(Travelling Salesman in R): Error in distMatrix[route] : subscript out of bounds

[omario]

I am working with the R programming language.

I found this link over here which shows how to recreate the "Travelling Salesman Problem" in R using the Genetic Algorithm (Optimized Delivery Route using Genetic Algorithm: Cost cutting for e-commerce)

library(GA)
data("eurodist", package = "datasets")
D <- as.matrix(eurodist)


tourLength <- function(tour, distMatrix) {
   tour <- c(tour, tour[1])
   route <- embed(tour, 2)[,2:1]
   sum(distMatrix[route])
}

#Firness function to be maximized

tspFitness <- function(tour, ...) 1/tourLength(tour, ...)

GA <- ga(type = "permutation", fitness = tspFitness, distMatrix = D,
          min = 1, max = attr(eurodist, "Size"), popSize = 50, maxiter = 5000,
          run = 500, pmutation = 0.2)

We can then see the results:

 summary(GA)

## Solutions              = 
##      x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15 x16 x17 x18 x19
## [1,] 20  7 11  3  4  5 18 13 12   9  14   2  15   8  16  19   1  21  17
## [2,]  6 10 20  7 11  3  4  5 18  13  12   9  14   2  15   8  16  19   1
## [3,] 10  6 17 21  1 19 16  8 15   2  14   9  12  13  18   5   4   3  11
##      x20 x21
## [1,]   6  10
## [2,]  21  17
## [3,]   7  20

I want to try and repeat this using my own data set.

Part 1: For instance, suppose I create a dataset that contains the longitude and latitude of cities :

set.seed(123)
data_1 = data.frame(id = c(1,2,3), long = rnorm(3, -74, 1 ), lat = rnorm(3, 40, 1 ))
data_2 = data.frame(id = c(4,5,6), long = rnorm(3, -78, 1 ), lat = rnorm(3, 42, 1 ))
final_data = rbind(data_1, data_2)
N <- nrow(final_data) # just for repeated convenience
final_data
#   id      long      lat
# 1  1 -74.56048 40.07051
# 2  2 -74.23018 40.12929
# 3  3 -72.44129 41.71506
# 4  4 -77.53908 41.55434
# 5  5 -79.26506 43.22408
# 6  6 -78.68685 42.35981 

Part 2: Then, I create a "distance matrix" that shows the distance between each pair of cities:

library(geoshpere)

dists <- outer(seq_len(N), seq_len(N), function(a,b) {
  geosphere::distHaversine(final_data[a,2:3], final_data[b,2:3]) # Notes 1, 2
})

dists
#           [,1]      [,2]     [,3]     [,4]     [,5]     [,6]
# [1,]      0.00  28876.24 255554.4 300408.5 525566.9 429264.3
# [2,]  28876.24      0.00 231942.7 320616.0 541980.9 448013.6
# [3,] 255554.43 231942.67      0.0 424449.9 584761.5 521210.7
# [4,] 300408.47 320616.03 424449.9      0.0 233840.9 130640.9
# [5,] 525566.87 541980.93 584761.5 233840.9      0.0 107178.2
# [6,] 429264.34 448013.57 521210.7 130640.9 107178.2      0.0

Part 3: Finally, I tried to run the Genetic Algorithm on this problem:

GA <- ga(type = "permutation", fitness = tspFitness, distMatrix = dists,
         min = 1, max = attr(eurodist, "Size"), popSize = 50, maxiter = 5000,
         run = 500, pmutation = 0.2)

My Problem: Unfortunately, this last line of code produces the following error:

Error in distMatrix[route] : subscript out of bounds
In addition: Warning messages:
1: In ga(type = "permutation", fitness = tspFitness, distMatrix = dists,  :
  'min' arg is deprecated. Use 'lower' instead.
2: In ga(type = "permutation", fitness = tspFitness, distMatrix = dists,  :
 
 Error in distMatrix[route] : subscript out of bounds 

Can someone please show me what am I doing wrong and how can I fix this?

Thanks!