多品类应急物资日均供需量预测模型数据集

研究物资物流供应保障感知预警问题，设计基于深度自注意力网络的多种类细粒度需求预测模型。本数据集包含北京和郑州两地的多个快递站点的需求和供应的订单数据，快递站点的距离数据，路径相似度数据以及相邻程度的数据，同时还包含快递站点间的用户相似度数据、快递站点发生的突发事件的数量，以及每天发生的新闻数量以及当天的节日类型。数据采集自2021.01.01至2022.08.31共608天北京282个快递站点以及2021.01.01至2022.12.31共730天郑州73个快递站点，数据量60MB。

This research addresses the perceptual early warning problem in material logistics supply support, and proposes a multi-type fine-grained demand forecasting model based on deep self-attention networks. This dataset includes order data of demand and supply from multiple courier stations in Beijing and Zhengzhou, distance data between courier stations, path similarity data, adjacency degree data, user similarity data between courier stations, the number of emergencies occurring at each courier station, the daily number of news reports, and the type of holiday on each day. The data was collected over two periods: from January 1, 2021 to August 31, 2022 (608 days in total) for 282 courier stations in Beijing, and from January 1, 2021 to December 31, 2022 (730 days in total) for 73 courier stations in Zhengzhou. The total size of this dataset is 60 MB.

搜集汇总

背景与挑战

背景概述

该数据集包含北京和郑州两地多个快递站点的订单数据、距离数据、路径相似度数据等，用于研究物资物流供应保障感知预警问题。数据采集自2021年至2022年，涵盖608天北京282个站点和730天郑州73个站点，总数据量60MB，支持需求预测和时空特征分析。