【数据可视化-53】电子商务运输数据可视化分析

🧑 博主简介：曾任某智慧城市类企业算法总监，目前在美国市场的物流公司从事高级算法工程师一职，深耕人工智能领域，精通python数据挖掘、可视化、机器学习等，发表过AI相关的专利并多次在AI类比赛中获奖。CSDN人工智能领域的优质创作者，提供AI相关的技术咨询、项目开发和个性化解决方案等服务，如有需要请站内私信或者联系任意文章底部的的VX名片（ID：xf982831907）

💬 博主粉丝群介绍：① 群内初中生、高中生、本科生、研究生、博士生遍布，可互相学习，交流困惑。② 热榜top10的常客也在群里，也有数不清的万粉大佬，可以交流写作技巧，上榜经验，涨粉秘籍。③ 群内也有职场精英，大厂大佬，可交流技术、面试、找工作的经验。④ 进群免费赠送写作秘籍一份，助你由写作小白晋升为创作大佬。⑤ 进群赠送CSDN评论防封脚本，送真活跃粉丝，助你提升文章热度。有兴趣的加文末联系方式，备注自己的CSDN昵称，拉你进群，互相学习共同进步。

一、引言

  在电子商务蓬勃发展的今天，物流运输作为连接商家与消费者的关键环节，其效率和可靠性直接影响着客户体验和企业竞争力。本报告聚焦于某国际电子商务公司的运输数据，通过可视化分析，深入挖掘影响产品准时交付的因素，旨在为企业优化物流策略、提升客户满意度提供数据支持。

二、数据集介绍

  本次分析所使用的数据集涵盖了10999个观测样本，包含12个关键特征：

• ID: 客户唯一标识符
• Warehouse_block: 仓库区块（A、B、C、D、E）
• Mode_of_Shipment: 发货方式（船运、航空、公路）
• Customer_care_calls: 客户服务通话次数
• Customer_rating: 客户评分（1-5分）
• Cost_of_the_Product: 产品成本（美元）
• Prior_purchases: 历史购买次数
• Product_importance: 产品重要性（低、中、高）
• Gender: 客户性别
• Discount_offered: 折扣力度
• Weight_in_gms: 产品重量（克）
• Reached.on.Time_Y.N: 是否准时到达（0表示准时，1表示延迟）

2.1技术工具

• Python版本: 3.9
• 代码编辑器: Jupyter Notebook
• 数据处理库: pandas, numpy
• 可视化库: matplotlib

2.2 导入数据

  我们将使用pandas库来加载数据，并进行初步的预处理：

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from wordcloud import WordCloud
from collections import Counter
import string

# 导入数据
df = pd.read_csv('ecommerce_delivery.csv')

# 查看数据大小
print("数据大小:", df.shape)

# 查看数据基本信息
print("\n数据基本信息:")
print(df.info())
print("\n数据描述性统计:")
print(df.describe())

# 统计缺失值
print("\n缺失值统计:")
print(df.isnull().sum())

# 统计重复值
print("\n重复值数量:", df.duplicated().sum())

三、单变量分析

3.1 准时交付情况分布

plt.figure(figsize=(8, 8))
plt.pie(df['Reached.on.Time_Y.N'].value_counts(), labels=['准时', '延迟'], autopct='%1.1f%%', startangle=90)
plt.title('准时交付情况分布', fontsize=14, fontweight='bold')
plt.tight_layout()
plt.show()

  结果显示，约%59.7的订单能够准时交付，而%40.3的订单出现延迟。

3.2 发货方式分布

plt.figure(figsize=(10, 6))
plt.hist(df['Mode_of_Shipment'], color='skyblue', edgecolor='black')
plt.title('发货方式分布', fontsize=14, fontweight='bold')
plt.xlabel('发货方式', fontsize=12)
plt.ylabel('频率', fontsize=12)
plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

  船运运输占比最高，其次是公路和航空。

3.3 客户评分分布

import seaborn as sns

plt.figure(figsize=(10, 6))
sns.countplot(df['Customer_rating'], bins=5, color='lightgreen', edgecolor='black')
plt.title('客户评分分布', fontsize=14, fontweight='bold')
plt.xlabel('评分', fontsize=12)
plt.ylabel('频率', fontsize=12)
plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

四、多变量分析

4.1 发货方式与准时交付的关系

plt.figure(figsize=(12, 6))
shipment_mode = df['Mode_of_Shipment'].unique()
准时交付 = [df[(df['Mode_of_Shipment'] == mode) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for mode in shipment_mode]
延迟交付 = [df[(df['Mode_of_Shipment'] == mode) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for mode in shipment_mode]

plt.bar(shipment_mode, 准时交付, label='准时交付', color='skyblue')
plt.bar(shipment_mode, 延迟交付, bottom=准时交付, label='延迟交付', color='salmon')
plt.title('发货方式与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('发货方式', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.2 客户评分与准时交付的关系

plt.figure(figsize=(12, 6))
customer_rating = df['Customer_rating'].unique()
准时交付 = [df[(df['Customer_rating'] == rating) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for rating in customer_rating]
延迟交付 = [df[(df['Customer_rating'] == rating) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for rating in customer_rating]

plt.bar(customer_rating, 准时交付, label='准时交付', color='lightgreen')
plt.bar(customer_rating, 延迟交付, bottom=准时交付, label='延迟交付', color='orange')
plt.title('客户评分与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('客户评分', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.3 产品重要性与准时交付的关系

plt.figure(figsize=(12, 6))
importance_levels = df['Product_importance'].unique()
on_time_counts = [df[(df['Product_importance'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Product_importance'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]

plt.bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
plt.bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')
plt.title('产品重要性与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('产品重要性', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.4 客户服务通话次数与准时交付的关系

plt.figure(figsize=(12, 6))

importance_levels = df['Customer_care_calls'].unique()
on_time_counts = [df[(df['Customer_care_calls'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Customer_care_calls'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]

plt.bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
plt.bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')


# plt.scatter(df['Customer_care_calls'], df['Reached.on.Time_Y.N'], c=df['Reached.on.Time_Y.N'], cmap='bwr', alpha=0.5)
plt.title('客户服务通话次数与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('通话次数', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()

4.5 产品成本与准时交付的关系

plt.figure(figsize=(12, 6))

importance_levels = df['Cost_of_the_Product'].unique()
on_time_counts = [df[(df['Cost_of_the_Product'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Cost_of_the_Product'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]

plt.bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
plt.bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')


# plt.scatter(df['Customer_care_calls'], df['Reached.on.Time_Y.N'], c=df['Reached.on.Time_Y.N'], cmap='bwr', alpha=0.5)
plt.title('产品成本与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('产品成本（USD）', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()

4.6 历史购买次数与准时交付的关系

plt.figure(figsize=(12, 6))

importance_levels = df['Prior_purchases'].unique()
on_time_counts = [df[(df['Prior_purchases'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Prior_purchases'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]

plt.bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
plt.bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')


# plt.scatter(df['Customer_care_calls'], df['Reached.on.Time_Y.N'], c=df['Reached.on.Time_Y.N'], cmap='bwr', alpha=0.5)
plt.title('历史购买次数与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('历史购买次数', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()

plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.7 性别与准时交付的关系

plt.figure(figsize=(10, 6))
genders = df['Gender'].unique()
on_time_counts = [df[(df['Gender'] == gender) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for gender in genders]
delayed_counts = [df[(df['Gender'] == gender) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for gender in genders]

plt.bar(genders, on_time_counts, label='准时交付', color='lightblue')
plt.bar(genders, delayed_counts, bottom=on_time_counts, label='延迟交付', color='peachpuff')
plt.title('性别与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('性别', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.8 折扣力度与准时交付的关系

plt.figure(figsize=(12, 6))

importance_levels = df['Discount_offered'].unique()
on_time_counts = [df[(df['Discount_offered'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Discount_offered'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]

plt.bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
plt.bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')


# plt.scatter(df['Customer_care_calls'], df['Reached.on.Time_Y.N'], c=df['Reached.on.Time_Y.N'], cmap='bwr', alpha=0.5)
plt.title('折扣力度与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('历史购买次数', fontsize=12)
plt.ylabel('折扣力度', fontsize=12)
plt.legend()

plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

4.9 仓库区块与准时交付的关系

plt.figure(figsize=(12, 6))
blocks = df['Warehouse_block'].unique()
on_time_counts = [df[(df['Warehouse_block'] == block) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for block in blocks]
delayed_counts = [df[(df['Warehouse_block'] == block) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for block in blocks]

plt.bar(blocks, on_time_counts, label='准时交付', color='lightcyan')
plt.bar(blocks, delayed_counts, bottom=on_time_counts, label='延迟交付', color='lightpink')
plt.title('仓库区块与准时交付的关系', fontsize=14, fontweight='bold')
plt.xlabel('仓库区块', fontsize=12)
plt.ylabel('订单数量', fontsize=12)
plt.legend()
plt.grid(linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

五、多维度组合分析

fig, axes = plt.subplots(2, 2, figsize=(18, 12))

# 发货方式与准时交付
shipment_modes = df['Mode_of_Shipment'].unique()
on_time_counts = [df[(df['Mode_of_Shipment'] == mode) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for mode in shipment_modes]
delayed_counts = [df[(df['Mode_of_Shipment'] == mode) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for mode in shipment_modes]
axes[0, 0].bar(shipment_modes, on_time_counts, label='准时交付', color='skyblue')
axes[0, 0].bar(shipment_modes, delayed_counts, bottom=on_time_counts, label='延迟交付', color='salmon')
axes[0, 0].set_title('发货方式与准时交付', fontsize=12, fontweight='bold')
axes[0, 0].set_xlabel('发货方式', fontsize=10)
axes[0, 0].set_ylabel('订单数量', fontsize=10)
axes[0, 0].legend()
axes[0, 0].grid(linestyle='--', alpha=0.7)

# 客户评分与准时交付
ratings = df['Customer_rating'].unique()
on_time_counts = [df[(df['Customer_rating'] == rating) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for rating in ratings]
delayed_counts = [df[(df['Customer_rating'] == rating) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for rating in ratings]
axes[0, 1].bar(ratings, on_time_counts, label='准时交付', color='lightgreen')
axes[0, 1].bar(ratings, delayed_counts, bottom=on_time_counts, label='延迟交付', color='orange')
axes[0, 1].set_title('客户评分与准时交付', fontsize=12, fontweight='bold')
axes[0, 1].set_xlabel('客户评分', fontsize=10)
axes[0, 1].set_ylabel('订单数量', fontsize=10)
axes[0, 1].legend()
axes[0, 1].grid(linestyle='--', alpha=0.7)

# 产品重要性与准时交付
importance_levels = df['Product_importance'].unique()
on_time_counts = [df[(df['Product_importance'] == imp) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for imp in importance_levels]
delayed_counts = [df[(df['Product_importance'] == imp) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for imp in importance_levels]
axes[1, 0].bar(importance_levels, on_time_counts, label='准时交付', color='palegoldenrod')
axes[1, 0].bar(importance_levels, delayed_counts, bottom=on_time_counts, label='延迟交付', color='hotpink')
axes[1, 0].set_title('产品重要性与准时交付', fontsize=12, fontweight='bold')
axes[1, 0].set_xlabel('产品重要性', fontsize=10)
axes[1, 0].set_ylabel('订单数量', fontsize=10)
axes[1, 0].legend()
axes[1, 0].grid(linestyle='--', alpha=0.7)

# 性别与准时交付
genders = df['Gender'].unique()
on_time_counts = [df[(df['Gender'] == gender) & (df['Reached.on.Time_Y.N'] == 0)].shape[0] for gender in genders]
delayed_counts = [df[(df['Gender'] == gender) & (df['Reached.on.Time_Y.N'] == 1)].shape[0] for gender in genders]
axes[1, 1].bar(genders, on_time_counts, label='准时交付', color='lightblue')
axes[1, 1].bar(genders, delayed_counts, bottom=on_time_counts, label='延迟交付', color='peachpuff')
axes[1, 1].set_title('性别与准时交付', fontsize=12, fontweight='bold')
axes[1, 1].set_xlabel('性别', fontsize=10)
axes[1, 1].set_ylabel('订单数量', fontsize=10)
axes[1, 1].legend()
axes[1, 1].grid(linestyle='--', alpha=0.7)

plt.tight_layout()
plt.show()

六、总结

  通过对电子商务运输数据的全面可视化分析，我们得出了以下关键结论：

1. 航空运输的准时率最高：航空运输的准时交付率显著高于船运和公路运输，表明运输方式对交付效率有重要影响。
2. 客户评分与准时交付正相关：评分较高的客户更有可能收到准时交付的产品，表明服务质量与客户满意度密切相关。
3. 高重要性产品准时率低：标记为高重要性的产品反而拥有最低的准时交付率，这提示我们需要重新审视高优先级订单的物流流程。
4. 女性客户体验更优：女性客户的准时交付率略高于男性，可能与购买行为或产品类型有关。
5. 季节性影响显著：不同月份的准时交付率存在明显波动，需进一步分析季节性因素对物流效率的影响。

  这些发现为企业优化物流策略、提升客户体验提供了重要参考。通过针对性地改进运输方式、优先处理高重要性订单以及考虑季节性因素，企业能够显著提升准时交付率，增强市场竞争力。

  注： 博主目前收集了6900+份相关数据集，有想要的可以领取部分数据：