使用 dplyr 总结数据框

Question

我有一个数据框，其中包含不同船只在一段时间内进行的航行。航程通常由一个起运港、一个或多个装货港和一个或多个卸货港组成。我的目标是找到开始港到装货港之间的英里数，如果有多个装货港，那么从一个装货港到下一个装货港，以及从装货港到卸货港。我还需要找到进行特定行程的月份。以下是一些测试数据：

test <- structure(list(Vessel = c("STI Selatar", "STI Selatar", "STI Selatar", 
"STI Selatar", "STI Selatar", "STI Selatar", "STI Selatar", "STI Selatar", 
"STI Selatar", "STI Selatar", "STI Selatar", "STI Selatar", "STI Amber", 
"STI Amber", "STI Amber", "STI Amber", "STI Amber", "STI Amber", 
"STI Amber", "STI Amber", "STI Amber"), VoyNum = c(14L, 14L, 
14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 62L, 62L, 62L, 
62L, 62L, 62L, 62L, 62L, 62L), Port = c("ROTTERDAM", "ROTTERDAM", 
"ENGLISH CHANNEL", "GIBRALTAR", "PIRAEUS", "ELEUSIS", "ELEUSIS", 
"AGIOI THEODOROI", "SUEZ CANAL", "SINGAPORE STRAIT", "YOSU", 
"DAESAN", "BOTANY BAY", "NAPA NAPA", "NAPA NAPA", "PNGLNG TERMINAL", 
"TORRES STRAIT", "SINGAPORE STRAIT", "SINGAPORE", "SINGAPORE STRAIT", 
"TOLO HARBOUR"), Function = structure(c(2L, 14L, 8L, 8L, 5L, 
14L, 6L, 6L, 1L, 8L, 4L, 4L, 2L, 14L, 6L, 6L, 1L, 8L, 4L, 8L, 
4L), .Label = c("Canal Transit", "Commencing", "Delivery", "Discharging", 
"Fuelling", "Loading", "Other", "Passing", "Port Call Cancelled", 
"Redelivery", "Repair", "Sampling", "Terminating", "Waiting"), class = "factor"), 
    Date_Arrival = structure(c(978307200, 1535198400, 978307200, 
    978307200, 1537237440, 1537696800, 1537736460, 1538161920, 
    1539118440, 978307200, 1541531520, 1541736000, 978307200, 
    1520194320, 1520258460, 1520654400, 1520800560, 978307200, 
    1521556560, 978307200, 1522321200), class = c("POSIXct", 
    "POSIXt"), tzone = "UTC"), Date_Departure = structure(c(1535196600, 
    1536346080, 978307200, 978307200, 1537688160, 1537736340, 
    1538155080, 1538938800, 1539190800, 978307200, 1541621880, 
    1541979000, 1519702560, 1520258340, 1520642520, 1520740800, 
    1520837280, 978307200, 1521894960, 978307200, 1522461600), class = c("POSIXct", 
    "POSIXt"), tzone = "UTC"), BallastLaden = c(0L, 0L, 0L, 0L, 
    0L, 0L, 1L, 1L, 1L, 1L, 1L, 1L, 0L, 0L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L), Miles = c(4, 2930, 0, 0, 24, 0.43, 23, 623, 7466, 
    0, 338, 0, 1720, 0.43, 43, 213, 2594, 0, 1424, 0, 0)), row.names = c(NA, 
-21L), class = c("tbl_df", "tbl", "data.frame"))

Miles 变量显示当前行与下一行之间的距离。 VoyNum是船舶所承担航次的ID。 Function 变量包含有关开始装载和卸载的信息。为了找出具体的行程是在哪一个月进行的，我正在寻找Date_Departure和Date_Arrival之间的间隔，将其减半并将其添加到Date_Departure。这是所需的输出：

# A tibble: 8 x 7
  Vessel      VoyNum FromPort        ToPort          Miles BallastLaden Month
  <chr>        <int> <chr>           <chr>           <dbl>        <int> <chr>
1 STI Selatar     14 ROTTERDAM       ELEUSIS         2958.            0 Sep  
2 STI Selatar     14 ELEUSIS         AGIOI THEODOROI   23             1 Sep  
3 STI Selatar     14 AGIOI THEODOROI YOSU            8089             1 Oct  
4 STI Selatar     14 YOSU            DAESAN           338             1 Nov  
5 STI Amber       62 BOTANY BAY      NAPA NAPA       1720.            0 Mar  
6 STI Amber       62 NAPA NAPA       PNGLNG TERMINAL   43             1 Mar  
7 STI Amber       62 PNGLNG TERMINAL SINGAPORE       2807             1 Mar  
8 STI Amber       62 SINGAPORE       TOLO HARBOUR    1424             1 Mar  

如何使用dplyr 结合group_by() 和summarize() 来实现这一目标？

我可以像这样总结总压载或满载里程，但这并不是我真正需要的，因为这些腿中的每一个都进一步分为多个端口：

test %>% group_by(Vessel, VoyNum) %>% summarise(BMiles=sum(test$Miles[BallastLaden==0]),                                                      LMiles=sum(test$Miles[BallastLaden==1]))

Answer 1

这是tidyverse 解决方案：

library(lubridate)
test %>% group_by(VoyNum) %>% arrange(Date_Departure) %>% mutate(cumMiles = cumsum(Miles)) %>%
  filter(Function %in% c("Commencing", "Loading", "Discharging")) %>% 
  mutate(Miles = c(diff(cumMiles - Miles), last(Miles)), ToPort = lead(Port),
         Month = month(Date_Departure + floor(c(Date_Arrival[-1], 0) - Date_Departure) / 2, label = TRUE)) %>%
  rename(FromPort = Port) %>% slice(-n()) %>%
  select(Vessel, VoyNum, FromPort, ToPort, Miles, BallastLaden, Month)
# A tibble: 8 x 7
# Groups:   VoyNum [2]
#   Vessel      VoyNum FromPort        ToPort          Miles BallastLaden Month
#   <chr>        <int> <chr>           <chr>           <dbl>        <int> <ord>
# 1 STI Selatar     14 ROTTERDAM       ELEUSIS         2958.            0 Sep  
# 2 STI Selatar     14 ELEUSIS         AGIOI THEODOROI   23             1 Sep  
# 3 STI Selatar     14 AGIOI THEODOROI YOSU            8089             1 Oct  
# 4 STI Selatar     14 YOSU            DAESAN           338             1 Nov  
# 5 STI Amber       62 BOTANY BAY      NAPA NAPA       1720.            0 Mar  
# 6 STI Amber       62 NAPA NAPA       PNGLNG TERMINAL   43             1 Mar  
# 7 STI Amber       62 PNGLNG TERMINAL SINGAPORE       2807             1 Mar  
# 8 STI Amber       62 SINGAPORE       TOLO HARBOUR    1424             1 Mar