2020-06-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/690573摘要：本研究計畫針對鳳梨產量與品質，收集關鍵環境與作物參數，結合統計分析、資料科學、機器學習等新近技術，建立產量與品質預測模式，預計未來應用在農業保險之風險評估。計畫執行方式，是在南部實際生產的鳳梨田區與合作的試驗場所建立監控設備，收集田間環境與與鳳梨品質及產量資料，接著透過適當分析方法找出影響鳳梨品質與產量的關鍵環境參數，並結合鳳梨生理機制相關知識，建立鳳梨生育期與產量模擬模組。依據累積收集的田間關鍵參數及作物生理參數，得以開發適用於鳳梨的產量與果品風險計算模型，並製作鳳梨主產地產量風險地圖。期待本計畫的研究成果，有助於規劃合理之鳳梨農業保險機制，降低農民因農產物遭受天然災害所產生之損失及維持穩定收益。<br> Abstract: This project aims at the development of predictive models for pineapple yield and quality. The models correlate the yield and quality to relevant environmental and crop parameters by recent statistical analysis, data science, and machine learning techniques, which are expected to be applied to agricultural insurance risk assessment in the future. We will set up IoT sensors at the test sites in the pineapple fields and/or cooperatives to collect environmental and phenotypic data. Appropriate methods will be applied to identify environmental factors that are critical to pineapple quality and yield. We will also investigate relevant plant physiological mechanisms to establish simulating modules for pineapple phenology and yield. After accumulating enough field data, we will be able to tackle the model construction for pineapple yield/quality risk assessment. The risk map for major pineapple-production counties will be demonstrated accordingly. We hope that the results of this project will help to plan a reasonable pineapple agricultural insurance plans to compensate farmers` losses due to the natural disasters with stable incomes.鳳梨產量預測品質預測作物模式農業保險pineappleyield predictionquality predictioncrop modelcrop insurance