Teaching mathematical modeling is one of the most effective ways to demonstrate to students the real-world applications of mathematics. It’s also one of the most time-consuming to prepare for. Finding relevant contexts, creating clean datasets, and designing high-quality, open-ended problems is a significant workload.

Generative AI (such as Gemini, ChatGPT, or specialized tools like Magic School AI) can serve as your personal "teaching assistant," handling the preparatory work so you can focus on what's most important: guiding your students' thinking.

Here’s how you can use it to build your next modeling lesson.

Generate Contextually Relevant Problem Ideas

This is the brainstorming phase. AI is brilliant at connecting student interests to math. Instead of starting from scratch, give the AI your students' context.

• Your Goal: Move from "What should I teach?" to "What are my students into, and how can I turn that into a math problem?"

• Key Action: Prompt the AI to generate ideas based on student interests, local events, or other relevant school subjects.

Example Prompt #1

My 10th-grade students are obsessed with the new 'Space Guardian' video game. Give me 5 ideas for a mathematical modeling problem related to the game (e.g., in-game economy, player growth rates, or mission success probability).

Example Prompt #2

I need a 45-minute modeling activity for my Algebra 1 class. My students live in Tampa, Florida. Brainstorm 3 problem ideas related to our local environment, like tidal patterns, tourism revenue, or business planning for a new surf shop.

Example Prompt #3

My algebra 1 class just finished a unit on endangered species in their biology class. Generate 3 modeling problem ideas that connect exponential functions to animal population data.

Generate Clean, Custom Datasets

This is one of the biggest time-savers. No more hunting the web for usable data or spending an hour cleaning a spreadsheet. AI can create perfectly formatted synthetic data to match your exact needs.

• Your Goal: Get a clean, relevant dataset in 30 seconds, not 30 minutes.

• Key Action: Specify to the AI exactly what type of data you need, including its format.

Example Prompt #1

I'm doing a problem on the relationship between hours of study and test scores. Create a synthetic dataset for 30 fictional students. The data should show a moderate positive correlation but not be perfectly linear. Include a few outliers. Format it as a simple table.

Example Prompt #2

Generate a CSV-formatted dataset for the 'cafeteria food waste' problem. Include 4 weeks of daily data with these columns: Day_of_Week, Main_Menu_Item, Weather_C, and Waste_kg.

Example Prompt #3

Take the following dataset and reformat it as a Python list of dictionaries:
[dataset]

Create Visual Aids and Visualizations

A good visual can make an abstract problem concrete. AI can help you create both illustrative "hooks" and data-driven visualizations.

• Your Goal: Help students "see" the problem and the data.

• Key Action: 

  • For Problem Context (Image AI): Use tools like DALL-E, Midjourney, or Adobe Firefly to create images for your problem.

  • For Data (Code Interpreters): Use AI that can analyze data (like Gemini's Code Interpreter or ChatGPT-4) to create graphs on the fly.

Example Prompt #1

Generate a simple, clear diagram of a 'leaky bucket' system to illustrate a modeling problem. Show a clear 'inflow' and 'outflow' with labels.

Example Prompt #2

Create a photo-realistic image of a crowded high school cafeteria at lunchtime, to be used as the title slide for a problem on food waste.

Example Prompt #3

Below is my dataset. Create a scatter plot of Weather_C vs. Waste_kg and calculate a line of best fit. Display the graph.
[paste data]