A Search for Major Factors Influencing on Cut Rose Production Using Past Daily Average Data and Predicting Production Using Topological Case-based Modeling.

DOI Web Site 2 Citations 11 References
  • HOSHI Takehiko
    Department of Biologiccal Science and Technology, School of High-Technology of Human Welfare, Tokai University
  • SUZUKI Takafumi
    Department of Biologiccal Science and Technology, School of High-Technology of Human Welfare, Tokai University
  • UMEDA Atsushi
    Department of Biologiccal Science and Technology, School of High-Technology of Human Welfare, Tokai University
  • FUSE Junya
    Taiyo-Kogyo Co.
  • TSUTSUI Hiroaki
    Yamatake Co.

Bibliographic Information

Other Title
  • 過去の日平均データを用いたバラ切花生産に影響する主要因の探索と位相事例ベースモデリングを用いた生産量予測
  • カコ ノ ヒ ヘイキン データ オ モチイタ バラ キリバナ セイサン ニ エイキョウ スル シュヨウイン ノ タンサク ト イソウ ジレイ ベースモデリング オ モチイタ セイサンリョウ ヨソク

Search this article

Abstract

A plant production support system was developed to search for major factors influencing on plant production and to predict daily harvest. A case study on cut-rose production using rock -wool hydroponics was performed. Daily environmental measurement data and daily production records were recorded for 260 days on two kinds of splay roses and analyzed. The squared correlation coefficients between all the past averaged environmental data and the amount of daily production were calculated, and the coefficients were then visualized with 3-D graphs. Major factors having influence on the cut-rose production system were selected from the data with high correlation coefficients by step-wise multiple regression analysis and cluster analysis. In a testing greenhouse, four key factors were identified : the number of produced cut flowers from 1 to 2 days before the harvest day, the number of produced cut flowers from 37 to 55 days before the harvest day, the daily difference in inside air temperature from 27 to 49 days before harvest and the daily minimum inside air temperature from 50 to 85 days before harvest had influence on the number of produced cut flowers in the case of cultivar “Little Silver”. A prediction model was constructed by topological case-based modeling (TCBM) using the chosen factors. The average error in predicted production was ±15.4 cut flowers day-1 (±28.5%). In the case of cultivar “Fantasy”, the average error of the predicted production was ±14.7 cut flowers day-1 (±23.2%). The process of identifying the major factors might be a useful method to identify the characteristics of plant production in an individual greenhouse. The computerized prediction model of daily harvest would help growers to promote business and trade to their advantage.

Journal

  • Shokubutsu Kojo Gakkaishi

    Shokubutsu Kojo Gakkaishi 15 (1), 20-26, 2003

    JAPANESE SOCIETY OF AGRICULTURAL, BIOLOGICAL AND ENVIRONMENTAL ENGINEERS AND SCIENTISTS

Citations (2)*help

See more

References(11)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top