Repository for a project about building a deep learning algorithm to predict shifts in hospitals.
Shift scheduling in hospitals is a time consuming process. Every month teams (including doctors and nurses) have to look for better shifts according to their preferences in order to make an optimal shift planning. This projects aims to develop a model to predict shifts using data from an hospital in Brazil in order to help the hospital teams to optimize their shift choosing.
The model uses the LSTM (Long Short Term Memory) to capture the patterns in the historic sequences of shifts chosen by a doctor for example. The algorithm, therefore, can capture the patterns and learn the preferences of each person to choose the best shift, so teams no longer have to spend time in shift planning, optimizing and automating this process.