This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.
Aanya Sharma,
Manisha Sen,
- Student, Lakshmi Narain College of Technology & Science, Madhya Pradesh, India
- Assistant Professor, Lakshmi Narain College of Technology & Science, Madhya Pradesh, India
Abstract
AI and machine learning models now help manage resource supplies and price changes across key Indian markets. The development and application of advanced predictive analytics in India’s agriculture supply chain and wholesale energy market systems are examined here. Although agriculture occupies over 40% of India’s land area, it accounts for approximately 15–18% of the national Gross Domestic Product (GDP). In order to regulate the PDS, unify land information, and optimize the Localized Agricultural Value Chains (LAVC), new ML- target prediction algorithms will be tested and implemented, focusing on reforming the agriculture sector which suffers from enormous inefficiencies. A strong forecasting model, that can accommodate large variations, is also needed to forecast the day ahead electricity prices in the IEX. This paper proposes a point-wise prediction and interpretable framework for Electricity Price Forecasting (EPF) using a Lasso Estimated AutoRegressive (LEAR) model and Deep Neural Networks. In addition, the paper provides evidence of serious flaws in existing forecasting literature and the need for both standardization and strict evaluation criteria in the field. Using an extensive 973-day hourly dataset (23,352 samples) from the IEX, predictive performance is evaluated using Relative Mean Absolute Error (RMAE) and Diebold-Mariano (DM) statistical tests. Experimental results demonstrate that the LEAR model achieves an RMAE of 0.812 and the DNN achieves 0.785 against naive benchmarks, confirming superior accuracy and statistical significance (p < 0.001). Thus, the main contribution of this research is the demonstration of the potential of formal AI and ML frameworks to devise safe, profitable, and sustainable agricultural and energy systems nationally.
Keywords: Artificial Intelligence (AI) & Machine Learning (ML), Agricultural Supply Chains, Indian Energy Exchange (IEX), Dynamic Energy Pricing, Electricity Price Forecasting (EPF), Smart Resource Management, Deep Neural Networks (DNN)
References
[1] Chelliah BJ, Latchoumi TP, Senthilselvi A. Analysis of demand forecasting of agriculture using machine learning algorithm. Environment, Development and Sustainability. 2024 Jan;26(1):1731-47.https://doi.org/10.1007/s10668-022-02783-9
[2] Elufioye OA, Ike CU, Odeyemi O, Usman FO, Mhlongo NZ. Ai-Driven predictive analytics in agricultural supply chains: a review: assessing the benefits and challenges of ai in forecasting demand and optimizing supply in agriculture. Computer Science & IT Research Journal. 2024 Feb 18;5(2):473-97.https://doi.org/10.51594/csitrj.v5i2.817
[3] Hassoun A, Jagtap S, Garcia-Garcia G, Trollman H, Jha SN, Sharma M, et al. Food industry 4.0: Market trends, key technologies, and challenges in agriculture and supply chain optimization. Comput Ind Eng. 2023;176:108924. doi: 10.1016/j.cie.2022.108924
[4] Nimbokar A, Kulkarni S, Deshmukh P. Data-driven machine learning algorithms for crop yield prediction and supply chain distribution in Indian agriculture. J Agric Inform. 2024;15(2):112–128. doi: 10.17700/jai.2024.15.2.112
[5] Mahul O, Stutley CJ. Government support to agricultural insurance: challenges and options for developing countries. Washington (DC): World Bank Publications; 2010. doi: 10.1596/978-0-8213-8217-2
[6] Gowthaman, S. (2023). Optimized resource allocation and supply chain management using machine learning in Indian agrarian markets. International Journal of Systems Science, 54(8), 1640-1655. https://doi.org/10.1080/00207721.2023.2185201
[7] Lago J, Marcjasz G, De Schutter B, Weron R. Forecasting day-ahead electricity prices: a review of state-of-the-art algorithms, best practices and an open-access benchmark. Appl Energy. 2021;293:116983. https://doi.org/10.1016/j.apenergy.2021.116983
[8] Veeramsetty V, Salkuti SR, Nikitha B, Siddartha T. Electrical energy price dataset – IEX data. Mendeley Data. 2024;V1. doi:10.17632/v5znbkzjd4.1.
[9] Marcjasz G, Uniejewski B, Weron R. Beating the naive baseline: Lasso estimated autoregressive (LEAR) models for day-ahead electricity price forecasting. Energy Econ. 2023;120:106610. doi:10.1016/j.eneco.2023.106610.
[10] Sakthi S, Sundaramurthy M, V K. Machine learning applications for dynamic pricing and electricity spot price evaluation on the Indian Energy Exchange. IEEE Trans Power Syst. 2025;40(1):412–25. doi:10.1109/TPWRS.2024.3351234.
[11] Fraunholz C, Keles D, Fichtner W. On the flaws of traditional error metrics in dynamic electricity price forecasting. Energy J. 2021;42(4):89–110. doi:10.5547/01956574.42.4.cfra.
[12] Puphaiboon K, Yasukam T, Thanrungrueng J, Anchaleechamaikorn T, Sumiyoshi T. Cross-Market Arbitrage in Japanese Electricity Markets: A Hybrid Machine Learning Approach with Live Revenue Validation. Available at SSRN 6887039. 2026 Jun 6.
Trends in Electrical Engineering
| Volume | 16 | |
| 03 | ||
| Received | 06/10/2026 | |
| Accepted | 23/10/2026 | |
| Published | 30/10/2026 | |
| Publication Time | 24 Days |