david-istvan / models-ws-scheduler

Quick&dirty solver for the workshop allocation problem with (i) OC scheduling preferences, (ii) OC/audience overlaps, and (iii) room capacity considered.

MODELS Workshops Scheduler

Quick&dirty (Dijkstra-would-not-have-liked-it) solver for workshop allocation with (i) OC scheduling preferences, (ii) OC/audience overlaps, (iii) room capacity, and (iv) full-day/half-day preferences considered.

• Step 1: Maximize satisfaction.
• Step 2: Generate all allocations that meet the maximal satisfaction goal.

Contents

• modelsws-python.mzn - Model file for Python-based execution.
• modelsws-nopython.mzn - Model file for manual execution.
• ws2022.dzn - Data file with the 2022 workshop preferences and resource constraints.

Automated run via Python

• pip install -r requirements.txt
• python runner.py

Manual run via MiniZinc IDE

• Run with solve maximize GOAL; first to find the maximal satisfaction score.
• Then run with constraint satisfactionRatio >= [score]; %solve satisfy; to generate all equally optimal schedules.

Power by MiniZinc.