xpertsystems/oil031-sample

OIL-031 是一个合成航运与物流数据集样本，专为油轮路线优化、AIS（自动识别系统）分析、运费率预测、港口拥堵机器学习、滞期费预测、瓶颈点风险建模和航次效率分类而设计。该样本包含12个表格，共156,285行数据，覆盖250艘船舶、500条航线、2,500个航次，模拟180天的操作，涉及6种油轮类型（VLCC、苏伊士型、阿芙拉型、LR2、MR、灵便型）。数据集基于行业标准物理原理生成，包括哈弗辛大圆距离与海事路由因子、BIMCO装卸费率、Worldscale运费定价、7个真实EIA瓶颈点（含实际交通份额）、特征耦合延迟分解和效率分级。数据表格涵盖船舶主数据、航线主数据、航次事件、货物移动、港口操作、港口拥堵、航运延迟、运费率、滞期费成本、天气干扰、瓶颈点事件和物流标签等，适用于多种机器学习任务，如分类、回归、时间序列预测和关系型学习。数据集经过校准，使用10个指标验证，符合BIMCO、INTERTANKO、Worldscale Association等行业标准，确保真实性和实用性。OIL-031 is a synthetic shipping and logistics dataset sample tailored for tanker route optimization, AIS (Automatic Identification System) analysis, freight rate forecasting, machine learning-driven port congestion analysis, demurrage prediction, bottleneck risk modeling, and voyage efficiency classification. This dataset sample comprises 12 tables with a total of 156,285 data rows, covering 250 vessels, 500 shipping routes, and 2,500 voyages, simulating 180 days of operational activities, and involving 6 tanker categories: VLCC, Suezmax, Aframax, LR2, MR, and Handysize. The dataset is generated in accordance with industry-standard physical principles, including Haversine great-circle distance and maritime routing factors, BIMCO loading and unloading rates, Worldscale freight pricing, 7 real EIA bottleneck points with actual traffic shares, feature-coupled delay decomposition, and efficiency grading. The included data tables cover vessel master data, route master data, voyage events, cargo movements, port operations, port congestion status, shipping delays, freight rates, demurrage costs, weather disturbances, bottleneck point events, and logistics tags, making it suitable for diverse machine learning tasks such as classification, regression, time series forecasting, and relational learning. The dataset has been calibrated and validated using 10 metrics, and complies with industry standards including BIMCO, INTERTANKO, and the Worldscale Association, ensuring its authenticity and practical utility.