使用ggplot绘制多过程数据：含日期、测量值及上下限可视化需求

Perfect, let's walk through how to get this done cleanly with R and ggplot2. Your dataset size is totally manageable, and breaking this into two key steps—date cleaning and plotting—will make it straightforward.

Step 1: Prep Your Data (Date & Factor Handling) #

First, we need to make sure your date column is recognized as a date type (not just text) and that your process types are treated as categorical factors (this helps with grouping/coloring in ggplot). We'll use tidyverse for data wrangling and lubridate for easy date parsing:

# Load required packages
library(tidyverse)
library(lubridate)

# Clean the data
df_clean <- df %>%
  # Convert date column to date format—adjust ymd() to match your actual date structure!
  # Use mdy() if dates are like "MM/DD/YYYY", dmy() for "DD/MM/YYYY"
  mutate(
    date = ymd(date),
    process_types = as.factor(process_types)
  )

Double-check the date conversion worked with head(df_clean$date)—you should see dates formatted as YYYY-MM-DD instead of strings.

Step 2: Build the ggplot Visualization #

Now we'll create a plot that shows measured values over time for each process, with upper/lower limits included. Since you have 14 process types, using faceting (one plot per process) will keep things readable instead of cramming all lines into one chart.

Option 1: If Upper/Lower Limits Vary by Date

If each row has a unique upper/lower limit tied to that date and process, use this code:

ggplot(df_clean, aes(x = date)) +
  # Add upper limit (gray dashed line)
  geom_line(aes(y = upper_limit), color = "gray50", linetype = "dashed") +
  # Add lower limit (gray dashed line)
  geom_line(aes(y = lower_limit), color = "gray50", linetype = "dashed") +
  # Add measured value line (colored by process type)
  geom_line(aes(y = measured_value, color = process_types), linewidth = 1) +
  # Facet by process type—adjust ncol to fit your screen
  facet_wrap(~process_types, ncol = 4) +
  # Customize labels and theme
  labs(
    title = "Measured Values vs. Control Limits by Process",
    x = "Date",
    y = "Measurement Value",
    color = "Process Type"
  ) +
  # Format date axis to avoid overlapping text
  scale_x_date(date_labels = "%Y-%m-%d", date_breaks = "1 month") +
  theme_minimal() +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1),
    plot.title = element_text(hjust = 0.5, size = 14)
  )

Option 2: If Upper/Lower Limits Are Fixed per Process

If each process has a single static upper/lower limit (not changing with date), we'll first extract those fixed values, then use geom_hline to add horizontal lines to each facet:

# Extract fixed limits per process
process_limits <- df_clean %>%
  group_by(process_types) %>%
  summarise(
    lower_limit = unique(lower_limit),
    upper_limit = unique(upper_limit)
  )

# Build the plot
ggplot(df_clean, aes(x = date, y = measured_value, color = process_types)) +
  geom_line(linewidth = 1) +
  # Add fixed lower limit to each facet
  geom_hline(data = process_limits, aes(yintercept = lower_limit), color = "gray50", linetype = "dashed") +
  # Add fixed upper limit to each facet
  geom_hline(data = process_limits, aes(yintercept = upper_limit), color = "gray50", linetype = "dashed") +
  facet_wrap(~process_types, ncol = 4) +
  labs(
    title = "Measured Values vs. Fixed Control Limits by Process",
    x = "Date",
    y = "Measurement Value",
    color = "Process Type"
  ) +
  scale_x_date(date_labels = "%Y-%m-%d", date_breaks = "1 month") +
  theme_minimal() +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1),
    plot.title = element_text(hjust = 0.5, size = 14)
  )

Quick Tips #

• If your date axis still looks cluttered, adjust date_breaks (e.g., use "2 months" instead of "1 month").
• If you prefer all processes on one single chart instead of facets, remove the facet_wrap() line—just note that 14 colors might get busy, so you could use scale_color_viridis_d() for more distinguishable hues.
• Always check for missing values with sum(is.na(df_clean))—missing dates or measurements can mess up the line plot.

内容的提问来源于stack exchange，提问作者Jenny