This is a brief documentation of how to set up an R environment in Windows in a remote workspace for an end user that will not have internet access.

We will assume that we can access the server workspace at G:\user_id\, that the user can access it from H:\, and that the directory structure looks like this:

├── applications       <- Applications added by the administrator
│   └── r
│       ├── libraries  <- R packages
│       ├── R          <- Base R files
│       └── RStudio    <- Optional R Studio
│
└── user               <- The users files and folders

Base R

Download the R installer from here:

https://cran.r-project.org/bin/windows/base/release.htm

Start the installer and choose the installation directory to be in the users applications directory:

G:\user_id\applications\r\R

RStudio

Download the RStudio zip archive from:

https://www.rstudio.com/products/rstudio/download/#download

Extract the contents of the archive to:

G:\user_id\applications\r\RStudio

Install Packages

To install packages to the libraries directory, start by specifying the install path and then install the packages using this path:

lib_path = "G:/user_id/applications/R/libraries"
install.packages("package_name", lib=lib_path)

Use Packages (for the end user)

For the user to be able to load the packages, they need to set the libpath in R to the correct folder in every session/script before loading the packages:

.libPaths("H:/applications/R/libraries")
library(package_name)

Another possible solution is to set the libpath in the .Rprofile file in the
users home directory.