A Bayesian-based hyper-heuristic algorithm for the integrated scheduling of distributed production assembly and delivery problem

Q-learning-based hyper-heuristic algorithm for priority and precedence dual-driven task assignment in spatial crowdsourcing

Optimizing Flexible Multi-Compartment Location Routing Problem for Waste Collection with Priority of Service Using a Hyper-Heuristic Algorithm and ε–Constraint Method

Q-learning-based hyper-heuristic algorithm for open dimension irregular packing problems

Dual-tree genetic programming-based hyper-heuristic algorithm for dynamic task scheduling in astronomical observation satellite

A Q-learning-based multi-objective hyper-heuristic algorithm for energy-efficient integrated distributed hybrid flow-shop scheduling with preventive maintenance

A hyper-heuristic algorithm based on genetic and greedy strategy for university course scheduling problem

Learning-based hyper-heuristic algorithm for space-free multi-row facility layout problem

A multi-objective double Q-learning-based hyper-heuristic algorithm for aluminum production and transportation integrated scheduling problem

Energy learning hyper-heuristic algorithm for cooperative task assignment of heterogeneous UAVs under complex constraints

Adaptive Resource Allocation for Cloud-native Microservice via Meta-learning and Hyper-heuristic Algorithms

The complex scenario of interweaving multimodal transport and direct transport in the actual transport network poses challenges for optimising transport scheme under many-objectives. We establish a many-objective optimisation model for heterogeneous networks with the objectives of transit time connection, economies of scale, total transport network cost and carbon emissions. In terms of high-level strategy, selection strategies based on Genetic Algorithm (GA) and simulated annealing algorithm (SA), adaptive re-heating by Simulated Annealing (SAARH), and supplementary solution by Simulated Annealing (SASS) are incorporated as the receiving criteria, and seven low-level heuristic operators are designed from three aspects. Taking six different transportation tasks of twelve city node networks along the Yangtze River Economic Belt as an example, we draw the following conclusions: compared with GA and NSGA-II, the hyper-heuristic algorithm under SAARH strategy is feasible and effective; in different scenarios, heterogeneous transport networks will also generate new transport schemes, and it may not be as effective as expected to increase the railway market share by simply improving the railway travel speed; A more favorable cost coefficient of economies of scale can significantly reduce the overall cost of the transport network. Nevertheless, its impact on transit time connection fluctuates considerably while exerting minimal influence on network carbon emissions.

Bayesian networks (BNs) are effective and universal tools for addressing uncertain knowledge. BN learning includes structure learning and parameter learning, and structure learning is its core. The topology of a BN can be determined by expert domain knowledge or obtained through data analysis. However, when many variables exist in a BN, relying only on expert knowledge is difficult and infeasible. Therefore, the current research focus is to build a BN via data analysis. However, current data learning methods have certain limitations. In this work, we consider a combination of expert knowledge and data learning methods. In our algorithm, the hard constraints are derived from highly reliable expert knowledge, and some conditional independent information is mined by feature selection as a soft constraint. These structural constraints are reasonably integrated into an exponential Monte Carlo with counter (EMCQ) hyper-heuristic algorithm. A comprehensive experimental study demonstrates that our proposed method exhibits more robustness and accuracy compared to alternative algorithms.

In the context of a low-carbon economy, scientific methods to reduce carbon emissions have become an important issue for many ports. Carbon emissions in port areas mainly arise from vessels and handling equipment. Therefore, an effective resource assignment and equipment arrangement system could not only reduce carbon emissions, but also improve the port’s operational efficiency. This study considers factors such as the arrival order of container trailers, the cargo weight, and the number of container rehandling operations. The objective is to minimize the carbon emissions and the number of container rehandling operations in ports, for which a mixed-integer linear programming model is built. Both heuristic algorithms and hyper-heuristic algorithms are employed to optimize the container storage plan, and their applicability in storage optimization is compared. The results indicate that hyper-heuristic algorithms outperform heuristic algorithms in terms of solution quality and stability, effectively satisfying the storage requirements of the yard while minimizing the carbon emissions and the number of container rehandling operations. The results provide theoretical support for port enterprises in improving their operational efficiency and achieving their goals regarding low carbon emissions.

An intelligent hyperheuristic algorithm for the berth allocation and scheduling problem at marine container terminals

Deep-Q-network-enhanced aquila-equilibrium hyper-heuristic algorithm for preventive maintenance integrated disassembly line balancing involving worker redeployment

A Q-learning-based multi-objective hyper-heuristic algorithm with fuzzy policy decision technology

This study takes Huawei's supply chain as a case to reveal the mechanism of AI-driven supply chain reconfiguration based on multidisciplinary theories. Huawei has achieved breakthroughs such as reducing demand forecasting error rate to 8%, optimizing inventory turnover ratio by 25%, etc., through cognitive network architecture and multi-technology integration. Technologies such as spatiotemporal graph neural networks and adversarial learning have demon-strated significant effectiveness in demand perception and resilient supply chain construction. Reparameterization of production functions has reduced capacity costs by 22%, while hyper-heuristic algorithms have optimized logis-tics costs by 15%. The research proposes innovative frameworks such as data middleware platform and double-loop learning, where knowledge graphs have shortened project cy-cles by 40% and smart contracts have reduced transaction costs by 70%. At the governance level, a transparency framework has been established, federated learning balances data sovereignty, and ESG embedding has achieved a compli-ance rate of 95%. In the future, quantum machine learning may break through real-time optimization bottlenecks, with the synergy between humanistic AI and ethical governance being key to development. This study provides theoreti-cal insights for intelligent supply chain research and practical references for corporate transformation.

Traditional spacecraft task planning has relied on ground control centers issuing commands through ground-to-space communication systems; however, as the number of deep space exploration missions grows, the problem of ground-to-space communication delays has become significant, affecting the effectiveness of real-time command and control and increasing the risk of missed opportunities for scientific discovery. Adaptive Space Scientific Exploration requires that spacecraft have the ability to make autonomous decisions to complete known and unknown scientific exploration missions without ground control. Based on this requirement, this paper proposes an actor–critic-based hyper-heuristic autonomous mission planning algorithm, which is used for mission planning and execution at different levels to support spacecraft Adaptive Space Scientific Exploration in deep space environments. At the bottom level of the hyper-heuristic algorithm, this paper uses the particle swarm optimization algorithm, grey wolf optimization algorithm, differential evolution algorithm, and positive cosine optimization algorithm as the basic operators. At the high level, a reinforcement learning strategy based on the actor–critic model is used, combined with the network architecture, to construct a framework for the selection of advanced heuristic algorithms. The related experimental results show that the algorithm can meet the requirements of Adaptive Space Scientific Exploration, and exhibits a quality solution with higher comprehensive evaluation in the test. This study also designs an example application of the algorithm to a space engineering mission based on a collaborative sky and earth control system to demonstrate the usability of the algorithm. This study provides an autonomous mission planning method for spacecraft in the complex and ever-changing deep space environment, which supports the further construction of spacecraft autonomous capabilities and is of great significance for improving the exploration efficiency of deep space exploration missions.

To improve workload balance and reduce cycle time in disassembly line balancing problems, a multi-cycle work-sharing strategy that allows tasks to be reassigned across sub-cycles is proposed for the first time. While this strategy enhances disassembly efficiency, it also increases total resource demands, as different tasks may require distinct resources for effective execution. Therefore, to effectively address this challenge, a bi-objective integer linear programming model is established with the goal of minimising both overall cycle time and total resource count. The epsilon constraint method is then employed to obtain solutions for small-scale problems. For large-scale problems, a Double Deep Q-Network-based Hyper-Heuristic (DDQN-HH) algorithm is developed. This approach incorporates specialised encoding and decoding strategies, two state functions, and eight heuristic action rules to enhance its effectiveness. And comparative experiments using multi-objective evaluation metrics demonstrate the DDQN-HH algorithm's superiority over three other leading algorithms: the Deep Q-Network-based Hyper Heuristic (DQN-HH) algorithm, the Random, Greedy-based Hyper Heuristic (RG-HH) algorithm, and the Multi-objective Equilibrium Optimiser (MOEO) algorithm. Furthermore, the DDQN-HH's practical significance in addressing real-world engineering problems is validated through managerial applications.