The MLE Broadcast Schedule Generator

Automatically builds an optimal and equitable broadcast schedule for MLE matches, replacing time-consuming manual scheduling efforts with a precise, constraint-based optimization approach.

What You'll Need

• Python 3 installed on your system.

• Required Python packages (install using pip install -r requirements.txt):

  • pandas
  • ortools

• Two inputs provided as CSV files:

  1. League Schedule CSV (league_schedule.csv)

     • This schedule is published by League Ops and is fixed prior to the season.
     • Columns required:
       • Match #: (integer ID) uniquely identifies matches.
       • Home Team: franchise + skill group + game mode (e.g., "Puffins AL 3s").
       • Away Team: franchise + skill group + game mode (e.g., "Foxes AL 3s").
       • Game Mode: either "3s" or "DBLs".
       • Skill Group: the league division, typically "AL", "CL", "ML", "FL", or "PL".

  2. Broadcast Slots CSV (broadcast_slots.csv)

     • Broadcast slots are determined by production availability.
     • Columns required:
       • Match #: indicates the match number that the broadcast slot corresponds to/from the league schedule.
       • Day: typically "Saturday" or "Sunday".
       • Time: scheduled broadcast start time, e.g., "7 PM".

• Recommended directory structure:

.
├── inputs
│   ├── league_schedule.csv
│   └── broadcast_slots.csv
├── requirements.txt
└── generate-schedule.py

How to Use It

1. Ensure you have the correct input files and place them under the inputs/ folder as described above.

2. Install dependencies:

pip install -r requirements.txt

3. Run the scheduling script:

python generate-schedule.py

4. Check your output:

• The script generates a broadcast_schedule.csv file in the working directory, containing tuples of assigned matches with the streaming slots.

Your output CSV (broadcast_schedule.csv) will have the following columns:

• Match #, Home Team, Away Team, Game Mode, Skill Group, Day, Time.

Example Row in broadcast_schedule.csv:

1,Puffins AL 3s,Foxes AL 3s,3s,AL,Saturday,7 PM

How It Works

The broadcast scheduling problem is, at heart, a constrained optimization problem: Select matches from a fixed league schedule and evenly assign them into available broadcast slots. The aim is an optimal and balanced distribution of appearances among all franchises, subject to practical constraints.

1. Problem Formalization

The tool models this problem mathematically, utilizing constraint programming.

• Objective: Minimize the maximum number of streamed appearances of any franchise/skill group combination across the season. This objective seeks to make coverage as equitable and fair as possible.

• Constraints:

  • Matches scheduled by the League Operations cannot be changed – we can only select certain matches from them to appear on stream.
  • Each broadcast slot must contain exactly one assigned match; no slots may remain empty or contain multiple matches.
  • No franchise (within a single skill group and game mode) may appear more than 3 times on stream throughout the entire season.
  • Matches can be broadcast only at their corresponding designated match number slots.

2. Mathematical Formulation

• Let ( x_{ij} ) be a binary variable, equal to 1 if match ( i ) is assigned to broadcast slot ( j ), and equal to 0 otherwise.
• Let ( appearances_k ) be the number of streamed appearances for franchise/skill group combinations ( k ): [ appearances_k = \sum_{i=1}^{M}\sum_{j=1}^{N} x_{ij} \cdot team_in_match(i,k) ] where ( team_in_match(i,k) ) checks whether franchise & skill group combination ( k ) appears in match ( i ).

We optimize by minimizing the largest ( appearances_k ): [ \text{Objective: } \min \max(appearances_1, appearances_2, ..., appearances_K) ]

Subject to:

• Slot usage constraint: Each slot exactly once: [ \sum_{i=1}^{M} x_{ij} = 1 \quad\forall j \in {1,...,N} ]

• Match schedule constraint: ( x_{ij} = 0 ) if the match number of schedule and slot do not match.

These variables, constraints, and objectives are directly encoded into the CP-SAT solver from Google's OR-Tools. OR-Tools efficiently solves large-scale constraints optimization problems and guarantees that the solution meets the constraints provided.

3. Implementation

The implementation (generate-schedule.py) performs the following tasks:

• Reads input CSV data using pandas.
• Creates optimization model variables and constraints using the OR-Tools CP-SAT model.
• Runs the solver to find an optimal assignment.
• Outputs the optimal schedule to a CSV file (broadcast_schedule.csv) in a readable and convenient format for production usage.