SELECTION OF AN ALGORITHM FOR THE PREDICTION OF STOPPAGES AND/OR FAILURE OF EXCAVATION UNITS USING SUPERVISED MACHINE LEARNING

التفاصيل البيبلوغرافية
العنوان: SELECTION OF AN ALGORITHM FOR THE PREDICTION OF STOPPAGES AND/OR FAILURE OF EXCAVATION UNITS USING SUPERVISED MACHINE LEARNING
المؤلفون: Misita, Mirjana, Spasojević Brkić, Vesna, Mihajlović, Ivan, Brkić, Aleksandar, Perišić, Martina, Papić, Neda, Janev, Nemanja
المصدر: IMCSM Proceedings - International May Conference on Strategic Management – IMCSM24, May 31, 2024, Bor
بيانات النشر: Technical Faculty in Bor
سنة النشر: 2024
مصطلحات موضوعية: excavator, stoppage, failure, machine learning
الوصف: The paper presents research into the possibility of applying machine learning algorithms in the prediction of stoppages and/or failure of excavator units. Regression trees, Random Forest and Support Vector Machine (SVM) algorithms were tested with different hyperparameter variations on the collected set of data on the causes and downtime of stoppages of the observed excavator units. The result indicates that the trained SVM algorithm with sufficient accuracy (MSE 0.106) can predict the stoppages of the observed excavator units. Further research is aimed at expanding the database and further improving the possibility of predicting the level of danger for various causes of stoppages and/or failure of the observed excavator units.
نوع الوثيقة: conference object
اللغة: English
Relation: Support Systems for Smart, Ergonomic and Sustainable Mining Machinery Workplaces – SmartMiner, #GRANT No. 5151 supported by the Science Fund of the Republic of Serbia; info:eu-repo/grantAgreement/MESTD/inst-2020/200105/RS//; https://machinery.mas.bg.ac.rs/handle/123456789/7810; http://machinery.mas.bg.ac.rs/bitstream/id/19668/bitstream_19668.pdf
DOI: 10.5937/IMCSM24008M
الاتاحة: https://machinery.mas.bg.ac.rs/handle/123456789/7810
https://doi.org/10.5937/IMCSM24008M
http://machinery.mas.bg.ac.rs/bitstream/id/19668/bitstream_19668.pdf
Rights: openAccess ; https://creativecommons.org/licenses/by/4.0/ ; BY
رقم الانضمام: edsbas.A40892D5
قاعدة البيانات: BASE