Optimal Group Assignment and Workload Allocation

Integer programming models to assign students to groups by maximising diversity or topic preferences, and to allocate multi-role teaching workloads while balancing role demand, preferences, fairness, and cohort protection.

grouper

The goal of grouper is to make it easier for an instructor to do student or workload allocations.

Installation

You can install the development version of grouper from GitHub with:

# install.packages("pak")
pak::pak("Zimmy313/grouper")

But to be honest, the package on CRAN is already the most up to date. Please refer to the vignettes for examples and documentation on the models in the package.

For best performance, it is best to get a license for Gurobi optimiser.

Using Gurobi on Linux

Something to take note: Rstudio does not seem to run the .bash_profile or .bashrc scripts. It may be necessary to set the environment variables explicitly in /etc/R/Renviron and /etc/R/Renviron.site

Shiny servers, on the other hand, seem to read environment variables from .bash_profile, so no extra modifications to Renviron needed. When running on shiny servers, it is also necessary to copy the files into the app directory. It does not seem to work with the use of shiny::runApp().