Stochastic Optimization of Ski Resort Operations

In this project, I worked with a team to develop a data-driven optimization model to support monthly operational decision-making for Vail Mountain Ski Resort under uncertain weather and demand conditions. Ski resort profitability is heavily influenced by factors outside managerial control, such as snowfall, temperature, and skier turnout, making it a strong candidate for stochastic modeling.

The model evaluates three key operational decisions each month, whether to open the resort, produce artificial snow, and host special events, while accounting for uncertainty in snow depth, temperature, dew point, demand, and costs. These uncertainties were modeled using probability distributions informed by historical data and reasonable assumptions.

To analyze risk and variability, I implemented Excel macros to automate Monte Carlo simulations, running 1,000 trials per scenario. Excel Solver was then used to identify the combination of monthly decisions that maximized expected seasonal profit while respecting real-world constraints, such as minimum snow depth requirements, wet-bulb temperature thresholds for snowmaking, and capacity limits.

The final model produced not only an optimal operating strategy, but also confidence intervals and downside risk estimates, enabling a more realistic understanding of profit outcomes. This project demonstrates my ability to combine stochastic modeling, optimization, and automation to translate uncertainty into actionable operational insights.