Keeping track of the huge amounts of patient data that are needed to make decisions and improve results is made possible by healthcare information systems. But it’s hard to optimize for prediction analytics because healthcare data is getting more complicated and bigger. It looks into optimization models that find a good mix between being able to guess accurately, using computers efficiently, and being able to grow. We look at linear and nonlinear programming, as well as genetic algorithms (GA) and particle swarm optimization (PSO) and other metaheuristic algorithms. To improve speed and efficiency, hybrid ways are being considered.  Multi-objective optimization is used to find fair, best solutions that help healthcare settings make better decisions and use their resources more efficiently.