Distributed dynamics scheduling based reinforcement learning: importance and challenges
Conférence : Communications avec actes dans un congrès international
When it comes to scheduling choices inside complex
industrial systems, the dynamic job shop scheduling problem
(DJSSP) poses substantial difficulties. Deep learning, artificial
intelligence (AI), and reinforcement learning approaches have
all shown promising solutions in recent years to enhance the
effectiveness and performance of DJSSP systems. This study
provides a detailed analysis of the DJSSP literature, with an
emphasis on these cutting-edge methods. The review adopts
a rational methodology that includes an exhaustive search of
many databases from 1995 to 2023. Articles were chosen based
on predetermined qualifying criteria, taking into account their
applicability to the DJSSP and the methodology used. This paper
also examines how software design affects research, how industry
influences research, and the types of research outputs that are
disclosed. The findings provide valuable insights into the current
state of research and offer guidance for future advancements in
optimizing dynamic job-shop scheduling using advanced learning
techniques