Skip to content

README.md

Latest commit

 

History

History
115 lines (80 loc) · 1.75 KB

README.md

File metadata and controls

115 lines (80 loc) · 1.75 KB
Raw

Student Internship Assignment (ASP + Clingo)

Assigns students to internship companies based on preferences using Answer Set Programming (ASP) with Clingo.

📌 Overview

Input:

  • Students
  • Student preferences (optional)
  • Companies with capacities

Goal: Assign every student while maximizing preference satisfaction and respecting constraints.

🧱 Structure

project/
├── main.py
├── rules.json
├── utils/
├── data/
├── model.lp
└── README.md

⚙️ Pipeline

  1. Load data from /data
  2. Clean & normalize names, generate IDs (vorname_nachname)
  3. Merge students with preferences
  4. Generate ASP files (data.lp, config.lp)
  5. Run Clingo to compute optimal assignment
  6. Parse results and map back to names
  7. Export assignments.csv

🧩 Rules (rules.json)

Configure assignment behavior without changing code:

{
  "min_per_company": 7,
  "max_prio": { "Caritas": 2 },
  "fill_first": { "Gesobau": 20, "ABB": 10 }
}

🚀 Usage

Install

pip install pandas

Install Clingo: https://potassco.org/clingo/

Run

python main.py

Options

python main.py --timeout 120 --debug
  • --timeout → solver time limit (seconds)
  • --debug → verbose logging

📄 Output

assignments.csv

Vorname,Nachname,company
Florian,Abraham,ABB
...

⚠️ Notes

  • All students are assigned (even without preferences)
  • Missing preferences = 0
  • Rules use fuzzy company name matching

🧠 Tech

  • Python + Pandas
  • Clingo (ASP solver)

✅ Status

✔ Modular structure
✔ Configurable rules
✔ Logging & CLI support
✔ Optimal assignment generation