下，我國鋼鐵產業的主要挑戰與因應策略，金屬工業研究發展中心，2014。
[11]. 曾成訓，初探卷積神經網路，https://chtseng.wordpress.com/2017/09/12/初探卷積神經網路/，2017。