Open AccessA Comparison of Machine Learning Algorithms in Manufacturing Production Process
Author(s) -
Rosalina Rosalina
Publication year - 2019
Publication title -
commit (communication and information technology) journal/commit journal
Language(s) - English
Resource type - Journals
eISSN - 2460-7010
pISSN - 1979-2484
DOI - 10.21512/commit.v13i1.5177
Subject(s) - machine learning , algorithm , computer science , reliability (semiconductor) , artificial intelligence , artificial neural network , process (computing) , decision tree , production (economics) , productivity , support vector machine , power (physics) , physics , quantum mechanics , economics , macroeconomics , operating system
This research aims to improve the productivity and reliability of incoming orders in the manufacturing process. The unclassified data attributes of the incoming order can affect the order plan which will impact to the low productivity and reliability in the manufacturing process. In order to overcome the problem, machine learning algorithms are implemented to analyze the data and expected to help the manufacturing process in deciding the incoming order arrangement process. Four machine learning algorithms are implemented (Decision Tree, Nave Bayes, Support Vector Machine, and Neural Network). These machine learning algorithms are compared by their algorithm performance to the manufacturing process problem. The result of the research shows that machine learning algorithms can improve the productivity and reliability rate in production area up to 41.09% compared to the previous rate without any dataset arrangement before. The accuracy of this prediction test achieves 97%.