Here is a tutorial to create a custom dot plot / dumbbell plot schedule with R and geom_dumbbell and geom_segment. With this template you can also color items by categories such as deliverable, status or quarter.

Original inspiration: and guidance on using geom_dumbbell.

Read and view the data

I started from an excel sheet.

schedule <- read_excel(here("posts", "2022-02-08-how-to-make-a-dumbbell-schedule-with-r", "schedule.xlsx"), sheet = "Sheet1")

kable(schedule) %>% 
  kable_paper(full_width = TRUE) %>% 
  row_spec(0, bold = T) %>% 
  kable_styling(latex_options = "HOLD_position")

milestone	deliverable	start_date	due_date	status	quarter
Draft Feasibility Study	Feasibility Study	2022-01-12	2022-02-07	complete	Q1
Final Feasibility Study	Feasibility Study	2022-02-12	2022-03-11	in progress	Q1
Draft Design Drawings	Design Drawings	2022-01-24	2022-03-18	in progress	Q1
Final Design Drawings	Design Drawings	2022-04-12	2022-05-20	not started	Q2
Draft Technical Documentation	Technical Documentation	2022-03-01	2022-03-18	not started	Q1
Final Technical Documentation	Technical Documentation	2022-03-28	2022-04-22	not started	Q2
Stakeholder Presentation	Presentation	NA	2022-03-25	not started	Q1
Draft Web Application	Web Application	2022-02-18	2022-03-16	not started	Q1
Final Web Application	Web Application	2022-03-28	2022-05-27	not started	Q2
Client Presentation	Presentation	NA	2022-06-03	not started	Q2

Clean the data

Pay attention to date formats with class(schedule$start_date). Cleaning steps created a field for month abbreviation + day of month to use as labels. The deliverable category was reordered based on due date using fct_reorder.

schedule <- schedule %>% 
  mutate(start_month_num = month(start_date)) %>% 
  mutate(end_month_num = month(due_date)) %>% 
  mutate(start_month_name = case_when(
    start_month_num == 1 ~ "Jan",
    start_month_num == 2  ~ "Feb",
    start_month_num == 3 ~ "Mar",
    start_month_num == 4 ~ "Apr",
    start_month_num == 5 ~ "May",
    start_month_num == 6 ~ "Jun",
  )) %>% 
  mutate(end_month_name = case_when(
    end_month_num == 1 ~ "Jan",
    end_month_num == 2 ~ "Feb",
    end_month_num == 3 ~ "Mar",
    end_month_num == 4 ~ "Apr",
    end_month_num == 5 ~ "May",
    end_month_num == 6 ~ "Jun",
  )) %>% 
  mutate(start_label = paste(start_month_name, day(start_date))) %>% 
  mutate(end_label = paste(end_month_name, day(due_date))) %>% 
  mutate(milestone = as_factor(milestone)) %>% 
  mutate(milestone = fct_reorder(milestone, as.numeric(due_date), .desc = TRUE))

This step was used for the bounds of the rectangle highlighting the second quarter.

Q2 <- schedule %>% 
  filter(quarter == "Q2")
xmin_Q2 <- as.POSIXct(min(Q2$start_date, na.rm = TRUE))
xmax_Q2 <- as.POSIXct(max(Q2$due_date, na.rm = TRUE))

This step was used to set the x-axis limits when using scale_x_datetime.

min <- as.POSIXct("2022-1-1")
max <- as.POSIXct("2022-6-15")

I originally used geom_dumbbell for the start and end dots, but using geom_point twice gave more flexibility for colors and made it easy add transparency to the start dots. The geom_dumbbell code is include for reference.

time_plot <- ggplot(data = schedule, aes(y = milestone)) +
  geom_rect(aes(xmin = xmin_Q2, ymin = -Inf,
                xmax = xmax_Q2, ymax = Inf),
                fill = "grey80", alpha = 0.5) +
  # create a thick line between x and xend instead of using default provided by geom_dumbbell
  geom_segment(aes(x = start_date, xend = due_date, y = milestone, yend = milestone), 
               color = "grey50", 
               size = 1.5) +
  # geom_dumbbell(color = "grey80", size_x = 5, size_xend = 5,
  #               colour_x = "blue", colour_xend = "red") +
  geom_point(data = schedule, aes(x = start_date, y = milestone,
                              color = deliverable), size = 5, alpha = 0.5,
             show.legend = FALSE) +
  # scale_colour_paletteer_d("Redmonder::qMSOPap") +
  # scale_colour_paletteer_d("Redmonder::qMSOPap") +
  # scale_colour_paletteer_d("Redmonder::qMSOMed") +
  # scale_colour_paletteer_d("palettetown::pelipper") +
  scale_colour_manual(values = c("#007a76", "#b7245c", "#ca7f0e", "#0d4fbd", "#785ceb")) + # manually specify colors or use an existing palette
  geom_point(data = schedule, aes(x = due_date, y = milestone,
                              color = deliverable), size = 5,
             show.legend = FALSE) +
  labs(x = NULL, y = NULL,
       title = "Milestone Timeline") +
  geom_text(color = "black", size = 3, hjust = 1.5,
            aes(x = start_date, label = start_label)) +
  geom_text(color = "black", size = 3, hjust = -0.5,
            aes(x = due_date, label = end_label)) +
  theme_minimal() +
  theme(panel.grid = element_blank()) +
  theme(plot.title.position = "plot") +
  scale_x_datetime(limits = c(min, max)) +
  annotate("text", x = as.POSIXct("2022-5-1"), y = schedule$milestone[1], label = "Q2", color = "black", size = 5)
time_plot

Warning: Removed 2 rows containing missing values (geom_segment).

Warning: Removed 2 rows containing missing values (geom_point).

Warning: Removed 2 rows containing missing values (geom_text).

Citation

BibTeX citation:

@online{rivers2022,
  author = {Rivers, Marie},
  title = {How to Make a Dumbbell Schedule with {R}},
  date = {2022-02-08},
  url = {https://marierivers.github.io/posts/2022-02-08-how-to-make-a-dumbbell-schedule-with-r/},
  langid = {en}
}

For attribution, please cite this work as:

Rivers, Marie. 2022. “How to Make a Dumbbell Schedule with R.” February 8, 2022. https://marierivers.github.io/posts/2022-02-08-how-to-make-a-dumbbell-schedule-with-r/.