How to Master Mathematical Modeling: Practical Tips for Students

Yesterday I returned to my alma mater, East China Normal University, to share my modeling experiences with junior students, hoping to help them avoid pitfalls and grow faster in the world of mathematical modeling.

1. Practice Is King – Learn by Doing

Mathematical modeling is an engineering‑style application of mathematics. Only through real‑world practice can you truly master it. Many students spend a lot of time reading textbooks and theory but still feel lost when faced with a modeling problem because the core is solving practical issues, not just book knowledge.

Try early hands‑on work, even if your first models are imperfect. Train with past competition problems or real‑life cases such as cafeteria queue analysis, local traffic congestion, or personal study‑plan optimization. Each practice session makes you more proficient.

2. Pursue Growth – Challenge Yourself Continuously

The learning process is cumulative; persistence leads to noticeable progress. Remember that modeling inherently involves trial‑and‑error, adjustment, and optimization.

Start with simple models (basic functions or sequences) and gradually move to differential equations and stochastic processes. After each project, summarize what went well, what could improve, and explore better solutions or more efficient computations.

Engage with teammates and senior students; discuss and seek advice to broaden your perspective and learn new problem‑solving methods.

3. Learn by Application – Focus on Use, Not Memorization

Avoid getting trapped in pure formula derivation; instead, consider how each mathematical tool applies to real problems. For example, think about least‑squares fitting for data, or differential equations for population or epidemic modeling. The goal is to solve real‑world issues, not just score high on exams.

Study with problems in mind, not problems after study.

4. Keep Curiosity Alive – Let Questions Drive Exploration

Modeling is full of hidden complexity. Ask “why” repeatedly: why a certain algorithm fits, why data trends appear, why a model describes reality. Curiosity pushes deeper research and richer understanding.

Also, explore modeling applications across fields such as finance, medicine, environmental science, and artificial intelligence to broaden your horizon.

5. Leverage Tools – Let Technology Boost Your Modeling

Modern modeling relies heavily on software tools for data processing, modeling, and solving. Mastering the right tools dramatically improves efficiency.

Programming languages : Python (with Pandas, NumPy, SciPy, Matplotlib) and MATLAB are mainstream.

Visualization tools : Matplotlib, Seaborn, Plotly for Python; TikZ for LaTeX; GeoGebra; plus diagram tools like draw.io or XMind.

Writing tools : LaTeX for papers; Markdown with editors such as Typora or VS Code for reports.

AI tools : Use ChatGPT, DeepSeek, etc., to generate code, optimize algorithms, and explore data features.

Effective tool use saves time, reduces repetitive work, and yields more accurate models, especially under competition time constraints.

6. Compete Early – Learn by Racing

Participating in modeling contests (e.g., MCM/ICM, CUMCM) accelerates learning. The primary value is the learning experience, not winning. Early participation builds experience, teamwork, time management, and writing skills.

Even if your first attempts are imperfect, the experience lays a solid foundation for future competitions.

Mathematical modeling is a comprehensive discipline that demands continuous practice, reflection, and optimization.