cat("\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ \n") # Programmer : Mohammad Lakhi # Student ID : 2512430 # Module : ACCA7022- Advanced Statistical Programming #File Title : Data Manipulation Assignment # Purpose : Managing an alien colony on a distant planet resources production data across # 30 Sols, 6 Stations and 4 Teams. Perform aggregation, anomaly detection, # peak production analysis and efficiency score using loops. # Month : February,2026 cat("\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ \n") #This make sure that we get same random numbers when we run the code set.seed(42) #Storing key values as constant no_sol <- 30 #number of days in the simulation no_station <- 6 #number of station in the colony no_teams <- 4 #number of Workers team per station base_output<- 50 #Daily production output for the team #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Task A :Generate Production Data #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Using three nested loops to fill 720 rows of data #First loop : goes through each day (1-30) #Second loop : goes through each station(1-6) #Third loop : goes through each team(1-4) #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Create a table with 720 rows(30x6x4=720) total_rows <- no_sol * no_station * no_teams #Creating 4 empty columns to be filled col_sol <- integer(total_rows) #storing the day numbers col_station <- integer(total_rows)#storing the station numbers col_teams <- integer(total_rows) #storing the team numbers col_output <- numeric(total_rows) #storing the output values row_index <- 1 #First loop : goes through each day for (sol in 1:no_sol) { #Second loop : goes through each station for (station in 1:no_station) { #Third loop : goes through each team for (team in 1:no_teams){ #RANDOM NOISE(picks a number between -5:5) r_noise <- sample(-5:5, size = 1) #OUTPUT FORMULA #Calculates the output for the day,station and team output_value <- base_output + (station*3) + (team*2) + r_noise #Stores values in the specific columns col_sol[row_index] <- sol col_station[row_index] <- station col_teams[row_index] <- team col_output[row_index] <- output_value #It moves to the next row row_index <- row_index +1 } } } #Combining all four columns into one table production_data <- data.frame( sol =col_sol, station = col_station, team = col_teams, output = col_output ) cat("~~~Task-A: First 10 rows of Production Data~~~ \n") print(production_data[1:10, ]) #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Task B :Total Production per Station #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #For each station we go through all 720 rows and sum up all outputs that belongs to that station #We have use a running total variable that starts at zero for each station #It keeps up adding until it goes through each row #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Creating two empty columns for station and total station_id <- integer(no_station) station_vectors <- numeric(no_station) #looping through each station and summing up all outputs for (station in 1:no_station) { station_id[station] <- station #Saves the station number running_total <- 0 #Loop to go through every single row in production_data for (row in 1:nrow(production_data)) { if (production_data$station[row] == station){ running_total <-running_total+production_data$output[row] } } #Saves the final total for the specific station station_vectors[station] <- running_total } #build a table for results station_totals <- data.frame( station = station_id, totaloutput = station_vectors ) #results cat("\n ~~~ Task-B: Station Totals ~~~ \n") print(station_totals) #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Task-C: Detect Production Anomalies #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #for each day and each station we have done three steps: #Step 1 : Add up all 4 teams output to get daily total #Step 2 : Divide by 4 to get daily average #Step 3 : Check each teams output against the average #-> If it is more 10 percent it is marked TRUE #-> If its below 10 percent it is marked FALSE #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Creating 5 empty columns for anomaly anomaly_sol <- integer(total_rows) anomaly_station <- integer(total_rows) anomaly_team <- integer(total_rows) anomaly_output<- integer(total_rows) anomaly <- logical(total_rows) #resetting row index to 1 for filling it with anomalies row_index <- 1 #Loop to go through each day for (sol in 1:no_sol) { #Loop to go through each station for (station in 1:no_station) { # Calculate the daily total output daily_total <- 0 for (row in 1:nrow(production_data)) { if (production_data$sol[row] == sol && production_data$station[row] == station) { daily_total <- daily_total + production_data$output[row] } } #Calculate the daily average by dividing total output/no. of teams daily_average <- daily_total/ no_teams #Comparison of each teams output beside daily avg for (row in 1:nrow(production_data)) { if (production_data$sol[row]== sol && production_data$station[row]== station){ team_op <- production_data$output[row] #calculation of deviation percentage deviation <- abs(team_op - daily_average)/daily_average*100 #Stores TRUE if deviation is more than 10% otherwise FALSE anomaly[row_index]<- deviation>10 #Saves all values at the current position anomaly_sol[row_index]<-sol anomaly_station[row_index]<-station anomaly_team[row_index]<-production_data$team[row] anomaly_output[row_index]<-team_op #Moves to next row row_index <- row_index +1 } } } } #Anomalies table for five vectors anomalies<- data.frame( sol = anomaly_sol, station = anomaly_station, team = anomaly_team, output = anomaly_output, Anomaly= anomaly ) #Prints only first 10 anomalies (which are true) cat("\n~~~ Task C: First 10 Anomalies \n") an_count<- 0 for (row in 1:nrow(anomalies)) { if (anomalies$Anomaly[row] == TRUE){ print(anomalies[row, ]) an_count<- an_count+1 #stops at 10 if(an_count == 10 ) break } } #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Task D : Identify Peak Productivity Days #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #We have gone through 30 days for each station #Add up all team output for that day #Tracked the highest total value using vector best_op #best_op start at the lowest possible value (-Inf) #Every time it finds a higher total value, best_op and best_sol is updated #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #3 empty tables for the results peak_station<- integer(no_station) #station number peak_sol <- integer(no_station) #day with the highest output peak_op <- numeric(no_station) #highest total output value #Loop to go through each station for (station in 1:no_station) { best_sol <- -1 best_op <- -Inf #lowest possible output value #Loop to go through each day for (sol in 1:no_sol) { #starts total at 0 for this day sol_total <- 0 for (row in 1:nrow(production_data)) { if (production_data$station[row]== station && production_data$sol[row] == sol) { sol_total<- sol_total + production_data$output[row] } } #Comparision of sol_total with the best current best value if (sol_total > best_op) { best_op<- sol_total best_sol <- sol } } # after checking all solutions, storing the peak solution values peak_station[station] <- station peak_sol[station] <- best_sol peak_op[station]<- best_op } #Building peak days Data frame peak_days <- data.frame( station = peak_station, sol = peak_sol, total_output =peak_op ) #Print peak days resukt cat("\n~~~ Task D: Peak Productivity Days per Station ~~~\n") print(peak_days) #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # Task E : Calculate Efficiency Score #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #Adding a new column called efficiency_score to production_data #Higher value means it have produce output relative to their station and team number combined #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #new column for efficiency score production_data$efficiency_score <- numeric(nrow(production_data)) #looping through every row and calculate efficiency score for (row in 1:nrow(production_data)) { #FORMULA production_data$efficiency_score[row]<- production_data$output[row]/(production_data$station[row]+production_data$team[row]) } #Printing first 10 rows of the updated dataframe cat("\n~~~ Task E : First 10 Rows with efficinecy score ~~~\n") print(production_data[1:10, ])