R ggplot：在柱状图上添加周服务总量趋势线报错求助

报错原因 #

你设置的geom_line(aes(y= qty, group = interaction(service_type,date)))把数据拆成了每个服务类型+每个日期的独立分组，每组只有1个观测值。折线图需要同一组内至少有2个连续的点才能连线，所以ggplot抛出这个提示。

更关键的是，你的需求是展示每周服务总数量的趋势线，但原代码里用的qty是单个服务类型的数量，逻辑上根本不符合需求。

修正步骤与代码 #

1. 先计算每周的服务总数量 #

先对原数据按周分组，算出每周所有服务类型的总qty：

library(dplyr)
library(ggplot2)

# 原数据
sample <- data.frame(service_type = c("A","B","C","A","B","C","A","B","C","A","B","C","A","B","C","A","B","C","A","B","C"),
                     qty = c(38,185,87,29,12,133,2,14,31,2,9,59,60,43,137,135,31,159,15,32,1),
                     year_week = c("2022 - CW02","2022 - CW02","2022 - CW02","2022 - CW03","2022 - CW03","2022 - CW03","2022 - CW04","2022 - CW04","2022 - CW04","2022 - CW05","2022 - CW05","2022 - CW05","2022 - CW06","2022 - CW06","2022 - CW06","2022 - CW07","2022 - CW07","2022 - CW07","2022 - CW08","2022 - CW08","2022 - CW08"),
                     xlabel2 = c("Jan-22","Jan-22","Jan-22","Jan-22","Jan-22","Jan-22","Jan-22","Jan-22","Jan-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22","Feb-22")
)

sample <- sample %>%
  mutate(date = as.Date(paste0(year_week, 1), "%Y - CW%U%u")) 

# 计算每周总服务量
weekly_total <- sample %>%
  group_by(year_week, date) %>%
  summarise(total_qty = sum(qty), .groups = "drop")

2. 绘制带总趋势线的柱状图 #

用原数据画分类型柱状图，用汇总后的weekly_total画趋势线，同时调整折线位置避免被遮挡：

sample %>%
  ggplot() +
  # 绘制分服务类型的柱状图
  aes(x = date, fill = service_type, weight = qty) +
  geom_bar(position = "dodge") +
  # 添加每周总数量的趋势线（用汇总数据）
  geom_line(data = weekly_total, aes(x = date, y = total_qty, group = 1), 
            color = "black", size = 1, linetype = "solid") +
  # 可选：添加总数量的标签，偏移10避免遮挡
  geom_text(data = weekly_total, aes(x = date, y = total_qty + 10, label = total_qty),
            size = 3, color = "black") +
  scale_fill_manual(
    values = c(A = "#FFC000",
               B = "#92D050",
               C = "#AE5AFF")
  )+
  scale_x_date(
    date_labels = "%Y - CW%U",
    date_breaks = "1 week",
    sec.axis = dup_axis(
      breaks = as.Date(paste0("15-", unique(sample$xlabel2)), "%d-%b-%y"),
      labels = \(x) format(x, "%b-%y")
    )
  ) +
  theme_minimal() +
  theme(legend.position = "left", 
        axis.text.x.bottom = element_text(angle=90, hjust=1),
        axis.title = element_blank(), 
        legend.title = element_blank())

关键改动说明 #

• 新增weekly_total数据框，专门存储每周的总服务量，解决了趋势线的数据来源问题
• geom_line使用汇总数据，设置group = 1确保所有点属于同一组，能正常连线
• 给折线添加了偏移的标签，方便查看每周总数量，可根据需求调整偏移量

内容的提问来源于stack exchange，提问作者romina