University Hospitals Coventry and Warwickshire NHS Trust

Solution

UHCW NHS Trust, IBM Consulting, and Celonis formed a blended team that combined process mining, operational research, and generative AI to improve outpatient services. Celonis process mining analyzed more than half a million pseudonymized patient journeys, and the team layered demographic data over the analysis so interventions would not worsen health inequalities. Using the IBM Garage model, the group then tested workflow changes such as adjusting SMS reminder timing to improve rebooking and slot reuse. In parallel, IBM watsonx.ai was piloted to train, tune, and deploy machine learning models that help schedule and validate patients on the elective backlog, while generative AI reads clinical letters to verify waiting-list status. The resulting architecture joins process data, patient communications, and document understanding into a faster operational loop, enabling UHCW to prioritize the right interventions and improve throughput without adding staff.

Data flow

Operational and demographic data feed the Celonis process-mining analysis, IBM Garage turns the findings into tested workflow changes such as reminder timing, and watsonx.ai models and generative AI process backlog records and clinical letters to support faster scheduling decisions.