The nursing shortage in the United States poses a serious problem to hospitals, given that nurses provide an indispensable service within the healthcare system. This issue is expected to worsen, especially given the aging population of baby-boomers, which includes those that are part of the nurse workforce. This has resulted in a wide variety of problems, including patient safety issues, inability to detect complications, and potential patient mortality rate increases. Nurse shortage implications go beyond healthcare quality, extending to health economics as well. Inaccurate estimates of the nursing resources required to satisfy patient demand in a hospital environment could make this already-challenging problem worse. In addition, mandatory nurse-to-patient ratios implemented in some states, though providing for simplification from a demand estimation perspective, create a risk of under- or overestimating required nurse resources. We develop a series of process flow-based models that take into account the inherent complexity in key hospital departments and hence become the basis of empirical models to estimate nurse demands and thereby the best use of the scarce available nurse resource pool. In addition, via an illustrative example of a simple intensive care unit system, we demonstrate the issues with mandatory nurse-to-patient ratios in addressing the nurse shortage crisis when subject to varying patient demand and hospital service quality goals.