Machine learning integrated patient flow simulation: why and how?

Stochastic distribution methods were used to construct patient flow simulation sub-models such as inflow, length of stay (LoS), cost treatment (CoT) and clinical pathways (CPs). However, the inflow rate demonstrates seasonality, trend, variation due natural human-made factors. LoS, CoT CPs are determined by social-demographics factors, laboratory test results, resource availability healthcare structure. For this reason, models developed using stochastic have limitations including uncertainty, not recognising heterogeneity, representing personalised value-based healthcare. This, in turn, results a low acceptance level implementation solutions suggested models. On other hand, machine learning becomes effective predicting CoT, CPs. This paper, therefore, describes why coupling with is important, proposes conceptual architecture for integrated its examples.