Management Decision Optimization Method and Empirical Study Based on Computer Big Data Intelligent Analysis

With the deepening of healthcare system reform and the promotion of smart healthcare construction, hospital management is facing multiple challenges such as diversified patient needs, refined resource allocation, and improved operational efficiency. The traditional decision-making model that relies on experience is no longer suitable for the development of modern hospitals. Problem: Currently, there are problems in hospital management such as imbalanced resource scheduling, redundant diagnosis and treatment processes, and lagging risk warning, resulting in high operating costs and low patient satisfaction. The structure and content of this article: firstly, this article constructed a multi-source data fusion system for hospitals, integrating electronic medical records, resource scheduling, operational finance and other data. Secondly, this article proposed a fusion prediction model based on an improved random forest and LSTM to achieve accurate prediction of core indicators such as patient flow and resource demand. Finally, this article constructed a big data intelligent decision support platform. The experimental survey results are as follows: the work efficiency of medical staff has significantly improved, the work efficiency of medical staff has significantly improved, the daily number of patients received by doctors has increased from 16.52 to 17.53, the average number of patients cared for by nurses has increased from 8.82 to 10.94, and the satisfaction of medical staff has increased from 77 points to 85 points, verifying the effectiveness and practicality of this method.