Prediction of Wheat Milling Yield Using Machine Learning: Identification of Seed Traits Associated with Milling Performance

Myoung-Goo Choi; Go Eun Lee; Jinhee Park; Chang-Hyun Choi; Jae-Han Son; Junseok Choi; Yurim Kim; Sukjin Kim; Chon-Sik Kang; Jeong-Heui Lee

doi:10.7740/kjcs.2025.70.4.291

All Issue

2025 Vol.70, Issue 4 Preview Page Next Page

Original Research Article

Prediction of Wheat Milling Yield Using Machine Learning: Identification of Seed Traits Associated with Milling Performance 기계학습 기반 밀 제분율 예측 및 연관 종자 형질 규명

1 December 2025. pp. 291-301

PDF XML

Abstract

Wheat (Triticum aestivum L.) is a cornerstone of global food security, and improving milling yield is vital for enhancing processing efficiency and the economic value of grain-based industries. However, milling yield is a multifactorial trait governed by complex interactions among physical and chemical kernel properties, which are often inadequately captured by conventional linear models. In this study, a machine learning framework based on the Random Forest (RF) regression algorithm was established to predict milling yield and to quantify the relative contribution of kernel attributes. Forty-five Korean wheat cultivars were evaluated for key physical traits (kernel weight, hardness, and particle size) and chemical traits (protein, ash, and amylose contents). The RF model achieved a high predictive accuracy (R² = 0.94; RMSE = 1.341), demonstrating its robustness in modeling the non-linear relationships among grain traits. Variable importance and partial dependence analyses identified kernel hardness, particle size, and kernel weight as the most influential predictors of milling yield, while protein and ash contents exerted relatively minor effects. A distinct non-linear response was detected in the hardness range of 25–40, where milling yield increased sharply before reaching a plateau. Principal component and cluster analyses further grouped cultivars into four distinct clusters according to trait combinations, underscoring that milling performance is determined by multivariate interactions rather than single trait effects. Collectively, this RF-based analytical framework provides a data-driven approach for integrating complex quality traits into wheat breeding and selection programs, thereby supporting the development of high-yielding and quality-stable Korean wheat cultivars optimized for milling performance.

Keywords

wheat

milling yield

kernel hardness

Random Forest

References

AACC International. 2000. Approved methods of analysis, 10th ed. AACC International : St. Paul, MN, USA.

Acar, O., T. Sanal, and H. Köksel. 2019. Effects of wheat kernel size on hardness and various quality characteristics. Qual. Assur. Saf. Crops Foods. 11 : 459-64.

10.3920/QAS2019.1552

Barrera, G. N., J. Méndez-Méndez, I. Arzate-Vázquez, G. Calderón-Domínguez and P. D. Ribotta. 2019. Nano- and micro-mechanical properties of wheat grain by atomic force microscopy (AFM) and nano-indentation (IIT) and their relationship with the mechanical properties evaluated by uniaxial compression test. J. Cereal Sci. 90 : 102830.

10.1016/j.jcs.2019.102830

Barroso Lopes, R., E. Salman Posner, A. Alberti, and I. Mottin Demiate. 2022. Pre milling debranning of wheat with a commercial system to improve flour quality. J. Food Sci. Technol. 59 : 3881-87.

10.1007/s13197-022-05411-636193367PMC9525502

Benos, L., A. C. Tagarakis, G. Dolias, R. Berruto, D. Kateris, and D. Bochtis. 2021. Machine Learning in Agriculture: A Comprehensive Updated Review. Sensors (Basel). 21(11) : 3758.

10.3390/s2111375834071553PMC8198852

Bohra, A., M. Choudhary, D. Bennett, R. Joshi, R. R. Mir, and R. K. Varshney. 2024. Drought-tolerant wheat for enhancing global food security. Funct. Integr. Genomics. 24 : 212.

10.1007/s10142-024-01488-8

Breiman, L. 2001. Random Forests. Machine Learning. 45 : 5-32.

10.1023/A:1010933404324

Cammerata, A., F. Sestili, B. Laddomada, and G. Aureli. 2021. Bran-Enriched Milled Durum Wheat Fractions Obtained Using Innovative Micronization and Air-Classification Pilot Plants. Foods. 10(8) : 1796.

10.3390/foods1008179634441573PMC8391628

Chen, F., Z. H. He, X. C. Xia, L. Q. Xia, X. Y. Zhang, M. Lillemo, and C. F. Morris. 2006. Molecular and biochemical characterization of puroindoline a and b alleles in Chinese landraces and historical cultivars. Theor Appl Genet 112(3) : 400-409.

10.1007/s00122-005-0095-z

Chen, T. and C. Guestrin. 2016. XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA. pp. 785-794.

10.1145/2939672.2939785

Choy, A. L., C. K. Walker and J. F. Panozzo. 2015. Investigation of Wheat Milling Yield Based on Grain Hardness Parameters. Cereal Chem. 92 : 544-50.

10.1094/CCHEM-04-14-0072-R

Delcour, J. A., I. J. Joye, B. Pareyt, E. Wilderjans, K. Brijs, and B. Lagrain. 2012. Wheat gluten functionality as a quality determinant in cereal-based food products. Annu. Rev. Food Sci. Technol. 3 : 469-92.

10.1146/annurev-food-022811-101303

Dhillon, M. S., T. Dahms, C. Kuebert-Flock, T. Rummler, J. Arnault, I. Steffan-Dewenter and T. Ullmann. 2023. Integrating random forest and crop modeling improves the crop yield prediction of winter wheat and oil seed rape. Front. Remote Sens. 3 : 1010978.

10.3389/frsen.2022.1010978

Dziki, D., A. Krajewska and P. Findura. 2024. Particle Size as an Indicator of Wheat Flour Quality: A Review. Processes. 12(11) : 2480

10.3390/pr12112480

Filip, E., K. Woronko, E. Stepien, and N. Czarniecka. 2023. An Overview of Factors Affecting the Functional Quality of Common Wheat (Triticum aestivum L.). Int. J. Mol. Sci. 24(8) : 7524.

10.3390/ijms2408752437108683PMC10142556

Friedman, J. H. 2001. Greedy function approximation: A gradient boosting machine. Ann. Stat. 29 : 1189-1232.

10.1214/aos/1013203451

Greenwell, B. M. 2017. pdp: An R package for constructing partial dependence plots, R Journal. 9 : 421-36.

10.32614/RJ-2017-016

Hassoon, W. H., D. Dziki, A. Miś and B. Biernacka. 2021. Wheat Grinding Process with Low Moisture Content: A New Approach for Wholemeal Flour Production. Processes. 9(1) : 32.

10.3390/pr9010032

Hook, S. C. W., G. T. Bone and T. Fearn. 1982. The conditioning of wheat. The effect of increasing wheat moisture content on the milling performance of uk wheats with reference to wheat texture. J. Sci. Food Agric. 33: 655-62.

10.1002/jsfa.2740330711

Inglis, A., A. Parnell and C. B. Hurley. 2022. Visualizing Variable Importance and Variable Interaction Effects in Machine Learning Models. J. Comput. Graph. Stat. 31 : 766-778.

10.1080/10618600.2021.2007935

Kang, C.S., J. H. Son, Y. K. Cheong, K. H. Kim, Y. H. Ko, J. C. Park, Y. J. Oh, K. H. Kim, B. K. Kim, and C. S. Park. 2015. Characterization of Korean Wheat Line with Long Spike. II. Flour Characteristics and Genetic Variations. Korean J. Breed. Sci. 47 : 229-37.

10.9787/KJBS.2015.47.3.229

Khalid, A., A. Hameed, and M. F. Tahir. 2023. Wheat quality: A review on chemical composition, nutritional attributes, grain anatomy, types, classification, and function of seed storage proteins in bread making quality. Front. Nutr. 10 : 1053196.

10.3389/fnut.2023.105319636908903PMC9998918

Kim, K. H., C. S. Kang, I. Choi, H. S. Kim, J. N. Hyun, and C. S. Park. 2016. Analysis of Grain Characteristics in Korean Wheat and Screening Wheat for Quality Using Near Infrared Reflectance Spectroscopy. Korean J. Breed. Sci. 48 : 442-449.

10.9787/KJBS.2016.48.4.442

Kim, K. M., K. H. Kim, C. Choi, J. Park, G. E. Lee, S. M. Yang, C. Cho, M. H. Lee, K. C. Jang, and C. S. Kang. 2023. A Wheat Cultivar, “Hwanggeumal” with Good Bread Quality, Partial Waxy and Tolerance to Lodging and Pre-Harvest Sprouting. Korean J. Breed. Sci. 55 : 272-80.

10.9787/KJBS.2023.55.3.272

Kweon, M., R. Martin and E. Souza. 2009. Effect of Tempering Conditions on Milling Performance and Flour Functionality. Cereal Chem. 86 : 12-17.

10.1094/CCHEM-86-1-0012

Lewko, P., A. Wójtowicz and M. Gancarz. 2023. Distribution of Arabinoxylans and Their Relationship with Physiochemical and Rheological Properties in Wheat Flour Mill Streams as an Effective Way to Predict Flour Functionality. Appl. Sci. 13(9) : 5458

10.3390/app13095458

Maningat, C. C. and P. A. Seib. 2010. Understanding the Physicochemical and Functional Properties of Wheat Starch in Various Foods. Cereal Chem. 87 : 305-314.

10.1094/CCHEM-87-4-0305

Mastanjević, K., Habschied, K., Dvojković, K., Karakašić, M., and Glavaš, H. 2023. Vitreosity as a Major Grain Quality Indicator—Upgrading the Grain-Cutter Method with a New Blade. Appl. Sci. 13 : 2655.

10.3390/app13042655

Moiraghi, M., L. S. Sciarini, C. Paesani, A. E. Leon, and G. T. Perez. 2019. Flour and starch characteristics of soft wheat cultivars and their effect on cookie quality. J. Food Sci. Technol. 56 : 4474-4481.

10.1007/s13197-019-03954-931686679PMC6801258

Morris, C. F. 2002. Puroindolines: the molecular genetic basis of wheat grain hardness. Plant Mol. Biol. 48 : 633-647.

10.1023/A:1014837431178

Parrenin, L., C. Danjou, B. Agard and R. Beauchemin. 2023. A decision support tool for the first stage of the tempering process of organic wheat grains in a mill. Int. J. Food Sci. Technol. 58 : 5478-5488.

10.1111/ijfs.16406

Probst, P., M. N. Wright and A.-L. Boulesteix. 2019. Hyperparameters and tuning strategies for random forest. WIREs Data Mining and Knowledge Discovery. 9 : e1301.

10.1002/widm.1301

Przybył, K., M. Gawrysiak-Witulska, P. Bielska, R. Rusinek, M. Gancarz, B. Dobrzański and A. Siger. 2023. Application of Machine Learning to Assess the Quality of Food Products—Case Study: Coffee Bean. Appl. Sci. 13(19) : 10786.

10.3390/app131910786

R Core Team. 2024. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: https://www.R-project.org/ (Accessed on Jan. 15, 2025)

Sakhare, S. D. and A. A. Inamdar. 2014. The cumulative ash curve: a best tool to evaluate complete mill performance. J. Food Sci. Technol. 51 : 795-799.

10.1007/s13197-011-0549-z24741178PMC3981997

Shewry, P. R. and S. J. Hey. 2015. The contribution of wheat to human diet and health. Food Energy Secur. 4 : 178-202.

10.1002/fes3.6427610232PMC4998136

Shiferaw, B., M. Smale, H.-J. Braun, E. Duveiller, M. Reynolds and G. Muricho. 2013. Crops that feed the world 10. Past successes and future challenges to the role played by wheat in global food security. Food Secur. 5 : 291-317.

10.1007/s12571-013-0263-y

Tanaka, H., C. F. Morris, M. Haruna and H. Tsujimoto. 2008. Prevalence of puroindoline alleles in wheat varieties from eastern Asia including the discovery of a new SNP in puroindoline B. Plant Genet. Resour. 6 : 142-152.

10.1017/S1479262108993151

Tian, X., X. Wang, Z. Wang, B. Sun, F. Wang, S. Ma, Y. Gu and X. Qian. 2022. Particle size distribution control during wheat milling: nutritional quality and functional basis of flour products—a comprehensive review. Int. J. Food Sci. Technol. 57 : 7556-7572.

10.1111/ijfs.16120

Wang, K., and B. X. Fu. 2020. Inter-Relationships between Test Weight, Thousand Kernel Weight, Kernel Size Distribution and Their Effects on Durum Wheat Milling. Semolina Composition and Pasta Processing Quality. Foods. 9(9) : 1308.

10.3390/foods909130832948041PMC7555757

Information

Publisher :The Korean Society of Crop Science
Publisher(Ko) :한국작물학회
Journal Title :The Korean Journal of Crop Science
Journal Title(Ko) :한국작물학회지
Volume : 70
No :4
Pages :291-301
Received Date : 2025-11-11
Revised Date : 2025-11-21
Accepted Date : 2025-11-22
DOI :https://doi.org/10.7740/kjcs.2025.70.4.291

The Korean Journal of Crop Science ISSN:0252-9777(Print) 2287-8432(Online) 한국작물학회지

All Issue