Smart manufacturing that involves tight integration of the physical system and cyber system is a hot topic in both industry and academia in the era of the Internet and big data. However, the dynamic and uncertain manufacturing environment introduces a significant adaptive issue of production scheduling, which is one of the pivotal tasks for smart manufacturing. This paper focuses on this problem and proposes a closed-loop adaptive scheduling solution based on the Cyber-Physical Production System (CPPS) with four phases: production data acquisition (PDA), dynamic disturbance identification (DDI), scheduling strategy adjustment (SSA), and schedule scheme generation (SSG). In the DDI phase, in view of the disturbance classification, a disturbance identification procedure based on CPPS monitoring is studied to ensure real-time response. In the SSA phase, an industrial big-data-driven scheduling strategy adjustment method is proposed, which consists of GA-based offline knowledge learning and KNN-based online adjustment, to enhance the system adaptability. We apply and verify the proposed adaptive scheduling solution on an experimental semiconductor manufacturing system, and the results demonstrate that the proposed method outperforms the dynamic scheduling method in terms of multiple objectives under different disturbance levels.
The dynamic scheduling problem of semiconductor manufacturing systems (SMSs) is becoming more complicated and challenging due to internal uncertainties and external demand changes. To this end, this paper addresses integrated release control and production scheduling problems with uncertain processing times and urgent orders and proposes a convolutional neural network and asynchronous advanced actor critic-based method (CNN-A3C) that involves a training phase and a deployment phase. In the training phase, actor–critic networks are trained to predict the evaluation of scheduling decisions and to output the optimal scheduling decision. In the deployment phase, the most appropriate release control and scheduling decisions are periodically generated according to the current production status based on the networks. Furthermore, we improve the four key points in the deep reinforcement learning (DRL) algorithm, state space, action space, reward function, and network structure and design four mechanisms: a slide-window-based two-dimensional state perception mechanism, an adaptive reward function that considers multiple objectives and automatically adjusts to dynamic events, a continuous action space based on composite dispatching rules (CDR) and release strategies, and actor–critic networks based on convolutional neural networks (CNNs). To verify the feasibility and effectiveness of the proposed dynamic scheduling method, it is implemented on a simplified SMS. The simulation experimental results show that the proposed method outperforms the unimproved A3C-based method and the common dispatching rules under the new uncertain scenarios.
23 June 2022 in IEEE Robotics and Automation Letters
Dynamic scheduling is one of the most important problems in the field of production scheduling. Existing ways to solve the problem are mainly based on experienced workers or automatic scheduling models (SMs). Because of the complementary advantages of workers and SMs, their combination has the potential to be a better solution. In this letter, a human-machine collaborative decision-making method based on confidence (HMCDM/C) is proposed. SMs are in charge of the automatic generation of decisions, and workers have the power to revise unreliable decisions. Furthermore, a threshold-based handover mechanism is proposed to determine when workers are involved in decision-making. Firstly, a measurement method is developed to quantify the confidence level of SM decisions. Secondly, an evaluation method is presented to determine the threshold of confidence levels, which can be used to discriminate whether the SM decisions are acceptable or not at a certain confidence level. Finally, several experiments are conducted in a smart workshop. The results show that the HMCDM/C can effectively coordinate workers with different experience levels and SMs, and has a very competitive performance.
22 February 2022 in International Journal of Computer Integrated Manufacturing
The data-driven method has been widely used to mine knowledge from the smart shop floors’ production data to guide dynamic scheduling. However, the mined knowledge may be invalid when the production scene changes. In order to address this problem and ensure the validity of the knowledge, this paper studies a data-driven scheduling knowledge life-cycle management (SKLM) method for the smart shop floor. The proposed method includes four phases: knowledge generation, knowledge application, online knowledge evaluation, and knowledge update. Specifically, the extreme learning machine (ELM) is applied to learn knowledge based on the composite scheduling rules. The quality control theory is used to evaluate the quality of scheduling knowledge. And the online sequential ELM (OS-ELM) is adopted to update the knowledge. Knowledge life-cycle management is implemented through the iterative knowledge update. The proposed method is validated on the MIMAC6, which is a simulation model of the semiconductor production line. Experimental results show that the proposed method could improve the effectiveness of scheduling knowledge and further optimize the performance of the smart shop floor.
29 November 2024 in Expert Systems with Applications