Artificial neural network-based repair and maintenance cost estimation model for rice combine harvesters

This research proposes an artificial neural network (ANN)-based repair and maintenance (R&M) cost estimation model for agricultural machinery. The proposed ANN can achieve high accuracy with small data requirement. In the study, is implemented to estimate R&M costs using a sample of locally-made rice combine harvesters. inputs are geographical regions, harvest area, curve fitting coefficients related historical data; output estimated cost. Multilayer feed-forward adopted as processing algorithm Levenberg-Marquardt backpropagation learning training algorithm. ANN-based model, results compared those conventional mathematical model. reveal that percentage error between models below 1%, indicating model’s predictive accuracy. useful setting service rates machinery, given significance in profitability. novelty this lies use curve-fitting improve Besides, could be further developed into web-based applications programming language enable ease greater user accessibility. Moreover, minor modifications, also applicable other areas tractors or harvesters different countries origin. Key words: cost, network, coefficients, DOI: 10.25165/j.ijabe.20231602.5931 Citation: Numsong A, Posom J, Chuan-Udom S. Artificial network-based Int J Agric & Biol Eng, 2023; 16(2): 38－47.