Method
Prophet
Additive trend-seasonality-holiday forecasting model.
10 cases from 8 companies apply Prophet. 3 report results in production.
Typical tasks: Forecast demand and load, Detect anomalies, Build a data platform. Companies: Walmart, Yandex Lavka, Avito, trivago, DoorDash, Uber and 2 more. Often combined with: Anomaly detection, Stacking, ARIMA / VAR, Linear regression, Threshold selection and 16 more.
| A Short-Term Anomaly Forecasting Model for Univariate Time Series and Its Use in the Sfera.IAS Product | general | 2025-05 | |
Forecasting Model Architecture in Yandex Lavka![]() | ![]() | food-delivery | 2025-05 |
| Autotuning: A comprehensive approach to Time Series Forecasting | retail | 2025-03 | |
| How We Cut Forecasting Time from Two Weeks to Five Minutes | classifieds | 2024-05 | |
| Real-world Insights: Anomaly Detection in Internet Traffic | travel | 2024-02 | |
| Order Forecasting on a Map Using a Single Time Series | general | 2023-10 | |
| How DoorDash Built an Ensemble Learning Model for Time Series Forecasting | food-delivery | 2023-06 | |
| ML for Forecasting and Analyzing Passenger Transport Behavior | — | mobility | 2023-05 |
| D3: An Automated System to Detect Data Drifts | mobility | 2023-02 | |
| Fitting a univariate time series model to a cluster of series | retail | 2022-03 |
