Modeling Lettuce Growth in Vertical Farms: Approaches and Strategies

Bohyun Sung; Kyungryoon Oh; Geunhyuk Ko; Hyejin Kim; Young-Yeol Cho

doi:10.7235/HORT.20250084

Online First

Preview Page Next Page

Research Article

Modeling Lettuce Growth in Vertical Farms: Approaches and Strategies 수직농장에서 상추 생육 모델링: 접근법과 전략

PDF XML

Abstract

Vertical farming is an advanced agricultural system that provides precise environmental control to promote uniform crop growth and enables planned year-round production. This study investigated lettuce growth in a fully controlled vertical farm over 31 days, measuring shoot fresh and dry weights to develop growth prediction models. Several mathematical models, in this case linear, quadratic, exponential, sigmoid, and expolinear functions, were evaluated for their suitability. Linear and quadratic models showed limited accuracy and practical limitations, while the exponential model did not account for the growth deceleration observed in lettuce, making it unsuitable. The sigmoid and expolinear models both effectively captured the biological growth pattern characterized by slow initial growth, a rapid mid-stage increase, and a late-stage plateau. Notably, the expolinear model demonstrated the highest predictive accuracy, with coefficients of determination exceeding 0.99 in the fitting data and 0.93 in validation data during the marketable growth period (up to 28 days after transplanting). Beyond this period, excessive leaf growth reduced market value, emphasizing the need for growth modeling focused on commercial quality maintenance. These findings highlight the expolinear model as a reliable tool for predicting lettuce growth and optimizing harvest timing in vertical farms.

Keywords

expolinear function

growth modeling

lettuce

predictive modeling

vertical farming

수직농장은 정밀한 환경 제어를 통해 균일한 작물 생장을 촉진하고 연중 계획된 생산을 가능하게 하는 첨단 농업 시스템이다. 본 연구에서는 완전제어형 수직농장에서 상추를 31일간 재배하며 생체중과 건물중을 측정하여 생육 예측 모델을 개발하였다. 선형, 2차, 지수, 시그모이드, 선형지수 함수 모델을 평가한 결과, 선형 및 2차 모델은 제한된 정확도를 보였고, 지수 모델은 상추의 성장 둔화를 반영하지 못해 부적합하였다. 시그모이드와 선형지수 모델은 느린 초기 성장, 급격한 중기 성장, 후기 성장 정체라는 생물학적 성장 패턴을 효과적으로 포착하였다. 특히 선형지수 모델은 학습 데이터에서 0.99 이상의 결정계수를 보였으며, 검증 데이터에서도 0.93 이상의 높은 예측 정확도를 나타내어, 시장성 있는 생육 기간(정식 후 28일) 내 예측에 적합하였다. 이후 과도한 잎 성장으로 상품 가치가 저하됨에 따라, 상업적 품질 유지 기간 내 생육 모델링의 중요성이 강조되었다. 본 연구 결과는 수직농장에서 상추 생육 예측과 수확시기를 최적화하기 위한 신뢰있는 모델로 선형지수 모델을 강조하였다.

키워드

선형지수 함수

생육 모델링

상추

예측 모델

수직농장

References

Archontoulis SV, Miguez FE (2015) Nonlinear regression models and application in agricultural research. Agron J 107:786-798. https://doi.org/10.2134/agronj2012.0506

10.2134/agronj2012.0506

Baek YT, Sul SG, Cho YY (2023) Estimation of days after transplanting using an artificial intelligence CNN (convolutional neural network) model in a closed-type plant factory. Hortic Sci Technol 41:81-90. https://doi.org/10.7235/HORT.20230008

10.7235/HORT.20230008

Cha MK, Cho JH, Cho YY (2013) Growth of leaf lettuce as affected by light quality of LED in closed-type plant factory system. Protected Hortic Plant Fac 22:291-297

Cha MK, Kim JS, Cho YY (2012) Growth response of lettuce to various levels of EC and light intensity in plant factory. J Bio-Env Con 21:305-311

Cha MK, Son JE, Cho YY (2014) Growth model of sowthistle (Ixeris dentata Nakai) using expolinear function in a closed-type plant production system. Hortic Sci Technol 32:165-170. https://doi.org/10.7235/hort.2014.13141

10.7235/hort.2014.13141

Li Q, Gao H, Zhang X, Ni J, Mao H (2022) Describing lettuce growth using morphological features combined with nonlinear models. Agronomy 12:860. https://doi.org/10.3390/agronomy12040860

10.3390/agronomy12040860

Ng ESQ (2018) Effects of light intensity and duration on physiology, productivity, quality and tipburn in lettuce. PhD Dissertation, National Institute of Education, Nanyang Technological University, Singapore

Shasteen K, Kacira M (2023) Predictive modeling and computer vision-based decision support to optimize resource use in vertical farms. Sustainability 15:7812. https://doi.org/10.3390/su15107812

10.3390/su15107812

Sul SG, Baek YT, Cho YY (2022) Effects of light intensity, light quality and photoperiod for growth of perilla in a closed-type plant factory system. J Bio-Env Con 31:180-187. https://doi.org/10.12791/KSBEC.2022.31.3.180

10.12791/KSBEC.2022.31.3.180

Sul SG, Baek YT, Cho YY (2023) Growth modeling of Perilla frutescens (L.) Britt. using expolinear function in a closed-type plant factory system. J Bio-Env Con 32:34-39. https://doi.org/10.12791/KSBEC.2023.32.1.034

10.12791/KSBEC.2023.32.1.034

Sung BH, Lee WK, Jeon SJ, Won JH, Cho YY (2024) Prediction lettuce growth using the artificial intelligence model XGBoost. J Agri Life Environ Sci 36:289-296. https://doi.org/10.22698/jales.20240023

10.22698/jales.20240023

Vatistas C, Avgoustaki DD, Bartzanas T (2022) A systematic literature review on controlled-environment agriculture: How vertical farms and greenhouses can influence the sustainability and footprint of urban microclimate with local food production. Atmosphere 13:1258. https://doi.org/10.3390/atmos13081258

10.3390/atmos13081258

Woo HK, Zhang F, Lee JW, Son JE (2016) Improvement of canopy light distribution, photosynthesis, and growth of lettuce (Lactuca Sativa L.) in plant factory conditions by using filters to diffuse light from LEDs. Hortic Sci Technol 34:84-93. https://doi.org/10.12972/kjhst.20160015

10.12972/kjhst.20160015

Information

Publisher :KOREAN SOCIETY FOR HORTICULTURAL SCIENCE
Publisher(Ko) :한국원예학회
Journal Title :Horticultural Science and Technology
Journal Title(Ko) :원예과학기술지
Received Date : 2025-09-17
Revised Date : 2025-12-08
Accepted Date : 2026-01-06
DOI :https://doi.org/10.7235/HORT.20250084

Horticultural Science and Technology 원예과학기술지 ISSN:1226-8763(Print) 2465-8588(Online)

Online First