This study proposes an effective dynamic OHT routing approach to handle a substantial volume of wafer transport in a modern large semiconductor fabrication plant. The proposed approach aims to overcome challenges faced by previous approaches whose applicability is limited in conditions where the underlying distribution of traffic conditions varies over a short period. The approach comprises two models to explicitly approximate the congestion-aware travel times of different parts of the candidate route based on the current rail conditions, to evaluate the traffic conditions when routing. First, the local path approximation model heuristically evaluates the travel time of paths within a short range. Second, the global path approximation model evaluates the travel time of a distant range using a deep neural network. The simulation experiments show that the proposed approach outperforms the benchmark algorithms regarding delivery time and throughput, exhibiting 11.34% lower delivery time compared to a reinforcement-learning-based benchmark model. The proposed approach successfully integrates environmental information to evaluate congestion in a complex fab and optimize the routes of a large fleet of OHTs while balancing the traffic throughout a dynamic system.

This study proposes an effective dynamic OHT routing approach to handle a substantial volume of wafer transport in a modern large semiconductor fabrication plant. The proposed approach aims to overcome challenges faced by previous approaches whose applicability is limited in conditions where the underlying distribution of traffic conditions varies over a short period. The approach comprises two models to explicitly approximate the congestion-aware travel times of different parts of the candidate route based on the current rail conditions, to evaluate the traffic conditions when routing. First, the local path approximation model heuristically evaluates the travel time of paths within a short range. Second, the global path approximation model evaluates the travel time of a distant range using a deep neural network. The simulation experiments show that the proposed approach outperforms the benchmark algorithms regarding delivery time and throughput, exhibiting 11.34% lower delivery time compared to a reinforcement-learning-based benchmark model. The proposed approach successfully integrates environmental information to evaluate congestion in a complex fab and optimize the routes of a large fleet of OHTs while balancing the traffic throughout a dynamic system.

Existing reinforcement learning (RL) methods have limited applicability to real-world industrial control problems because of their various constraints. To overcome this challenge, in this article, we devise a novel RL method to enable the optimization of a policy while strictly satisfying the system constraints. By leveraging a value-based RL approach, our proposed method is not limited by the challenges faced when searching a constrained policy. Our method has two main features. First, we devise two distance-based Q-value update schemes, incentive and penalty updates, which enable the agent to decide on controls in the feasible region by replacing an infeasible control with the nearest feasible continuous control. The proposed update schemes can adjust the values of both continuous and original infeasible controls. Second, we define the penalty cost as a shadow price-weighted penalty to achieve efficient, constrained policy learning. We apply our method to the microgrid control, and the case study demonstrates its superiority.

Traffic signal control is one of the most important challenges for resolving traffic congestion in modern cities. Many previous studies that attempted to solve this problem were based on a simulation optimization approach. However, these studies focused on the optimization procedure without considering two fundamental limitations of simulation: designing a simulation environment and running a simulator during optimization. To overcome these limitations, this study proposes a simulationoptimization framework using a deep-neural-network-based metamodel. The proposed framework can be summarized in five steps: (1) parameterizing real-world features, such as road networks and traffic demands; (2) constructing sample intersections as simulation environments; (3) gathering training data by running a simulator; (4) fitting the metamodel with the data to predict the output of the simulator; and (5) immediately producing the optimized traffic signal via the trained model whenever a new real instance is provided. Based on numerical experiments, we evaluated the performance of the proposed framework in terms of the prediction accuracy of the metamodel and optimized the solution quality. The experiments show that the proposed framework is comparable with the existing simulation optimization framework in terms of solution quality; however, it significantly reduces the computation time. The experiments also verify that its performance is robust for unseen road networks and traffic demands. Moreover, the metamodel can easily accommodate variants of problems with multi-objective functions or additional constraints, such as the minimum duration of pedestrian crosswalks and fixed cycle lengths.

• Korea’s carbon neutrality requires a pivotal role of the power sector in balancing the dual challenges of rapidly growing electricity demand and full decarbonization . • A renewable-dominated power system consistently emerges as a key strategy across various models. • Achieving rapid decarbonization in the power sector presents significant technological and economic challenges. • Carbon capture and storage (CCS) and nuclear power are essential to addressing these challenges. • Current energy and climate policies fall short of meeting the mid-century mitigation target. In October 2021, Korea announced its mid-century carbon mitigation target of achieving carbon neutrality by 2050, reaffirming its commitment by enhancing its 2030 Nationally Determined Contribution (NDC). This study employs six energy-economic and integrated assessment models to explore net-zero emission pathways and strategies for Korea’s power sector, while assessing the associated costs and challenges. The findings underscore the complexity and urgency of this transition, with the power sector playing a pivotal role in balancing the dual challenges of rapidly growing electricity demand and full decarbonization. A shift toward a renewable-dominated power sector emerges as a robust strategy, though it poses unprecedented technological and economic challenges. Large-scale low-carbon technologies, such as carbon capture and storage (CCS) and nuclear power, are identified as crucial solutions to reduce reliance on variable renewable energy sources and mitigate associated costs. Additionally, the study finds that current energy and climate policies are insufficient to meet the mid-century mitigation target, highlighting the urgent need for policy enhancements to bridge the gap and ensure the feasibility of Korea’s carbon neutrality goal.

The advancement of 2.5-D integration technology has proven to be an efficient means of combining heterogeneous chiplets to achieve advanced design objectives. In the context of 2.5-D integration, especially in high-frequency systems, minimizing timing skew among multiple nets has become imperative. Considering this, the deviation of the wirelength for numerous nets must be constrained within tight bounds to ensure the optimal performance of 2.5-D systems. The interposer used in 2.5-D integration offers enhanced flexibility in octagonal wire routability and interlayer transitions; however, it introduces the challenge of minimizing routing bends to enhance both manufacturability and signal integrity. In our study, we first addressed an interposer routing problem, minimizing both wirelength and bends in octagonal routable interposer design, while adhering to constraints on the length-deviation limits among multiple nets. We formulated a mixed-integer linear programming (MILP) model by considering the requirements of the routing problem. However, the model has several challenges, and the routing problem is NP-hard, indicating that no exact polynomial time method may exist. Consequently, we proposed an efficient routing algorithm that utilizes an assignment heuristic and bidirectional shortest path merge technique. Our algorithm provides a near-optimal solution in terms of the total wirelength and number of bends within an acceptable computation time. As a result, a design task that would have taken a month to complete was finished in less than two days. Experiments on real-world test cases verified that the proposed algorithm consistently delivers satisfactory solutions.