eprintid: 20076 rev_number: 2 eprint_status: archive userid: 1 dir: disk0/00/02/00/76 datestamp: 2024-06-04 14:19:49 lastmod: 2024-06-04 14:19:49 status_changed: 2024-06-04 14:16:33 type: conference_item metadata_visibility: show creators_name: Okwu, M.O. creators_name: Otanocha, O.B. creators_name: Edward, B.A. creators_name: Oreko, B.U. creators_name: Oyekale, J. creators_name: Oyejide, O.J. creators_name: Osuji, J. creators_name: Maware, C. creators_name: Ezekiel, K. creators_name: Orikpete, O.F. title: Investigating the Accuracy of Artificial Neural Network Models in Predicting Surface Roughness in Drilling Processes ispublished: pub keywords: Amides; Forecasting; Grinding (machining); Hardening; Infill drilling; MATLAB; Mean square error; Neural networks; Shrinkage, Artificial intelligence; Artificial neural network; Artificial neural network modeling; Computational intelligence; Drilling operation; Drilling process; Error values; Machining operations; Means square errors; Oil hardened non-shrinking, Surface roughness note: cited By 0; Conference of 5th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2023 ; Conference Date: 22 November 2023 Through 24 November 2023; Conference Code:198427 abstract: In recent times, the application of Artificial Intelligence (AI) has become widespread across various fields, including machining operations like milling, drilling, shaping, turning, grinding, counter drilling, and counter sinking. Among the techniques used in these applications, Artificial Neural Networks (ANN) have gained popularity. This study focuses on utilizing ANN to optimize drilling hardened non-shrinking (OHNS) die steel. The methodology presented in this research involves the implementation of ANN to analyze the impact of drilling process parameters on surface roughness. MATLAB 2023 software was used to effectively train the neural network. The dataset for this study consists of input and output expressions derived from experimental analysis. The input data includes variables such as cutting speed, feedrate, drill size, and depth of cut. The objective of this study is to predict surface roughness based on input datasets, with 70 of the samples allocated as training data and the remaining 30 for testing and validation. The experimental results demonstrate that ANN provides a reliable prediction capability, with Coefficient of Determination (R2) and Mean Square Error (MSE) values of 0.997 and 0.231893, respectively. The low MSE value indicate better model performance, also the R2 value obtained indicate a strong correlation between the predicted and actual values, suggesting that the ANN model provides a satisfactory fit to the data, indicating a high level of confidence in the statistical results. Thus, it can be concluded that the results obtained from the ANN model are statistically significant, slightly superior to the classical model, and exhibit a good fit. © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) date: 2024 publisher: Elsevier B.V. official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189831808&doi=10.1016%2fj.procs.2024.02.020&partnerID=40&md5=8648e9cabcf1df68a7070f5417f7d107 id_number: 10.1016/j.procs.2024.02.020 full_text_status: none publication: Procedia Computer Science volume: 232 pagerange: 1982-1990 refereed: TRUE issn: 18770509 citation: Okwu, M.O. and Otanocha, O.B. and Edward, B.A. and Oreko, B.U. and Oyekale, J. and Oyejide, O.J. and Osuji, J. and Maware, C. and Ezekiel, K. and Orikpete, O.F. (2024) Investigating the Accuracy of Artificial Neural Network Models in Predicting Surface Roughness in Drilling Processes. In: UNSPECIFIED.