Forecasting Engine

Predict bed census and arrivals

A powerful engine to set up good capacity management

Not knowing how many patients to expect and the impact on bed utilization and staff makes it difficult to properly implement capacity management.

You want to have the right capacity in the right place at the right time. To do this, you first need good predictions of the expected flow of patients and bed occupancy in the next two weeks.

We took up this challenge and built a forecasting engine that is capable of generating accurate forecasts. This is based on advanced mathematical techniques and years of experience in healthcare.


For each department, for the next 14 days, we provide hourly expected bed census including confidence intervals and daily number of emergency arrivals, elective arrivals (already planned and yet to be planned) and number of patients to be transferred.

Arrival flows

Different types of flows create demand for bed capacity. We tailor our forecasting methodologies specifically to this, so that the best forecast is made for each flow. Ultimately, we combine all flows for the overall picture.


Each type of patient goes through their own path through the hospital with an associated length of stay. We learn these characteristics to accurately determine what capacity is needed at what point in the path.

State-of-the art methods

We use various self-learning prediction and simulation techniques which automatically adapt to any changes in the current process. To this end, we can provide forecasts taking into account the current situation.

API integration

Simply send the required data via an API endpoint and get the forecasts back immediately. This way you can easily integrate the forecasts into your existing BI systems or process the forecasts in a different way.

Multiple data source formats

The engine can connect directly to a database (MySQL and SQL) to retrieve the required data or the data can be provided as a CSV file. The engine can write results back to a database (MySQL and SQL) or provide the output as json.


By means of various parameters which can be provided through the API, you can influence certain parts of the forecasts in order to set them up specifically for your use.


We value secure data. Data does not contain any personal information and data might be encrypted. We do not store this data. They are only used to make the predictions and after that the data is immediately inaccessible to us.