Injection-Molded Polypropylene Parts: Experimental Process Dataset with Part Weights, Energy Consumption and CAD Geometry

Injection-Molded Polypropylene Parts: Experimental Process Dataset with Part Weights, Energy Consumption and CAD Geometry Overview This dataset contains experimental data from polypropylene parts produced by plastic injection molding under controlled variations of process parameters. The dataset is intended to support research in data-driven manufacturing, process modeling, surrogate modeling, multi-objective optimization, process variability analysis, energy efficiency, and quality assessment in plastic injection molding. For each injection batch, the dataset provides: Process parameters used during the injection cycle. Individual part-weight measurements for batch (5 injected parts by batch). Average part weight. Energy consumption. Cycle time. Qualitative viability label. CAD geometry of the injected parts and sprue system. The dataset includes one structured CSV dataset and a CAD file describing the injected part geometry together with the sprue. Associated publication. This dataset originates from the experimental campaign reported in the following peer-reviewed article, which should be consulted for the full description of the experimental setup, process variables, and methodology: Pérez-González, J., Sánchez-Calleja, I., Fernández-Gorgojo, A., Ferrero-Guillén, R., Martínez-Gutiérrez, A., & Díez-González, J. (in press). Robust Multi-Objective Optimization of Injection Molding Parameters: An Experimental SHAP-Guided Memetic NSGA-II Approach. The International Journal of Advanced Manufacturing Technology. DOI: [to be added upon publication]. Companion experimental work. The same 1000 injection batches and the same injected parts described in this dataset were also analyzed for surface defect characterization using robotic inspection and quantitative multidefect modeling. The corresponding image data (raw and labeled) and quantitative defect measurements are published in a separate Zenodo record associated with: Companion Zenodo dataset: 10.5281/zenodo.20322729 Sánchez-Calleja, I., Ferrero-Guillén, R., Martínez-Gutiérrez, A., Díez-González, J., & Perez, H. (in press). Explainable Closed-Loop Injection Molding Parameter Tuning via Robotic Inspection and Quantitative Multidefect Modeling. DOI: [to be added upon publication]. Both datasets share the same batch identifier (1–1000) and can be merged to obtain the complete experimental record — process parameters, weights, energy and cycle time on this side; surface defect quantification on the companion side. Users of this dataset are encouraged to read both articles before any reanalysis, and to consult the companion dataset when surface defect information is required. Experimental Setup Item Description Material SABIC 579S homopolymer polypropylene Manufacturing process Plastic injection molding with a MiniMat 60 injection machine, from Mateu & Solé Mold type Two-cavity multicavity mold Parts per injection cycle Two injected parts Part geometry Provided as CAD file CAD content Injected part geometry and sprue Dataset type Experimental process dataset The experiments were carried out using a two-cavity multicavity mold, producing two parts per injection cycle. The CAD file included in this dataset provides the geometry of the injected pieces together with the sprue, supporting interpretation, reuse, and potential simulation-based studies. Dataset Files /Dataset.csv/Parts_geometry.stp File Description Dataset.csv Unified process dataset containing injection molding parameters, individual part weights, average weight, energy consumption, cycle time, and viability label for each batch. Parts_geometry CAD model containing the injected part geometry and sprue (in mm). Experimental Design The dataset contains 1000 injection batches, each batch contains 5 injected parts. 3 prior parts were also injected and discarded before the 5 registered parts to account for process transitory stability. Batch range Design strategy Description 1–768 Design of Experiments (DoE) The dataset was constructed by varying the injection molding process parameters according to their predefined ranges and step values. 769–1000 Maximin sampling criterion Additional configurations were generated by maximizing the distance between nearest individuals. This strategy was used to avoid a perfectly regular grid of data and to improve the coverage of the process parameter space. The first part of the dataset follows a structured DoE based on the variation of the main injection molding parameters. The second part introduces additional parameter combinations selected using a maximin criterion, increasing the diversity of configurations and reducing the dependency on a perfectly squared parameter mesh. Unified Process Dataset The file Dataset.csv contains one row per injection batch. Each row includes the process configuration used during the injection cycle and the corresponding measured or derived process outputs. Structure of Dataset.csv Column Description Unit / Values batch Injection batch identifier - label_viability Qualitative viability label of the process/result:G: Feasable injected part, can contain defects.Y: Feasable injected part, but with very noticeable short shots.R: Unfeasible injected part, injection incomplete or too erratic. G,Y,R Tinj Injection temperature °C tinj Injection time s Pinj Injection pressure bar Ph Holding pressure bar Bp Back pressure bar th Holding time s P1_weight Individual weight measurement of part 1 g P2_weight Individual weight measurement of part 2 g P3_weight Individual weight measurement of part 3 g P4_weight Individual weight measurement of part 4 g P5_weight Individual weight measurement of part 5 g weight_avg Average part weight associated with the batch g energy Energy consumption associated with the batch Wh cycle_time Injection molding cycle time s Data Description The dataset combines process inputs and manufacturing outputs. Process Parameters Parameter Description Range of Values Step Value* Tinj Injection temperature [180-240] ºC 20ºC tinj Injection time [0.5 - 2] s 0.5 s Pinj Injection pressure [5 - 35] bar 10 bar Ph Holding pressure [5 - 35] bar 15 bar Bp Back pressure [15 - 30] bar 15 bar th Holding time [4 - 6] s 2 s *Step Values corresponding to the first 768 batches. Process Outputs Output Description Unit / Values P1_weight-P5_weight Individual part-weight measurements for the batch g weight_avg Average part weight computed from the available valid measurements g energy Energy consumption associated with the injection batch Wh cycle_time Total cycle time of the injection process s label_viability Qualitative viability label describing the process/result condition G,Y,R CAD Geometry The dataset includes a CAD file describing the injected geometry. Item Description Geometry type Injected polypropylene parts Included geometry Injected pieces and sprue Mold configuration Two-cavity multicavity mold Production output Two parts per injection cycle Purpose Geometry documentation, interpretation, reuse, and simulation support The CAD geometry allows users to relate the experimental process data to the physical shape of the injected component and the feeding system used during the injection process. Intended Use This dataset is suitable for: Data-driven modeling of plastic injection molding processes. Surrogate model development and validation. Multi-objective optimization of injection molding parameters. Analysis of part weight, energy consumption, and cycle time. Study of process variability and robustness. Quality assessment in injection-molded polypropylene parts. Benchmarking of machine learning and explainable AI methods in manufacturing. Educational and research purposes in manufacturing, polymer processing, and smart manufacturing. Citation If you use this dataset, please cite it as: Pérez-González, J., Sánchez-Calleja, I., Fernandez-Gorgojo, A., Ferrero-Guillén, R., Martínez-Gutiérrez, A., & Díez-González, J. (2026). Injection-Molded Polypropylene Parts: Experimental Process Dataset with Part Weights, Energy Consumption and CAD Geometry [Data set]. Zenodo. https://doi.org/10.5281/zenodo.20309380 License This dataset is released under the Creative Commons Attribution 4.0 International License. CC BY 4.0 Contact For questions about the dataset, please contact the corresponding author. Corresponding author: Jesús Pérez-GonzálezEmail: [e-mail]: 0009-0007-7683-6401 Funding This work was partially supported by the Spanish Research Agency (AEI) under grant number PID2023-153047OB100, by the Department of Education of the Regional Government of Castile and León, and the Universidad de León. The author Iván Sánchez-Calleja acknowledges funding for doctoral studies of the University of León.注塑成型聚丙烯零件：包含零件重量、能耗与CAD几何模型的实验过程数据集 ## 概述 本数据集收录了通过塑料注塑成型工艺制备的聚丙烯零件的实验数据，实验过程中对工艺参数进行了可控化调整。本数据集可支撑塑料注塑成型领域的数据驱动制造、工艺建模、代理模型构建、多目标优化、工艺波动分析、能效提升以及质量评估等相关研究。 针对每个注塑批次，数据集提供以下内容： - 注塑周期内使用的工艺参数 - 单批次5个注塑零件的单件重量测量值 - 平均零件重量 - 能耗 - 注塑周期时长 - 工艺/结果合格性标签 - 注塑零件与浇道系统的CAD几何模型 本数据集包含一份结构化CSV数据集，以及一份描述注塑零件几何形状与浇道的CAD文件。 ## 关联发表论文 本数据集源自以下同行评议论文中报道的实验研究，实验设置、工艺变量与研究方法的完整说明请参考该论文： > Pérez-González, J., Sánchez-Calleja, I., Fernández-Gorgojo, A., Ferrero-Guillén, R., Martínez-Gutiérrez, A., & Díez-González, J.（已录用待刊）. 《注塑成型参数的鲁棒多目标优化：基于实验SHAP引导的模因NSGA-II方法》. 《国际先进制造技术期刊》. DOI：[出版后补充] ## 配套实验研究 本数据集涉及的1000个注塑批次与注塑零件，同时通过机器人检测与定量多缺陷建模开展了表面缺陷表征分析。对应的原始与标注图像数据、定量缺陷测量结果已发布于另一项Zenodo数据集，相关信息如下： > 配套Zenodo数据集：10.5281/zenodo.20322729 > Sánchez-Calleja, I., Ferrero-Guillén, R., Martínez-Gutiérrez, A., Díez-González, J., & Perez, H.（已录用待刊）. 《基于机器人检测与定量多缺陷建模的可解释闭环注塑成型参数调优》. DOI：[出版后补充] 两个数据集共享相同的批次标识符（1~1000），可合并以获取完整的实验记录：本数据集包含工艺参数、重量、能耗与周期时长；配套数据集包含表面缺陷量化结果。鼓励本数据集使用者在开展二次分析前阅读两篇论文，若需要表面缺陷相关信息，请参考配套数据集。 ## 实验设置 | 项目 | 描述 | | ---- | ---- | | 材料 | SABIC 579S 均聚聚丙烯 | | 制造工艺 | 使用Mateu & Solé公司的MiniMat 60注塑机开展塑料注塑成型 | | 模具类型 | 两型腔多腔模具 | | 单次注塑周期产出零件数 | 2个注塑零件 | | 零件几何形状 | 以CAD文件形式提供 | | CAD内容 | 注塑零件几何形状与浇道 | | 数据集类型 | 实验过程数据集 | 实验采用两型腔多腔模具开展，单次注塑周期可产出2个零件。本数据集附带的CAD文件包含注塑零件与浇道的几何模型，可辅助实验结果解读、数据复用以及基于仿真的后续研究。 ## 数据集文件 `/Dataset.csv` `/Parts_geometry.stp` | 文件 | 描述 | | ---- | ---- | | Dataset.csv | 结构化工艺数据集，包含每个注塑批次的注塑工艺参数、单件重量测量值、平均重量、能耗、周期时长与合格性标签 | | Parts_geometry.stp | CAD模型，包含注塑零件与浇道的几何形状（单位：毫米） | ## 实验设计 本数据集包含1000个注塑批次，每个批次包含5个注塑零件。为消除工艺过渡阶段的不稳定性，在正式记录的5个零件前，会先注塑并弃置3个零件。 | 批次范围 | 设计策略 | 描述 | | ---- | ---- | ---- | | 1~768 | 试验设计（Design of Experiments，DoE） | 数据集通过按照预设范围与步长调整注塑工艺参数构建 | | 769~1000 | 最大最小采样准则 | 通过最大化最近邻样本间的距离生成额外参数配置，该策略可避免数据分布完全规则的网格状，提升工艺参数空间的覆盖度 | 数据集的前768个批次基于主注塑工艺参数的结构化试验设计构建，后232个批次引入了基于最大最小准则选择的额外参数组合，提升了配置多样性，降低了对规则参数网格的依赖。 ## 统一工艺数据集 `Dataset.csv`文件以每行对应一个注塑批次的格式存储，每行包含该批次使用的工艺配置，以及对应的测量或衍生工艺输出结果。 ### Dataset.csv文件结构 | 列名 | 描述 | 单位/取值范围 | | ---- | ---- | ---- | | "batch" | 注塑批次标识符 | 无 | | "label_viability" | 工艺/结果合格性标签：<br>G：可行注塑零件，可含缺陷；<br>Y：可行注塑零件，但存在明显缺料；<br>R：不可行注塑零件，注塑不完全或波动过大 | G, Y, R | | "Tinj" | 注塑温度 | °C | | "tinj" | 注塑时长 | s | | "Pinj" | 注塑压力 | bar | | "Ph" | 保压压力 | bar | | "Bp" | 背压 | bar | | "th" | 保压时长 | s | | "P1_weight" | 零件1的单件重量测量值 | g | | "P2_weight" | 零件2的单件重量测量值 | g | | "P3_weight" | 零件3的单件重量测量值 | g | | "P4_weight" | 零件4的单件重量测量值 | g | | "P5_weight" | 零件5的单件重量测量值 | g | | "weight_avg" | 批次平均零件重量 | g | | "energy" | 批次能耗 | Wh | | "cycle_time" | 注塑周期时长 | s | ## 数据说明 本数据集整合了工艺输入与制造输出数据。 ### 工艺参数 | 参数 | 描述 | 取值范围 | 步长* | | ---- | ---- | ---- | ---- | | "Tinj" | 注塑温度 | [180~240] °C | 20°C | | "tinj" | 注塑时长 | [0.5~2] s | 0.5s | | "Pinj" | 注塑压力 | [5~35] bar | 10bar | | "Ph" | 保压压力 | [5~35] bar |15bar | | "Bp" | 背压 | [15~30] bar |15bar | | "th" | 保压时长 | [4~6] s |2s | *步长对应前768个批次的参数设置。 ### 工艺输出 | 输出项 | 描述 | 单位/取值范围 | | ---- | ---- | ---- | | "P1_weight"~"P5_weight" | 批次单件重量测量值 | g | | "weight_avg" | 基于有效测量值计算的批次平均零件重量 | g | | "energy" | 注塑批次能耗 | Wh | | "cycle_time" | 注塑工艺总周期时长 | s | | "label_viability" | 描述工艺/结果状态的合格性标签 | G, Y, R | ## CAD几何模型 本数据集包含一份描述注塑零件几何形状的CAD文件。 | 项目 | 描述 | | ---- | ---- | | 几何类型 | 注塑聚丙烯零件 | | 包含几何 | 注塑零件与浇道 | | 模具配置 | 两型腔多腔模具 | | 生产产出 | 单次注塑周期产出2个零件 | | 用途 | 几何文档记录、结果解读、数据复用与仿真支撑 | CAD几何模型可帮助使用者将实验工艺数据与注塑零件的物理形状以及注塑过程中的进料系统建立关联。 ## 适用场景 本数据集适用于以下场景： 1. 塑料注塑成型工艺的数据驱动建模 2. 代理模型的开发与验证 3. 注塑成型参数的多目标优化 4. 零件重量、能耗与周期时长分析 5. 工艺波动与鲁棒性研究 6. 注塑成型聚丙烯零件的质量评估 7. 制造业中机器学习与可解释人工智能方法的基准测试 8. 制造、聚合物加工与智能制造领域的教学与研究用途 ## 引用方式 若使用本数据集，请按以下方式引用： > Pérez-González, J., Sánchez-Calleja, I., Fernández-Gorgojo, A., Ferrero-Guillén, R., Martínez-Gutiérrez, A., & Díez-González, J. (2026). 注塑成型聚丙烯零件：包含零件重量、能耗与CAD几何模型的实验过程数据集 [数据集]. Zenodo. https://doi.org/10.5281/zenodo.20309380 ## 许可协议 本数据集采用知识共享署名4.0国际许可协议（CC BY 4.0）发布。 ## 联系方式 若对本数据集有疑问，请联系通讯作者： - 通讯作者：Jesús Pérez-González - 邮箱：[e-mail] - ORCID：0009-0007-7683-6401 ## 资助信息 本研究得到西班牙研究署（AEI）项目PID2023-153047OB100、卡斯蒂利亚-莱昂自治区教育厅以及莱昂大学的部分资助。作者Iván Sánchez-Calleja感谢莱昂大学博士研究奖学金的资助。