Addressing the dynamic flexible job shop scheduling (DFJSS) problem to generate a high-quality rescheduling scheme has gained increasing significance in various applications in recent years, such as the intelligent manufacturing system for aluminum profiles. In the intelligent manufacturing system for aluminum profiles, random events such as machine faults and variable processing times significantly affect production completion times. However, existing methods for solving DFJSS face challenges in generating high-quality rescheduling schemes in real time. To address these issues, a search strategy that combines graph neural networks (GNNs) and neighborhood search features is employed in the Monte Carlo tree search (MCTS) algorithm (GNS-MCTS) for real-time solving of the DFJSS. The search strategy in the GNS-MCTS aims to build a model that analyzes the scheduling disjunctive graph and predicts the searching probability, guiding the MCTS to important locations in the solution space rather than searching unimportant areas and wasting computing time. Specifically, the search strategy of GNS-MCTS incorporates encoders that fuse neighborhood search features and graph embedding vectors (GEVs) to improve the quality of the searching probability and make decisions about optimizing the scheduling disjunctive graph by breaking low-importance node pairs and reallocating low-importance nodes to other machines. Experimental results indicate that GNS-MCTS surpasses baseline algorithms across various problem sizes and real-time constraints, significantly enhancing computational efficiency and solution quality for the DFJSS. Further ablation studies and analysis reveal the impact of neighborhood search features on the enhancement of the learning-based search strategy when combined with neighborhood search for real-time DFJSS solutions.

As manufacturing processes continue to evolve, distributed and energy-efficient flexible manufacturing systems have become the central paradigm of intelligent manufacturing. To effectively meet the demands of intricate production scheduling requirements, it is crucial to take into account practical factors, such as two-stage assembly production and finite transportation resources. However, this imposes greater demands on the efficiency of scheduling algorithms. In this paper, a knowledge-based bi-hierarchical optimization algorithm (KBOA) is specially designed for solving the distributed energy-efficient assembly flexible job shop problem with transportation constraints (DEAFJSP-T). The objectives of minimizing makespan and total energy consumption are simultaneously taken into account. The search process of the KBOA is structured into two distinct modes. In the first mode, single-objective searches are conducted along several specified directions. To elevate the overall quality of the population, knowledge-based initialization strategies are employed, tailored for different objectives. Furthermore, the knowledge-based local search is designed to strengthen the exploitation of the KBOA by leveraging the interconnections among subproblems. Incorporating the search results from the first mode, the multi-objective search is conducted based on Pareto dominance in the second stage. Considering the characteristics of the problem, individual cooperation is employed to enhance the search efficiency of the KBOA. Extensive experiments are conducted to assess the performances of the KBOA. The statistical comparisons demonstrate that the KBOA is superior in solving the DEAFJSP-T in terms of efficiency and effectiveness.Note to Practitioners—Drawing inspiration from a real-world case in an electrical machinery equipment factory, we have observed the implementation of distributed energy-efficient flexible manufacturing with assembly and transportation (DEAFJSP-T). In this practical scenario, we must concurrently optimize two conflicting objectives of time and energy consumption, taking into account the intricate coupling relationship among processing, assembly and transportation, and satisfying complex constraints. Traditional scheduling methods often fall short in directly resolving such problems. Consequently, this research holds significant practical application value and academic importance. Given the distinctive nature of the problem, we have introduced a bi-hierarchical optimization approach to reduce redundant searches while retaining critical Pareto front information. To enhance the overall quality of the population, we have developed several knowledge-based heuristics for initialization. Additionally, we have devised problem property-driven individual cooperation and knowledge-based local search to augment the exploration for non-dominated solutions and strike a balance between exploration and exploitation. Our experimental results demonstrate the superior performance of our approach, surpassing state-of-the-art algorithms. In essence, this article presents a bi-hierarchical optimization algorithm infused with problem-specific knowledge to tackle an intricate scheduling problem derived from real-world scenarios, namely DEAFJSP-T Looking ahead, we aim to address scheduling in uncertain scenarios and explore the integration of machine learning into optimization methods.

In this paper, by taking a locally active memristor (LAM) as the neuronal synapse to link two identical/disparate ReLU-type Hopfield neural networks (RHNNs), the memristive homogeneous and heterogeneous neural networks are presented and their coupling dynamics are discussed in succession. The homogeneous RHNN model consists of two 3D RHNNs with a LAM, in which the coupling strength-and initial value-induced synchronous transitions are revealed. Besides, the heterogeneous RHNN model is focused on which is constructed by using a LAM to couple a 3D and 2D RHNN, where complex and rich dynamics are revealed in numerical, including hyperchaos, chaos, quasi-period, and multi-stable patterns. Particularly, attractor control of offset boosting as well as color image encryption application are realized, indicating the high controllability and security of the LAM-coupled neural networks. Finally, the electrical neuron depending on the digital circuit is implemented and hardware experimental results verify the numerical measurements well.

Deterministic processing time are no longer applicable under realistic circumstances because of the uncertainties involved in manufacturing and production processes. The present study aims to address a multiobjective distributed assembly flexible job shop scheduling problem with type-2 fuzzy time (DAT2FFJSP), focusing on the optimization objectives of minimizing the makespan and total energy consumption. To address this problem, a mixed-integer linear programming model is presented. Then, a population-based iterative greedy algorithm (PBIGA) with a Q-learning mechanism is proposed, which possesses the following characteristics: 1) a hybrid initialization method is used to generate the population; 2) six local search operators, crossover operators, and mutation operators are applied to explore and exploit the solution space; and 3) the Q-learning mechanism intelligently utilizes historical information on the success of local search operator updates to determine the most suitable perturbation operator; and 4) an energy-saving strategy is applied to improve the candidate solutions. Finally, the effectiveness of the proposed components is validated through extensive experiments that are conducted on 30 instances. The PBIGA outperforms the state-of-the-art algorithms on the DAT2FFJSP.

Learning differential evolution (DE) algorithms are widely adopted to address flexible job-shop scheduling problems (FJSPs) because of the optimization ability. However, traditional learning DEs are not sufficient to develop the feature information of the problem. In this article, a feature-based learning DE algorithm (FLDE) is proposed to address FJSP considering worker health. Occupational repetitive actions index (OCRA) is an indicator that describes the degree of worker fatigue. The OCRA is utilized to ensure the feasibility of scheduling solutions generated by FLDE. A feature-based decision model (FDM) is designed to select the appropriate optimization operator for a scheduling solution. A critical operation search method is introduced to extract feature information from the scheduling solution. Experimental results reveal that FDM is critical to improving the local optimization ability of FLDE, and that FLDE outperforms the comparison algorithms on 40 problem instances.

The production process of steelmaking continuous casting (SCC) is a typical heterogeneous distributed manufacturing system. The scheduling problem in heterogeneous distributed manufacturing systems is a complex combinatorial optimization problem. In this article, the energy-aware distributed heterogeneous flexible flow shop scheduling problem (EADHFFSP) with variable speed constraints is studied with objectives, including total tardiness (TTD) and total energy consumption (TEC). A mixed-integer linear programming (MILP) model is constructed for the EADHFFSP. A co-evolution algorithm with dueling reinforcement learning mechanism (DRLCEA) is presented to address EADHFFSP. In DRLCEA, a knowledge-based hybrid initialization operation is proposed to generate the initial population of the problem. A global search based on adversarial generative learning is designed to search the solution space. The dueling double deep Q-network (DDQN) is applied to select the operator for the local search. A speed adjustment strategy and an energy-saving strategy based on knowledge are proposed to reduce TTD and TEC of the EADHFFSP with regard to the properties of EADHFFSP. The results of experiments show that the performance of DRLCEA is superior to certain state-of-the-art comparison algorithms in solving EADHFFSP.

This paper discusses a strategy for estimating Hammerstein nonlinear systems in the presence of measurement noises for industrial control by applying filtering and recursive approaches. The proposed Hammerstein nonlinear systems are made up of a neural fuzzy network (NFN) and a linear state`-space model. The estimation of parameters for Hammerstein systems can be achieved by employing hybrid signals, which consist of step signals and random signals. First, based on the characteristic that step signals do not excite static nonlinear systems, that is, the intermediate variable of the Hammerstein system is a step signal with different amplitudes from the input, the unknown intermediate variables can be replaced by inputs, solving the problem of unmeasurable intermediate variable information. In the presence of step signals, the parameters of the state-space model are estimated using the recursive extended least squares (RELS) algorithm. Moreover, to effectively deal with the interference of measurement noises, a data filtering technique is introduced, and the filtering-based RELS is formulated for estimating the NFN by employing random signals. Finally, according to the structure of the Hammerstein system, the control system is designed by eliminating the nonlinear block so that the generated system is approximately equivalent to a linear system, and it can then be easily controlled by applying a linear controller. The effectiveness and feasibility of the developed identification and control strategy are demonstrated using two industrial simulation cases.

As one representative of last-mile logistics in intelligent transportation systems, the on-demand food delivery (OFD) service has gained rapid market growth but also faces multiple challenges. One of the critical issues is the order dispatching problem (ODP) with an NP-hard nature, which refers to dispatching a large number of orders to riders reasonably in real time with very limited decision time. To address the ODP, this paper proposes an optimization algorithm based on graph neural networks (GNN) by combining the advantages of machine learning (ML) techniques and operational research (OR) methods: 1) The ML component learns to reduce the solution space by filtering out inappropriate riders for each order, handling the large-scale complexity of ODP. Specifically, we present a rider modeling approach by using GNN to better characterize rider information; besides, two attention mechanisms are designed to adaptively learn the matching relationship between riders and orders. 2) The OR component ensures the solution quality with a greedy and regret value-based dispatching heuristic. Extensive experiments are conducted on real-world datasets to evaluate the performance of the proposed method by comparing it with other existing models and algorithms. The results show that the design of our ML model is effective in yielding better prediction results, and the proposed GNN-based optimization algorithm can effectively and efficiently solve the ODP by improving delivery efficiency and customer satisfaction.