What This Tool Does

This planner answers those questions by scheduling my work over the next several months under two different scheduling strategies: Spreading work evenly across projects proportionally to their budget (paced), or knocking them out one by one (frontload).

Data

The planner reads from a simple JSON file listing my projects, their deadlines, and how much work remains. That’s it.

Each project needs an end date and an estimate of remaining days. Optionally, you can say that some projects will get “renewed”. This is useful for grants that get renewed and go for at least another year with a set budget.

The json file looks like this:

{
  "projects": [
    {
      "name": "Project A",
      "end_date": "2026-03-30",
      "remaining_days": 28,
      "renewal_days": 10
    },
    {
      "name": "Project B",
      "end_date": "2026-08-31",
      "remaining_days": 23
    },
    ...
  ]
}

Running the Scheduler:

Once you have your input data, you read it in and create a schedule based on the data, choosing one of the methods.

The paced approach spreads work across all active projects, making steady progress on everything simultaneously.

This is usually how researchers do their work. It prioritizes projects by deadline (earliest due date first) but tries to make sure every project gets worked on every two weeks.

The frontload approach completes projects one at a time in deadline order. The point of this method is just to provide a bounding case and give a better sense of priority and when exactly I’d need more work if I were trying to knock everything out as soon as possible.

Both approaches assign a full day of work for each project—a simplifying assumption I’m comfortable with.

Let’s run both approaches and compare them:

from datetime import date
import pandas as pd

# Import planner modules
from planner.scheduler import Scheduler
from planner.analysis import (
    load_projects,
    compute_weekly_availability,
    compute_monthly_unassigned_days,
    create_calendar_heatmap,
    create_availability_plot,
)

# Configuration
CONFIG_FILE = "projects.json"
NUM_WEEKS = 52

# Load and schedule
projects = load_projects(CONFIG_FILE)
scheduler = Scheduler(projects, start_date=date.today())
schedule_paced = scheduler.create_schedule(num_weeks=NUM_WEEKS, method="paced")
schedule_frontload = scheduler.create_schedule(num_weeks=NUM_WEEKS, method="frontload")

# Get statistics
stats_paced = scheduler.get_statistics(schedule_paced)
stats_frontload = scheduler.get_statistics(schedule_frontload)

# Compute availability
paced_availability = compute_weekly_availability(schedule_paced, NUM_WEEKS)
frontload_availability = compute_weekly_availability(schedule_frontload, NUM_WEEKS)