An improved deep reinforcement learning approach: a case study for optimisation of berth and yard scheduling for bulk cargo terminal

URN:NBN:SI:doc-6VN6YA832023Ai, T.Huang, H. F.Huang, L.Jiao, F.Ma, W. G.Song, R. J.documents/doc/6/URN_NBN_SI_doc-6VN6YA83_001.pdfdocuments/doc/6/URN_NBN_SI_doc-6VN6YA83_001.txt183articlestr. 303-31610.14743/apem2023.3.4741854-6250268736003URN:NBN:SI:doc-6VN6YA83engFakulteta za strojništvo, Inštitut za proizvodno strojništvoAdvances in production engineering and managementBYalgoritmibulk cargo terminaldeep reinforcement learningdouble deep Q-networkgloboko okrepljeno učenjekontejnerski terminaliMarkov decision process (MDP) modelnalaganjeoptimisationoptimizacijaprioritised experience replay and softmax strategy-based duelingpristaniščarazlaganjeschedulingAn improved deep reinforcement learning approach: a case study for optimisation of berth and yard scheduling for bulk cargo terminal