CONCLUSION

In this paper, the group developed a stochastic simulation model using real hospital data to support the model. Using this model, hospitals would be able to determine the optimal number of resources required to effectively serve all COVID-19 patients, therefore maximizing patient treatment and minimizing patient deaths. The main purpose of this model is focusing on simulating a model that yields a result close to the real data set. However, the group is also able to use this simulation to calculate different variables that the data set did not provide in the first place and to observe the effects of changes in variables on the system performance. Therefore, this simulation model can be altered to meet the specific needs of different systems.

With future work, the group hopes to be able to expand the model to include more accurate measures of system performance using more advanced simulation techniques while also including additional resources and entity attributes such as hospital staff, ventilators and other hospital equipment, and patient severity levels. This will allow the model to be more flexible and useful to the end-users. Some possible end users of the system include hospital managers who may use the simulation model for tasks such as determining the number of resources needed at the hospital, determining the effects of process changes on the system, improving patient death rates, or even for preparing for future surges in patient hospitalization. The model could also be used by researchers to understand how the various characteristics of a hospital impact COVID death rates and to learn more about COVID in terms of severity trends by observing both the ICU and regular bed data.

However, there are various limitations associated with the simulation model. It is undeniable that the more complex the model, the more significant time and money people have to invest. If the organization using the model (hospital or researcher) needs to make an adjustment to the model or fix an error, this will require an individual who is knowledgeable and trained in simulation methodology which may not be directly available. In addition, simulation models do not directly provide any answers or solutions to the user. The results of the simulation model instead must be interpreted which requires some previous knowledge of both the system and model. Sometimes, results may be difficult to interpret and some more work will be required to make sure the end users can make the right decision for their companies or organizations.

In conclusion, the simulation model might not satisfy all the requirements for the end user, but it provides a main foundation using the simulation tools and techniques learned throughout the semester. Through this work, the group learned how to effectively develop models, make the results more efficient and accurate, and to interpret the output.