This work investigates the space-limited aircraft assembly scheduling problem (SAASP) based on real-world cases. A computational model, minimizing the makespan, is developed to formulate the complex operational relationships and space constraints of the workstations in SAASP. To address SAASP, we propose a reinforcement and statistical learning-assisted evolutionary algorithm (RSLEA). First, hierarchical encoding and queue decoding methods are designed to capture the intricate sequencing and workstation space constraints. Second, a statistical learning strategy is implemented to accelerate the convergence of early exploration. Next, a reinforcement learning strategy is introduced to control the sampling size of the statistical learning process. Additionally, another reinforcement evolutionary learning strategy is developed for exploitation. Three crossover operators are employed, with their computational resources adaptively allocated by the agent. To validate the performance of the proposed algorithm, RSLEA is tested on 520 instances and three real-world cases from a Chinese aircraft manufacturing factory. These cases are the assembly workstations of wings, partial bodies, and tails of aircraft, which are critical and complex. The RSLEA is compared with state-of-the-art algorithms and obtains better performance on 89% instances, and reduces the average gap to the baseline by 2.72%. Moreover, RSLEA achieves the best solution and overall performance in real-world cases.

With the increasing demand for scheduling large-scale Earth Observation Satellites (EOS), traditional scheduling models are no longer sufficient to meet the requirements for flexibility and order-driven operations. To address this challenge, this paper proposes a Large Language Model (LLM)-based Bi-level Online Order Dispatching and Scheduling (LLM-BODS) algorithm. The proposed algorithm consists of three key components: a bi-level online optimization framework, an LLM-based order dispatching algorithm, and a metaheuristic algorithm for single-satellite order scheduling. Within the bi-level optimization framework, the upper level dynamically allocates orders to satellites based on real-time order demands and satellite resource states using the LLM-based order dispatching algorithm. The lower level employs an adaptive large neighborhood search (ALNS) metaheuristic to generate optimal task execution sequences for each satellite. Experimental results demonstrate that the proposed algorithm not only maintains scheduling quality but also enhances scheduling efficiency, effectively addressing the flexible scheduling needs of large-scale Earth Observation Satellites.

The quadratic unconstrained binary optimization (QUBO) is a well-known NP-hard problem that takes an $n\times n$ matrix $Q$ as input and decides an $n$-dimensional 0-1 vector $x$, to optimize a quadratic function. Existing learning-based models that always formulate the solution process as sequential decisions suffer from high computational overload. To overcome this issue, we propose a neural solver called the Value Classification Model (VCM) that formulates the solution process from a classification perspective. It applies a Depth Value Network (DVN) based on graph convolution that exploits the symmetry property in $Q$ to auto-grasp value features. These features are then fed into a Value Classification Network (VCN) which directly generates classification solutions. Trained by a highly efficient model-tailored Greedy-guided Self Trainer (GST) which does not require any priori optimal labels, VCM significantly outperforms competitors in both computational efficiency and solution quality with a remarkable generalization ability. It can achieve near-optimal solutions in milliseconds with an average optimality gap of just 0.362\% on benchmarks with up to 2500 variables. Notably, a VCM trained at a specific DVN depth can steadily find better solutions by simply extending the testing depth, which narrows the gap to 0.034\% on benchmarks. To our knowledge, this is the first learning-based model to reach such a performance.

The agile earth observation satellite scheduling problem (AEOSSP) with time-dependent transition times is a complex combinational optimization problem that has emerged from the development of large-scale satellite management techniques. To address this problem, we propose a deep reinforcement learning-based construction model (DRL-CM) that consists of five parts: 1) a Markov decision process (MDP); 2) a feature engineering; 3) a constructive heuristic neural network (CHNN); 4) an RL training method; and 5) an evaluation system. Specifically, the CHNN comprises six modules containing three special components that we propose: a dynamic encoder, a dynamic global layer, and a two-stage attention layer. First, we build the MDP of the AEOSSP and the feature engineering with effective features required for decision-making. Second, we design the CHNN to function as the MDP policy and train it with an RL model. Finally, we propose a comprehensive evaluation system for the validation of our model. The experimental results indicate that the proposed DRL-CM outperforms the state-of-the-art algorithm in terms of both optimization speed and quality. In addition, the feature engineering and network architecture built in our model are verified to be effective in comprehensive experiments.

To address the large-scale and time-consuming multiple agile earth observation satellite (multi-AEOS) scheduling problems, this article proposes a data-driven parallel scheduling approach, which is composed of a probability prediction model, a task assignment strategy, and a parallel scheduling manner. In this approach, given the historical data of satellite scheduling, a prediction model is trained based on the cooperative neuro-evolution of augmenting topologies (C-NEAT) to predict the probabilities that a task will be fulfilled by different satellites. Driven by the probability prediction model, an assignment strategy is adopted for dividing the multi-AEOS scheduling problem into several single-AEOS scheduling subproblems, which can adaptively assign each task to the satellite with the highest predicted probability and greatly decrease the problem size. In a parallel manner, the single-AEOS scheduling subproblems are optimized, respectively, leading to an acceleration in the optimization efficiency of the original problem. Computational experiments indicate that the proposed approach presents better overall performance than other state-of-the-art methods within a very limited scheduling time. As the two main components of the proposed approach, the prediction model based on C-NEAT and the task assignment strategy also outperform other models with traditional training algorithms and inadaptive assignment strategies, respectively.

Satellite Imaging Scheduling, Satellite Downlinking Scheduling and Ground Resources Scheduling are important components in satellites daily management. Considering the highly interlinking of these three types of scheduling, an Integrated Satellite Scheduling model is formulated and proved NP-complete in this paper. To address the large scale and oversubscription of the Integrated Satellite Scheduling in an actual background, an evolvable real-time system of Integrated Satellite Scheduling is constructed based on Cooperative Neuro Evolution of Augmenting Topologies (C-NEAT). With the help of the C-NEAT, the system learns from historical scheduling data and adaptively assigns each request to the satellite or the ground antenna which is most likely to fulfill this request. Moreover, the real-time scheduling function of the system is actualized by the windowed scheduling framework. Experimental results indicate that the system greatly reduces the problem size of Integrated Satellite Scheduling and improves the scheduling efficiency, where daily and emergent requests are arranged over time.