In the current economic climate public health services throughout Europe face an ever increasing challenge to deliver an effective and efficient service to patients. A specific example of this is that the Health Service Executive, the Irish Government agency responsible for healthcare in Ireland, has a dedicated division focusing on this challenge. The HSE Directorate of Quality and Clinical Care has developed a strategy for chronic disease management which focuses on improving quality, increasing access while reducing costs. This is an ambitious program which seeks to change how service is delivered in order to provide a better model of care for patients and clients in real hospital and clinical contexts within budgetary constraints.

We focus on two key areas to achieve these goals: hospital bed management and elective surgery. In each of these areas there is a complex relationship between the delivery of the service and the resources and processes into which it fits within the hospital. For example, in bed management, there is a temporal dimension to demands for beds related to seasonal illnesses. In addition, there is a usually a complex process associated with the discharge of patients and the availability of minor administrative and clinical documents, results and decisions. For example, in many hospitals a patient cannot be discharged without the authorisation of a consultant, however, the consultant needs to have specific information available in order to issue a discharge order. If documents arrive a little too late, a hospital bed might be unproductively occupied for another 24 hours. While there is a scheduling problem at the heart of this process – we would like to schedule the consultant such that all available information is at his disposal in order to recommend discharge – this is not practical. Using machine learning and data mining we can, instead, learn the preferences of the consultant and schedule the availability of key documents to coincide with his/her rounds. Similarly, in elective surgery planning, the availability of portering staff and recovery room capacity has a constraining impact on the scheduling of an operating theatre. These impacts are usually poorly understood. However, using the techniques developed in this project we are able to learn and acquire these constraints so they can be properly handled in our optimisation models.