In this university project, I used open source data provided by Public Health Scotland to forecast A&E attendances and waiting times across Scottish hospitals using generalised additive models. It includes visualizations, details of the pre-processing and model fitting, all of which is done in R and knited to html, which is available to download and see.
Because this is a university project, not everything is perfect and there are some flaws in the models which have resulted in large uncertainty bands for the forecasted results. I plan on revisiting this project sometime soon to address such errors. Additionally, the data for load (A&E attendance data) is too large to upload to github, if you'd like to replicate my results with the data I used, feel free to contact me.