Scalable Scheduling of Semiconductor Packaging Facilities Using Deep Reinforcement Learning.

Reinforcement learning (RL) has emerged as a promising approach for scheduling semiconductor operations. Yet, it is still challenging to solve large-scale scheduling problems based on an RL method since learning complexity grows fast as the size of shop floor increases. This challenge becomes more apparent when solving the scheduling problems with a diverse number of job types, which leads to the difficulties in exploration and function approximation in RL. This article presents a scheduling method for semiconductor packaging facilities using deep RL in which an agent allocates a job to one of machines in a centralized manner. Specifically, a novel state representation is introduced to effectively accommodate the variations in the number of available machines and the production requirements. Furthermore, we propose a continuous representation of an action to maintain the size of the action space even when the numbers of jobs, machines, and operation types are subject to change. Extensive experiments on large-scale datasets demonstrate that the proposed method mostly outperforms the metaheuristics and rule-based methods, as well as the other RL approaches considered in terms of makespan while requiring much less computation time than the metaheuristics.

Effects of Nozzle Details on Print Quality and Hardened Properties of Underwater 3D Printed Concrete.

This study developed a 3D concrete printing (3DCP) system that can print not only in air but also underwater. This underwater 3DCP system is equipped with many distinct technologies, such as a technology to supply the printing material to the nozzle tip at a constant rate by detecting its amount in the printer hopper. Using the developed 3DCP system, the effect of nozzle details on underwater print quality and hardened properties was investigated. The straight-line printing performance underwater was evaluated using five nozzles: a nozzle without a trowel (Nozzel#1), a nozzle with fixed trowels attached to both sides (Nozzle#2), a nozzle with trowels attached to the back and both sides to constrain five sides (Nozzle#3), a nozzle with a three-sided trowel inclined by 30° (Nozzle#4), and a nozzle with a roof added to Nozzle#4 opening (Nozzle#5). Nozzle#4 yielded the best print quality and hardened properties. In addition, an underwater curved shape printing test was performed using Nozzle#4, the problems that occurred in this test were analyzed and solutions were suggested.