AI Tools for Researchers

Working with AI in Supply Chain Research

AI can help with your research if you know how to use it right. This guide shows you what actually works when you're analyzing data, reading papers, or trying to solve problems.

Get Started Browse Resources

What You'll Learn

Whether you're just starting out or already using AI, here's what you need to know to actually get useful results from it.

Prompt Engineering

How to write prompts that actually give you useful answers. The difference between getting garbage and getting insights is usually in how you ask.

  • Structuring effective prompts
  • Chain-of-thought reasoning
  • Few-shot learning examples

Data Analysis

Getting AI to help you make sense of your datasets. Finding patterns you might miss and turning messy supply chain data into something you can actually understand.

  • Cleaning and structuring data
  • Pattern recognition
  • Generating visualizations

Literature Review

Reading 50 papers takes forever. AI can help you quickly summarize what each one says, pull out the important bits, and spot where there's still work to be done.

  • Summarizing academic papers
  • Comparing methodologies
  • Finding research gaps

Core Concepts

Knowing a bit about how these tools work makes it easier to get what you want out of them

1 What are Large Language Models?

LLMs like ChatGPT, Claude, and Gemini are AI systems trained on massive amounts of text. They understand context, generate human-like responses, and can help with complex reasoning tasks.

Think of them as:
  • • Very smart autocomplete (but way more capable)
  • • Pattern matchers trained on billions of examples
  • • Tools that understand both what you say and what you mean

2 Strengths and Limitations

Good at:
  • • Summarizing information
  • • Brainstorming ideas
  • • Explaining concepts
  • • Writing code
  • • Structuring data
Watch out for:
  • • Hallucinations (making things up)
  • • Outdated knowledge
  • • Math errors
  • • Bias in training data
  • • Can't browse the web (usually)

3 Prompt Engineering Basics

The way you ask questions matters. Good prompts are clear, specific, and provide context.

Weak prompt:
"Tell me about supply chains"
Strong prompt:
"Explain how blockchain technology can improve transparency in coffee supply chains, focusing on smallholder farmers in Ethiopia. Include 3 real-world examples and potential challenges."

4 Ethical Use

AI is powerful but it's not perfect. Be honest about when you use it, double-check what it tells you, and think about the bigger picture.

  • Always cite when AI helps with your work
  • Verify facts and figures independently
  • Don't share confidential or sensitive data
  • Understand that AI can be biased
  • Use AI to augment, not replace, critical thinking

Practical Techniques

Real prompts and workflows you can use in your supply chain research

Technique

Analyzing Supply Chain Data

AI can help you clean up messy data, spot patterns you might miss, and figure out what the numbers are actually telling you about your supply chain.

Example Prompt:
I have a CSV file with columns: supplier_name, country, product_category, co2_emissions_kg, cost_usd, delivery_days. Please help me: 1. Identify which suppliers have the highest emissions per dollar spent 2. Find patterns between country and delivery times 3. Suggest visualizations to show the relationship between cost and emissions 4. Recommend 3 suppliers to investigate for sustainability improvements Data sample: [paste first 10 rows here]
Pro tip:
Break complex analysis into steps. Ask the AI to explain its reasoning. Always verify statistical calculations independently.
Technique

Accelerating Literature Review

Reading academic papers takes forever. AI can give you the main points fast, help you compare how different studies approached the same problem, and show you what questions haven't been answered yet.

Example Prompt:
I'm researching circular economy models in the textile industry. I have 5 papers below. For each paper, extract: 1. Main research question 2. Methodology used 3. Key findings (3 bullets max) 4. Limitations mentioned by authors 5. How it relates to smallholder producer inclusion Then create a comparison table showing which methodologies were used and what gaps exist. Paper 1: [paste abstract and conclusion] Paper 2: [paste abstract and conclusion] ...
Warning:
AI summaries can miss nuance. Always read the original papers for anything you'll cite. Use AI to organize, not replace, your literature review.
Technique

Structured Problem Solving

When you're stuck on a complicated problem, AI can help you break it into smaller pieces and work through different ways to solve it. Like talking through your problem with someone who asks good questions.

Example Prompt:
I'm working on reducing packaging waste for a local bakery network. Context: - 12 bakeries across the region - Currently using single-use plastic packaging - Average 500 customers per bakery per day - Budget constraint: max 15% increase in packaging costs Please help me: 1. Break this problem into smaller components 2. List potential circular economy solutions 3. For each solution, estimate feasibility (low/medium/high) and explain why 4. Identify what data I'd need to collect to evaluate each option 5. Suggest 3 pilot programs I could propose Think step by step and explain your reasoning.
Best practice:
Add "Think step by step" or "Explain your reasoning" to get more detailed analysis. Use AI to generate options, then apply your own judgment.

Tools & Resources

Tools and resources that are actually useful for supply chain research

AI Platforms

  • ChatGPT
    OpenAI's conversational AI. Great for general research tasks.
  • Claude
    Anthropic's AI. Excellent for long documents and analysis.
  • Gemini
    Google's AI. Integrates with Google Workspace.

Coding Assistants

  • GitHub Copilot
    AI pair programmer. Free for students.
  • Cursor
    AI-first code editor built on VS Code.
  • JetBrains AI
    Built into PyCharm, IntelliJ, and other IDEs.

Learning Resources

Explore Our GitHub Resources

We maintain a collection of open-source AI tools and resources specifically for supply chain research. All code is available for you to use, modify, and learn from.

  • Prompt templates for common research tasks
  • Sample datasets for practice
  • Python notebooks with worked examples
  • Tools for evaluating AI outputs
Browse Repositories
Featured Repositories:
big-AGI
Advanced AI interface with multiple model support
Prompt Library
Ready-to-use prompts for research tasks
ChainForge
Visual prompt engineering toolkit

Common Questions

Answers to questions students frequently ask about using AI in research

Can I use AI for my thesis or research paper?

Yes, but be transparent about it. Many universities now allow AI use as long as you cite it properly and don't claim AI-generated content as your own original work. Check your institution's academic integrity policy. Use AI as a tool to help you think, not to think for you.

How do I cite AI in my research?

Different citation styles have different approaches. Generally, you should mention the AI tool, the company, the date, and how you used it. Example:

OpenAI. (2024). ChatGPT (GPT-4) [Large language model]. Used to help analyze survey data patterns on March 15, 2024. https://chatgpt.com

What if the AI gives me wrong information?

AI makes mistakes. Always verify facts, statistics, and citations independently. Use AI to generate ideas and structure thinking, but check everything important against reliable sources. If you're unsure, ask the AI to explain its reasoning or provide sources (though be aware it might make up sources too).

Is my data safe when I use AI tools?

It depends on the tool. Some AI companies use your inputs to train their models, others don't. Never share confidential data, personal information, or anything covered by an NDA. If working with sensitive supply chain data, consider using tools with enterprise privacy settings or run models locally.

Which AI tool should I use?

Different tools have different strengths. ChatGPT is versatile and widely used. Claude is better for long documents. Gemini integrates well with Google tools. Try a few and see what works for your workflow. Many are free or offer student discounts.

Ready to Start Using AI in Your Research?

Connect with other students and researchers who are figuring this out too. Share what works, ask questions, and learn from each other.

Join Our Discord Attend a Workshop