Monte Carlo Simulation and Human Risk


Many industries—reinsurance, finance and commercial development, for example—rely on risk analysis technology that utilizes Monte Carlo simulation. To boil down the likelihood of success or failure of any given project, the simulation process considers a series of random events, or risks, and calculates thousands of scenarios to project the probability of different outcomes occurring.

For the aforementioned industries, the risks often deal with insurance decisions, expected shortfalls in capital or supply chain disruptions. While important issues, they are not life threatening.

However, Monte Carlo simulation, which was created by scientists attempting to determine radiation shielding, has assisted in the assessment of risk that impacts more than just the bottom line. In many instances, it has helped save lives and guide decision makers in times of disaster.

While the basic premise and procedures of Monte Carlo simulation remain constant, the technology used has continued to evolve. In turn, more industries have discovered unique ways—with remarkable results-—to extend its utility beyond the business world.

Hurricane Katrina Call Center

When Hurricane Katrina struck New Orleans in 2005, Louisiana officials created a family crisis phone center to help callers locate family members and collect information regarding missing persons. Given the expected influx of calls, Monte Carlo simulation helped officials quickly determine the call load the center should expect and create a staffing schedule to manage the activity.

Critical data points, such as timing of calls, duration of calls and number of deaths were used to predict potential call load. As new data became available, the schedule was adjusted to anticipate necessary schedule changes.

The call center helped reunite 8,000 of the 11,000 people reported missing in Louisiana and identify 900 of the estimated 1,100 people who died as the result of the devastating hurricane.

Guatemala Volcano Evacuation

Volcan de Fuego is an active volcano located in a relatively populated area of Guatemala. While the volcano’s activity typically poses little risk to those living nearby, it has the potential to be destructive. Recently, Guatemala’s National Institute for Seismology, Volcanology, Meteorology and Hydrology worked with the University of Bristol to create a risk assessment of potential volcanic-related evacuation risks. The study considered the likelihood of a successful evacuation by inputting several variables, such as the time between a possible eruption and a possible hazard hitting a location, along with communication times from authorities and evacuation times.

These variables were each modeled by experts using probability distributions. In September 2012, after its largest eruption since 1999, more than 33,000 people within 12 miles of the volcano were evacuated.

Pandemic Preparation

In 2008, the New York City Health Emergency Preparedness Program examined the ability of the city’s 64 hospitals to properly staff nurses in the event of a pandemic. By using Monte Carlo simulation, it was possible to ascertain a series of probabilities regarding nursing shortages during a future pandemic. Officials identified a 26.5% chance that shortages would exist, a 10% chance that there would be a shortage of at least 46 nurses and a 1% chance of a shortage of more than 102 nurses.

The simulation took into account variables such as the number of hospital beds, the maximum number of shifts a nurse can perform in a week and the total number of shifts available in a week. Nurse absenteeism, due to taking care of family members also suffering from the pandemic and nurses’ desire to protect themselves from being exposed, was also factored into the simulation.

Randy Heffernan

More articles by »

About the Author

Randy Heffernan is vice president at Palisade Corporation, a provider of risk and decision analysis software.



  • That kind of actions are perfectly organized. Hail to the people who did this.

  • Raúl Castro

    Excellent article Randy!
    There is a great window of opportunity still pending to explore in many industries.I'm sure that future managers will be to acknowledge the growing advantages of Montecarlo simulation!

    Best Regards,


  • Michel Fliess

    Monte Carlos techniques do necessite a 'good' probabilistic model. This is almost always impossible for 'rare' events. Other methods which do not rely on such assumptions exist and night be user instead.


Leave a reply