Proof-of-concept of harvest peak control using a strawberry cultivation emulator with artificial weather chambers

Abstract

Abstract Strawberry ( Fragaria × ananassa ) production requires a precise regulation of harvest-peak timing to meet market demand; however, fruit ripening remains sensitive to environmental fluctuations. We developed a peak-shift control system using a maturation simulator and conducted a proof-of-concept trial in climate chambers reproducing greenhouse conditions. Two control scenarios were designed: delayed flowering, wherein the predicted peaks lagged by a week and were corrected using heating offsets, and premature flowering, wherein the peaks advanced by a week and were adjusted using cooling offsets. Temperature offsets were explored within ± 5 °C, updated approximately twice weekly, to converge the predicted peaks with the target date on December 21, 2019. Results demonstrated successful alignment within ± 1 day in three of four treatments, surpassing previously reported prediction models. Cooling and heating treatments broadened and shortened harvest distributions by 2–3 and 2–4 days, respectively, suggesting the potential for balancing yield concentration. Post-harvest evaluation revealed no significant differences in morphology, grade distribution, or class proportion, although heating significantly reduced the soluble solid content, indicating a trade-off between accelerated ripening and sweetness. To our knowledge, this study provides the first experimental demonstration of simulation-in-the-loop, proactive harvest-peak control in strawberry, forming a basis for digital-twin-based cultivation control.