The factors affecting the behavior response of service providers in manure returning based on machine learning: Empirical evidence from the green planting-breeding cycle agriculture pilot program
To enhance the behavior responses of service providers in manure returning and address the “last mile problem” in establishing green planting-breeding cycle agriculture, this study utilizes micro-level survey data from the green planting-breeding cycle agriculture pilot program. Multiple machine learning methods, including penalized regression, ensemble learning, and deep learning were used to systematically identify the factors influencing the behavior responses of service providers in manure returning. The research showed that: 1) Compared with traditional multiple linear regression, machine learning methods can significantly improve the trustworthiness and interpretability of model results. Economic incentives, fixed assets, operating income, economic benefit cognition, and service radius are the key factors affecting the behavior response of service providers in manure returning, all of which exert significant positive impacts. 2) Through Accumulated Local Effects (ALE) plots and SHapley Additive exPlanations (SHAP) value polynomial fitting dependence plots, it was found that there are obvious non-linear relationships between three factors: fixed assets, economic incentives, and service radius, and the behavior response of service providers in manure returning. Among them, the fitting curves of fixed assets and service radius present an inverted U-shaped pattern, while that of economic incentives shows a U-shaped pattern. 3) The interaction effect between fixed assets and economic incentives can significantly enhance the behavior response of service providers in manure returning. When the stock of fixed assets exceeds 2.5 million yuan, the positive effect of economic incentives on service providers is significantly strengthened, with SHAP values increasing from below 0.3 to 0.5-1.0. 4) The interaction effect between service radius and operating income also significantly promotes the behavior response of service providers in manure returning. When the service radius covers 10 to 30 villages, service providers with high operating income can more efficiently convert the expanded coverage into economies of scale benefits, thereby improving their willingness to supply services. Based on these findings, this study puts forward the following policy recommendations: Implementing tiered economic incentive policies based on asset scale and service radius, promoting the layout of integrated regional crop-livestock farming, optimizing the manure transportation network, and strengthening the capacity building of service providers.
Office of the Ministry of Agriculture and Rural Affairs, Office of the Ministry of Finance. Notice on carrying out pilot work of green planting-breeding and circular agriculture[EB/OL]. [--]. in Chinese)
[3]
刘趁. 我国农业绿色种养循环发展新模式初见成效[N]. 农民日报, --(3)
[4]
LiuC. The new model of green planting, breeding and circular development in agriculture in China has achieved initial results[N]. Farmers’ Daily, --(3) (in Chinese)
ZhouX, DuS, XuY, FuG H, HuJ P, WuY Q. Route exploration of manure extension and apllication on three major grain crops in China[J]. Chinese Journal of Agricultural Resources and Regional Planning, 2025, 46(3): 82-90 (in Chinese)
LiuZ T, PeiW J. Analysis on the mode and restricting factors of manure returning to field in the development of green circular agriculture by combining planting with feeding[J]. Agricultural Comprehensive Development in China, 2024(12): 44-47 (in Chinese)
[9]
van SelmB, HijbeekR, van IttersumM K, van HalO, van MiddelaarC E, de BoerI J M. Recoupling livestock and feed production in the Netherlands to reduce environmental impacts[J]. Science of the Total Environment, 2023, 899: 165540
[10]
RichardsD, YabarH. Promoting energy and resource recovery from livestock waste: Case study Yuge Farm, Japan [J]. Case Studies in Chemical and Environmental Engineering, 2023, 7: 100299
[11]
YoonI, OhS H, KimS W. Sustainable animal agriculture in the United States and the implication in Republic of Korea[J]. Journal of Animal Science and Technology, 2024, 66(2): 279-294
[12]
HeZ Q, ZhangH L. Applied Manure and Nutrient Chemistry for Sustainable Agriculture and Environment[M]. Dordrecht: Springer Netherlands, 2014
[13]
JinS Q, ZhangB, WuB, HanD M, HuY, RenC C, ZhangC Z, WeiX, WuY, MolA P J, ReisS, GuB J, ChenJ. Decoupling livestock and crop production at the household level in China[J]. Nature Sustainability, 2021, 4(1): 48-55
JinS Q, ZhangJ H, PangJ, HuY. The historical context, theoretical logic and practical trend of the relationship between livestock and crop production in China[J]. China Rural Survey, 2025(5): 23-41 (in Chinese)
MiG H, HuoY W, ZengA J, LiG H, WangX, ZhangF S. Integration of agricultural machinery and agronomic techniques for crop nutrient management in China[J]. Scientia Agricultura Sinica, 2022, 55(21): 4211-4224 (in Chinese)
WuD M, ZhangZ Y, HuangZ, LanB H, BaiR B. Technology of manure returning to field in large scale pig farm based on nutrient management[J]. Journal of Domestic Animal Ecology, 2025, 46(8): 96-100 (in Chinese)
[20]
郭庆海. 渐行渐远的农牧关系及其重构[J]. 中国农村经济, 2021(9): 22-35
[21]
GuoQ H. The fading relationship between agriculture and animal husbandry and its reconstruction[J]. Chinese Rural Economy, 2021(9): 22-35 (in Chinese)
XiaJ Q, HeK, ZhangJ B. Effect of environmental regulations and village regulations on farmers’ green production willingness: Taking the resource use of livestock and poultry waste of large-scale pig farmers as an example [J]. Chinese Journal of Eco-Agriculture, 2019, 27(12): 1925-1936 (in Chinese)
WangJ H, TaoJ Y, ChenL. Resource utilization method and influencing factors of farmers’ behaviors towards livestock and poultry waste [J]. China Population, Resources and Environment, 2019, 29(5): 127-137 (in Chinese)
HuN J, SunX L, XuY T, ZhouZ Y, ZhuL Q. Influencing factors of farmers’ organic fertilizer application behavior and their stratification based on Logistic-ISM model [J]. Resources Science, 2019, 41(6): 1120-1130 (in Chinese)
TaoY, QiuX W, ZhouY X, HuJ L. The conflict between risk perception, social trust and farmers’ organic fertilizer substitution behavior [J]. Journal of Agrotechnical Economics, 2022(5): 49-64 (in Chinese)
LuY, ZhangY Q, LiB, ZhaoH Y. Managerial individual characteristics and corporate performance: Evidence from a machine learning approach[J]. Journal of Management Sciences in China, 2020, 23(2): 120-140 (in Chinese)
[32]
HolmstromB, MilgromP. Multitask principal-agent analyses: Incentive contracts, asset ownership, and job design [J].Journal of Law, Economics,& Organization, 1991, 7: 24-52
[33]
HuselidM A. The impact of human resource management practices on turnover, productivity, and corporate financial performance [J]. The Academy of Management Journal, 1995, 38(3): 635-672
DuY W, JingJ. On the farmers’ willingness for green production and their relevant behaviors in the perspective of the rural revitalization strategy [J]. Journal of Yunnan Minzu University: Philosophy and Social Sciences Edition, 2019, 36(1): 95-103 (in Chinese)
[36]
BurtonR J F. The influence of farmer demographic characteristics on environmental behavior: A review[J]. Journal of Environmental Management, 2014, 135(4): 19-26
XueC X. The leading effect of farmers’ political identity on green agricultural production technology [J]. Journal of Northwest A&F University: Social Science Edition, 2022, 22(3): 148-160 (in Chinese)
[39]
周其仁, 罗德胤. 城乡中国[J]. 世界建筑, 2017(12): 121-121
[40]
ZhouQ R, LuoD Y. Urban rural China[J]. World Architecture, 2017(12): 121-121 (in Chinese)
TongZ H, WenT J. The organizational path of capital and sector to the countryside and smallholder economy-and questioning the path of professional cooperatives[J]. Open Times, 2009(4): 5-26 (in Chinese)
[43]
许佳彬. 中国奶牛养殖业绿色转型的驱动机制研究[D]. 哈尔滨: 东北农业大学, 2023
[44]
XuJ B. Study on the driving mechanism of green transformation of the Chinese dairy farming industry[D]. Harbin: Northeast Agricultural University, 2023 (in Chinese)
LuY, ShiH Q. Determinants of financing for high-tech enterprises in China: Machine learning analysis based on the STAR market [J]. Journal of Financial Research, 2022(9): 132-151 (in Chinese)
QiX L, WangX R. Factors influencing the citizenization of rural-urban migrants: A study based on machine learning [J]. Chinese Rural Economy, 2024(5): 128-150 (in Chinese)
[52]
ChenT Q, GuestrinC. XGBoost: A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. -,, San Francisco, California, USA. ACM, 2016, 785-794
YangL, WangY. Review of various cross-validation estimation methods of generalization error[J]. Application Research of Computers, 2015, 32(5): 1287-1290, 1297 (in Chinese)
[55]
BertomeuJ, CheynelE, FloydE, PanW Q. Using machine learning to detect misstatements[J]. Review of Accounting Studies, 2021, 26(2): 468-519
ChenY S, ZhouJ Y, HuangJ Q. How does the generosity of enterprises come: Evidence from machine learning[J]. Journal of Finance and Economics, 2023, 49(6): 153-169 (in Chinese)
[58]
Martín-Baos J Á, López-GómezJ A, Rodriguez-BenitezL, HillelT, García-RódenasR. A prediction and behavioural analysis of machine learning methods for modelling travel mode choice[J]. Transportation Research Part C: Emerging Technologies, 2023, 156: 104318
ZhangT C, YanT W, HeK, ZhangJ B. Impact of capital endowment on peasants’ willingness to invest in green production: Taking crop straw returning to the field as an example [J]. China Population, Resources and Environment, 2017, 27(8): 78-89 (in Chinese)
YangZ H. Ageing, social network and the adoption of green production technology: Evidence from farm households in six provinces in the Yangtze River Basin [J]. China Rural Survey, 2018(4): 44-58 (in Chinese)
ZhangC, HuangB, ZhongZ. The decision mechanism of agricultural socialized service radius: Evidence from four agricultural cooperatives[J]. Reform, 2020(12): 121-131 (in Chinese)
HuangY Z, LuoX F, LiR R, ZhangJ B. Farmer cognition, external environment and willingness of green agriculture production: Based on the survey data of 632 farmers in Hubei Province [J]. Resources and Environment in the Yangtze River Basin, 2018, 27(3): 680-687 (in Chinese)