面向航天产品的工艺规划与车间调度协同优化系统与应用验证数据集

本数据集基于2022年11月至2025年10月在典型航天产品制造企业实际研发与试制环境中，通过企业工业互联网平台、PDM及MES系统等多源集成获取的工艺、任务与设备状态数据构建而成，聚焦于航天产品制造过程中工艺规划与车间调度协同优化问题。数据集核心内容包括基于时序关系学习的资源状态预测方法、基于正排/倒排的建模方法、面向多约束环境的协同优化调度算法、以及集成了三者的面向航天产品的工艺规划与车间调度协同优化系统。数据来源于航天结构件等典型产品的试制产线，涵盖从工艺设计、任务下发、资源分配的全流程，并在高性能计算平台上基于Python/Java开发环境完成数据处理与模型训练，所有算法与代码均通过严格验证确保可复现性。数据集以结构化表格、时序序列及系统日志等多种形式存储，完整呈现了从多源数据处理到协同决策生成的闭环信息流程。通过与传统企业内部的排产调度方法的对比，验证了本数据集所支持方法在缩短制造周期、提升资源利用率等方面的显著优势。本数据集为航天制造领域智能调度系统的研发、验证与应用提供了高质量的数据基础与实验平台，对推动航天产品制造向柔性化、智能化、高效化转型具有重要的理论研究价值与工程实践意义。数据量为12.5MB。

This dataset is constructed from process, task and equipment status data collected via multi-source integration through the enterprise industrial internet platform, PDM and MES systems in the actual R&D and trial-manufacturing environment of a typical aerospace product manufacturing enterprise from November 2022 to October 2025, focusing on the collaborative optimization problem of process planning and workshop scheduling in aerospace product manufacturing processes. The core contents of this dataset include a resource status prediction method based on temporal relation learning, a modeling method based on forward/backward scheduling, a collaborative optimal scheduling algorithm for multi-constraint environments, and an aerospace product-oriented process planning and workshop scheduling collaborative optimization system integrating the three aforementioned methods. The data is sourced from trial-manufacturing production lines of typical products such as aerospace structural components, covering the full process from process design, task distribution to resource allocation. Data processing and model training were completed on a high-performance computing platform with Python/Java development environments, and all algorithms and codes have undergone strict verification to ensure reproducibility. Stored in various forms such as structured tables, time series and system logs, this dataset fully presents the closed-loop information flow from multi-source data processing to collaborative decision generation. Compared with traditional internal enterprise scheduling methods, the methods supported by this dataset have been verified to have significant advantages in shortening manufacturing cycles and improving resource utilization rates. This dataset provides a high-quality data foundation and experimental platform for the R&D, verification and application of intelligent scheduling systems in the aerospace manufacturing field, and holds important theoretical research value and engineering practice significance for promoting the transformation of aerospace product manufacturing towards flexibility, intelligence and efficiency. The data volume of this dataset is 12.5 MB.