Smart Crowding predictions

Offentlig prosjektsammendrag

Over-crowding and inundation are major problems and risk factors in healthcare institutions; increasing waiting times and length of stay, leading to corridor patients and high use of temporary staffing solutions. In addition, while local hospitals can be stretched to the maximum, nearby facilities are left with unused capacity. Poor preparedness and oversights add to the problems; resources are not pooled, systems and people are not aligned, real-time data is not available, and bottlenecks in patient flow are rife across primary, secondary and tertiary care settings. The Covid-19 pandemic has exacerbated the need for good planning solutions, showing an urgent need for action and increased planning horizons. No software solution today gives the "probable" combined with the "definite", the "incoming" paired with the "current", and the data and power needed to act in advance of becoming overwhelmed. The ‘predictions’ innovation project is launched to expand an already disruptive value proposition with high-value-adding functionality, extending the window of opportunity where Smart Crowding is ahead of the competition, both in terms of the overall approach and technical solution. The project will result in a prototype software module to be used in future pilots, allowing for robust predictions by integrating AI/ML capabilities based on new data sources, and an activity and equipment tracking system. Weather events are particularly important, as adverse conditions such as gales or slippery ice may increase emergency room visits. Connecting predictive engines to the platform will enhance the ability to look forward days and even weeks into the future. The functionality will allow hospital staff to better prepare for what is to come and accelerate the flow of patients as they arrive. The largest challenge in the project is to create robust processes for data acquisition and integration before expanding the innovation effort based on live operations and data streams.