Scenario Continuation
Now that we've identified potential risks, the next step is to quantify their financial impact. Using the cost range and likelihood of each risk, we can use Monte Carlo simulations to determine how much contingency buffer should be added to the project budget.
What is a Monte Carlo Simulation?
A Monte Carlo simulation is a method that uses random sampling to show how a project's costs might vary based on risk. By running many simulations, we can estimate the total risk cost and see how much budget we need to be confident we've covered it - for example, having enough budget 50% (P50) or 90% (P90) of the time.
Review the risk data provided by the Plant Manager, including cost ranges and likelihoods, and use it to calculate the P50 and P90 total risk cost estimates.
Tip: This can be done effectively in Python using Google Colab, where you can build a simulation using random sampling techniques based on the triangular distributions provided.
Objective
Create a visual diagram that communicates the results of your Monte Carlo simulation and supports a contingency funding recommendation. To get there, you will:
- Run a Monte Carlo simulation using the updated 2025 risk distributions provided.
- Calculate the P50 (median) and P90 (conservative) estimates for total project risk cost.
- Visualize the results with a histogram (or similar chart) that shows the full distribution of possible outcomes.
- Use the results to recommend how much contingency should be added to the project budget.
As you design your visualization, keep in mind:
- Decision focus: Sponsors want to know “How much contingency should we set aside?”
- Audience perspective: Make the spread of costs and the meaning of P50 vs P90 clear to non-technical stakeholders.
- Design for insight: Titles, markers, and annotations should highlight the story, not just the data.
AI tools can help with coding, but remember:
- You are the analyst. Decide first what message your diagram should communicate.
- Review critically. Ask yourself: Would this diagram help a sponsor make a confident funding decision?
Note: Normally, it is not recommended to share sensitive company data with public AI tools. Since these numbers are fictional, this exercise provides a safe opportunity to practice AI workflows. A common best practice is to prototype with dummy data, then apply the workflow on a local machine with real data.