Hi @Drdinhluong,
Here is how to use ifelse in the situation you describe:
suppressPackageStartupMessages(library(tidyverse))
df <- data.frame(A = c(NA, "2021-04-09", NA, "2021-04-21", NA),
B = c("2021-05-03", NA, "2021-05-05", "2021-05-02", "2021-05-05"))
df
#> A B
#> 1 <NA> 2021-05-03
#> 2 2021-04-09 <NA>
#> 3 <NA> 2021-05-05
#> 4 2021-04-21 2021-05-02
#> 5 <NA> 2021-05-05
# Check that the ifelse logic is correct
ifelse(is.na(df$A), df$B, df$A)
#> [1] "2021-05-03" "2021-04-09" "2021-05-05" "2021-04-21" "2021-05-05"
# Make changes to dataframe
df %>%
mutate(A = ifelse(is.na(A), B, A) ) -> df
df
#> A B
#> 1 2021-05-03 2021-05-03
#> 2 2021-04-09 <NA>
#> 3 2021-05-05 2021-05-05
#> 4 2021-04-21 2021-05-02
#> 5 2021-05-05 2021-05-05
Created on 2021-06-18 by the reprex package (v2.0.0)
As @andresrcs suggests, your Reproducible Example should include some relevant data in text form (not just a screenshot from your monitor) because otherwise in order to help you someone has to type that data in to demonstrate a solution.
HTH