TY - GEN
T1 - Optimized Task Planning of Transfer Robots Using Reinforcement Learning
AU - Park, Ji Whan
AU - Do Noh, Sang
N1 - Publisher Copyright:
© 2023, IFIP International Federation for Information Processing.
PY - 2023
Y1 - 2023
N2 - Industry 4.0, which has been actively studied in recent years, requires various digital transformation technologies. These technologies collect various data from processes and machines, provide useful information to the manufacturing system, and integrate them into the virtual world by synchronizing the physical manufacturing resources and their digital models. It has become a very important and necessary method to reflect the status of a manufacturing shop floor using data, analyzing it, and finding better approaches through simulations. In this study, a transfer robot used in various production lines was analyzed to determine the range of tasks. Generally, each transfer robot in a production line has a specific range of tasks, and task allocation has a significant impact on the overall productivity. In this study, the range of tasks performed by transfer robots was optimizes by applying reinforcement learning algorithms and simulations, and the result was applied to a real display panel production line. Based on the results, this study shows that the convergent application of data analytics, production simulation, and artificial intelligence algorithms contributes to increased productivity of production lines.
AB - Industry 4.0, which has been actively studied in recent years, requires various digital transformation technologies. These technologies collect various data from processes and machines, provide useful information to the manufacturing system, and integrate them into the virtual world by synchronizing the physical manufacturing resources and their digital models. It has become a very important and necessary method to reflect the status of a manufacturing shop floor using data, analyzing it, and finding better approaches through simulations. In this study, a transfer robot used in various production lines was analyzed to determine the range of tasks. Generally, each transfer robot in a production line has a specific range of tasks, and task allocation has a significant impact on the overall productivity. In this study, the range of tasks performed by transfer robots was optimizes by applying reinforcement learning algorithms and simulations, and the result was applied to a real display panel production line. Based on the results, this study shows that the convergent application of data analytics, production simulation, and artificial intelligence algorithms contributes to increased productivity of production lines.
KW - reinforcement learning
KW - simulation
KW - transfer robot
UR - https://www.scopus.com/pages/publications/85174445766
U2 - 10.1007/978-3-031-43670-3_41
DO - 10.1007/978-3-031-43670-3_41
M3 - Conference contribution
AN - SCOPUS:85174445766
SN - 9783031436697
T3 - IFIP Advances in Information and Communication Technology
SP - 591
EP - 602
BT - Advances in Production Management Systems. Production Management Systems for Responsible Manufacturing, Service, and Logistics Futures - IFIP WG 5.7 International Conference, APMS 2023, Proceedings
A2 - Alfnes, Erlend
A2 - Romsdal, Anita
A2 - Strandhagen, Jan Ola
A2 - von Cieminski, Gregor
A2 - Romero, David
PB - Springer Science and Business Media Deutschland GmbH
T2 - IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2023
Y2 - 17 September 2023 through 21 September 2023
ER -
