Data-Driven Decision-Support for Process Improvement through Predictions of Bed Occupancy Rates

Managing bed utilization by ensuring the supply keeps up with demand is not an easy task in a large public hospital with many medical disciplines. The bed managers who make decisions on reserving and allocating beds centrally require high-dimensional data from several hospital information systems supporting processes in emergency room, specialized clinics and wards. In this work, we put together an automated process for cleaning, consolidating and integrating data from several information systems into reports required by the bed managers to analyse the bed occupancy situations across more than thirty medical disciplines. To prevent bed crunch situation where patients wait more than ten hours for beds, we developed two predictive models based on Principal Component Analysis and Multiple Linear Regression to provide hospital with the foreknowledge of the bed occupancy situations. Our aim is to move the hospital from reactive to proactive bed management. Our data-driven solution focuses on consistent and accurate reporting, content co-creation with the bed managers, and an approach which enables seamless transition towards proactive decision-making. The solution is implemented in a large public hospital in Asia. It provides high value to stakeholders in hospital by reducing the time-taken to get the information on hand for decision-making from at least four days to half a day. The selected prediction model for bed occupancy rate has achieved 80% accuracy (at an error tolerance of 5%). We hope our results will encourage and benefit hospitals with similar settings to adopt data-driven methods to tackle high bed occupancy situations in their premises.