Back to projects

Forecasting / Operations Analytics

System Capacity and Care Load Analytics for UAC

A capacity analytics dashboard for monitoring care-load pressure, backlog flow, anomalies, and 7-day operational forecasts.

Main result0.8767R2 score
MAE27.82
RMSE33.83
forecast horizon7-day
regressorRF
Live dashboard walkthrough, embedded without autoplay.
Problem

Operational teams need a clearer view of care-load pressure, backlog trends, and near-term capacity risk across changing system conditions.

Solution

Built a Streamlit analytics system with KPI monitoring, pressure anomaly views, a Random Forest 7-day prediction model, and scenario simulation.

Outcome

The dashboard connects current load, forecasted arrivals, and model insight views so capacity pressure is easier to monitor and explain.

Workflow

01Load signals
02Backlog flow
03Pressure anomaly
047-day model
05Scenario view

Role and technical focus

Forecasting model, KPI design, scenario simulator, dashboard build

PythonRandom ForestStreamlitPandasAnalyticsForecasting

Notes

  • Random Forest Regressor used for 7-day arrival prediction.
  • Dashboard includes pressure, volatility, backlog, and scenario simulation views.
  • Metrics are reported as model performance, not causal claims.
Project media

Real screens and analysis visuals