Post-op bed reservations held against unclear need

1. System failure identification

   Intensive care beds are reserved for scheduled surgical cases that may never need them. The bed is unavailable to the emergency department while the reservation stands, whatever the eventual outcome. Reserved capacity is functionally occupied for the whole of the operating list.

2. Data pipeline diagnostics

   A reservation is a binary hold with no probability attached. The system cannot distinguish a case that will certainly need intensive care from one that probably will not. Certainty and speculation are recorded identically once a hold exists.

3. Predictive computational model

   Attach a predicted probability of post-operative intensive care need to each scheduled case, from procedure type and patient factors. Expected demand becomes a sum rather than a count of holds. Aggregating probabilities across a list gives a far tighter estimate than counting holds.

4. Workflow integration

   Let bed management release capacity against expected demand rather than reserved count, with the probability visible so the judgement stays human and auditable. Bed managers will not act on a number whose reasoning they cannot inspect.

Back to the case study database