This regional society meets regularly to enable members to network with other operational researchers in their local area and to expand their knowledge of OR in breadth and depth.

We put on meetings, works visits and other events, which are generally free admission and open to all, giving members the freedom to invite clients and others outside OR who are interested in the subject.

Connect with us

You can find the latest news of this group’s activities as well as dates and details of upcoming and past events. You can also join this group’s email distribution list and get in direct touch with the group’s organisers through the ‘get in touch’ form. Any documents, such as presentation slides and promotional leaflets, are available through the ‘related documents’ box below.

Committee details

Join us for our next event

Probabilistic Forecast Reconciliation For Emergency Services Demand

Date: 3 May 2023, 12.00
Venue: Seminar Room 0.04 (ground floor), Abacws building, Cardiff University, Senghennydd Road, Cardiff

Register here

About the Event

Accurate forecasts of ambulance demand are crucial inputs when planning and deploying staff and fleet. Such demand forecasts are required at national, regional and sub-regional levels, and must take account of the nature of incidents and their priorities. These forecasts are often generated independently by different teams within the organization. As a result, forecasts at different levels may be inconsistent, resulting in conflicting decisions and a lack of coherent coordination in the service. To address this issue, we exploit the hierarchical and grouped structure of the demand time series, and apply forecast reconciliation methods to generate both point and probabilistic forecasts that are coherent and use all the available data at all levels of disaggregation. The methods are applied to 5 years of daily data, disaggregated by nature of incident, priority, managing health board, and control area. We use an ensemble of forecasting models, and show that the resulting forecasts are better than any individual forecasting model. We validate the forecasting approach using time-series cross-validation.

Following the presentation, a number of us will be going for lunch at a local pub. If you would like to join please email Paul Harper ([email protected]).