CROP YIELD PREDICTION BASED ON GROWTH AND ENVIRONMENTAL FACTORS: A CASE STUDY OF ORIENTAL MELONS (CUCUMIS MELO L. VAR. MAKUWA MAKINO) IN SEONGJU REGION

Abstract

In the Republic of Korea, Gyeongsangbuk-do accounts for 95% of the total oriental melon (Cucumis melo L. var. makuwa Makino) cultivation area and 91% of the yield. Recently, a smart farm system has been introduced that improves efficiency in farm management by collecting information on the growth and environmental factors for each facility farm. This study aimed to identify the factors that influence the increase in the crop yield of Seongju oriental melon through oriental melon information obtained from such a smart farm system, and to construct crop yield prediction models. The data used in the study was collected from four Seongju oriental melon facility farms measured from 2018 to 2019. To predict the crop yield of each farm per week, this study used principal component analysis, distributed-lag model, and auto-regressive integrated moving average with explanatory variables model (ARIMAX) based on growth and environmental variables. ARIMAX with principal components provided stable predictability. We expect the results of this study to be useful in developing optimal farming techniques that can improve the crop yield for farms cultivating Seongju oriental melon in facilities.