Enhancing agriculture supply chains through regional demand forecasting with Machine Learning Algorithms
Keywords:
Agriculture Supply Chain; Machine Learning; Demand Forecasting; Fair Pricing; recommendation System; Farmer-Consumer PlatformAbstract
With the stats analysis of 2014 of Our Country Agricultural supply chain faces a significant inefficiencies as the intermediate stakeholders capture the large portion of the value, in this scenario farmers only earn about 31 to 43 percent of final market price. This imbalance contributes to higher consumers expenditures and unequal income distribution and substantial post harvest waste. To tackle these challenges this research introduces kisaankart which is a machine learning- driven a digital framework that is designed to enhance the demand prediction accuracy and make transparent fair pricing between farmers and consumers connectivity. This approach explore the potential of artificial intelligence models check price forecasting and market trends analysis. By Using govt mandi prices data, simulated orders and users interaction records, and preprocessing and feature engineering are applied to both categorical and temporal variables. The performance of some machine learning like ARIMA, XGBOOST, and LSTM/GRU are used assessed through RMSE and MAE comparison supported by pilot feedback and cross validation and for model validation and measurement of user responses. Resulting indicates that ML based system outperform traditional stats models in identifying the some complex seasonal and market dynamics. With the Additional the hybrid recommended engine significantly enhances users engagement metrices such as click through the conversion rates. Simulations of market outcomes suggest that direct digital trade could raise farmers earning by 25 to 35 and reduced in consumers prices by 15 to 25 percent, So Overall, kissankart demonstrate that implementing of machine learning and real world validation are recommended to enhance scalability and long term sustainability in transforming Indias Agriculture value network.
References
India Data Portal. Area, production, and yield (APY) dataset. Available from: https://indiadataportal.com/
Ministry of Agriculture and Farmers Welfare, Government of India. Agricultural statistics. New Delhi: Government of India.
Kurumatani T. Time series forecasting of agricultural product prices based on recurrent neural networks and its evaluation method. SN Appl Sci. 2020;2(8):1–17.
Ly R, Traore F, Dia K. Forecasting commodity prices using long short-term memory neural networks. Int Food Policy Res Inst. 2021;2000.
Dairi J, et al. Deep learning approaches for time series prediction: A comparative study. J Biomed Inform. 2021;118:103791.
T A, et al. Forecasting financial time- series data using BiLSTM and comparative analysis with ARIMA and LSTM models. J Finance Data Sci. 2022.
Sabu K, et al. Predictive analysis in agriculture: Commodity price forecasting using LSTM in Kerala, India. J Agri Food Syst. 2020.
Chen D, et al. Comparative analysis of machine learning approaches. J Food Eng Sci. 2021.
Rahman O. Forecasting financial time series using deep learning networks: Evidence from long-short term memory and gated recurrent unit. Int J Financ Stud. 2022.
BDT. A survey of hybrid forecasting models combining statistical methods and deep learning for complex time series. Appl Sci (MDPI). 2024.
Gu YH, Jin D, Yin H, Zheng R, Piao X, Yoo SJ. Forecasting agricultural commodity prices using dual input attention LSTM. Agriculture. 2022;12(2):256.
Kumar H, et al. Post-harvest loss mitigation through development and testing of solar-grid hybrid cold storage system. Energies. 2021;14(22):7633.
Manogna RL. Innovation and firm growth in agricultural inputs industry: Empirical evidence from India. J Agribus Dev Emerg Econ. 2021;11(5):506–519.
GKP, et al. Problems and prospects of agricultural marketing in India. J Econ Dev Res. 2021.
CMT. Crop yield prediction using random forest algorithm and XGBoost machine learning model in Indian agriculture. J Smart Farming. 2024.
Hochreiter S, Schmidhuber J. Long short- term memory. Neural Comput. 1997;9(8):1735–1780.
Gers FA, Schmidhuber J, Cummins F. Learning to forget: Continual prediction with LSTM. Neural Comput. 2000;12(10):2451–2471.
Hsu D. Time series forecasting based on augmented long short-term memory. arXiv preprint arXiv:1707.00666; 2017.
Guo T, Xu Z, Yao X, Chen H, Aberer K, Funaya K. Robust online time series prediction with recurrent neural networks. In: Proc IEEE Int Conf Data Sci Adv Anal (DSAA); 2016. p. 816–825.
Li Y, Zhu Z, Kong D, Han H, Zhao Y. EALSTM: Evolutionary attention-based LSTM for time series prediction. Knowl Based Syst. 2019;181:104785.
Duan Q, Zhang L, Wei F, Xiao X, Wang L. Forecasting model and validation for aquatic product price based on time series GA-SVR. Trans Chin Soc Agric Eng. 2017;33(1):308–314.
Karia A, Bujang I. Progress accuracy of CPO price prediction: Evidence from ARMA family and artificial neural network approach. Int Res J Finance Econ. 2011;64:66–79.
TTAH. Comparing random forest and XGBoost for sales prediction: A technical review. IEEE Trans Knowl Data Eng. 2023.
SAG, et al. Artificial intelligence in agriculture: Ethics, impact, possibilities, and pathways for policy. J Sustain Agri Food Syst. 2024.
AKR, et al. Role of explainable AI in crop recommendation technique of smart farming. IEEE Access. 2024.
Gowthaman T, Kumar KS, Adarsh VS, Bhattacharyya B. Trend analysis and ARIMA models for water quality parameters of Brahmani River, Odisha, India. Int J Environ Clim Change. 2022;12(12):219–228.
Jin D, Yin H, Gu Y, Yoo SJ. Forecasting of vegetable prices using STL-LSTM method. In: Proc Int Conf Syst Inform (ICSAI); 2019. p. 866–871.
PPC, et al. The role of D2C organic food and agri-tech sectors in India's Green Revolution 2.0. J Agribus Dev. 2023.
