In NiklasCarlos1994/batchanalytics: Functions That Represent Metrics For Analysing Batching Behavior

#load packages
library(batchanalytics)
library(bupaR)
library(bamalog)
library(tidyr)
library(readr)
library(lubridate)

#read in the data

#check the batch_anayltics sample for different read in from so after git import everyone can use this -> no path visible
setwd("C:\\Users\\Niklas\\Desktop\\BachelorArbeit\\EventLogs\\real_world_event_logs")

#lib batch ana verwenden - check ob neuster stand



csv_log <- read.csv("Sepsis Cases - Event Log.csv")



print(head(csv_log))

#Format----> case_id,"activity","resource","arrival","start","complete","instance_id" format required


# where resource col na vals repace with resource at the beginging


#create start, complete col und transform timestamp into correct format





#arename cols

#convert time

#csv_log$ <- as.POSIXct(csv_log$Start.Timestamp, format = "%Y-%m-%d %H:%M:%S", tz = "GMT") 
csv_log$Complete.Timestamp <-  as.POSIXct(csv_log$Complete.Timestamp, format = "%Y-%m-%d %H:%M:%S", tz = "GMT")

#check if posxct
head(csv_log)

#rename cols?

#add fake arrivaltimes?

#names(task_log) <- c("case_id", "activity", "arrival", "start", "complete", "resource", "input_batch_number", "input_batch_type")
names(csv_log)

#insert new cols with fake times



#timeConversion



csv_log$arrival <- as.POSIXct(csv_log$Complete.Timestamp, format = "%Y-%m-%d %H:%M:%S", tz = "GMT") 



csv_log$start <- as.POSIXct(csv_log$Complete.Timestamp, format = "%Y-%m-%d %H:%M:%S", tz = "GMT") 
csv_log$complete <-  as.POSIXct(csv_log$Complete.Timestamp, format = "%Y-%m-%d %H:%M:%S", tz = "GMT")



head(csv_log)

names(csv_log)

#create new table

newtable <- cbind( csv_log[1], csv_log[2], csv_log[35], csv_log[36], csv_log[37])
head(newtable)

#add arrival col with minus 5 min from start -< posixct = seconds from..

newtable$arrival <- as.POSIXct(csv_log$Start.Timestamp - 5*60 )

newtable2 <- cbind( newtable[1], newtable[2],   newtable[3],newtable[6], newtable[4],newtable[5])
head(newtable2)

#add activity instance

newtable2$instance_id <- 1:nrow(newtable2) 

#ma <- cbind(ma, "observation"=1:nrow(ma)) 

evtl complete daten bearbeiten weil compl = start immer

task_log <- newtable2

#rename cols
names(task_log) <- c("case_id", "activity", "resource", "arrival", "start", "complete","instance_id" )


head(task_log)

#write csv copy for web app

#write.csv(task_log,"C:\\Users\\Niklas\\Desktop\\BachelorArbeit\\EventLogs\\real_world_event_logs\\transformedBPIC2017_event_log_ready_for_analysis.csv", row.names = FALSE)

#test 

result_log <- my_detect_batching(task_log)

head(result_log)

#metriken einzeln ausprobieren

NiklasCarlos1994/batchanalytics documentation built on Dec. 17, 2021, 5:25 a.m.