replenishment

replenishment is a Python library for stock replenishment simulation, policy optimization, and decision visualization.

It is designed for teams that want to:

• simulate replenishment policies against historical or synthetic demand
• optimize mean-forecast safety stock, k*RMSE, and percentile-target policies
• compare policies visually on replenishment time series
• move cleanly between CSV/DataFrame inputs and simulation configs

Supported Python Versions

The project is tested on Python 3.10, 3.11, 3.12, and 3.13.

Yes, Python 3.13 is out and stable. Python 3.13.0 was officially released on October 7, 2024.

Installation

Install from PyPI

pip install replenishment

Create a Development Environment with uv

Choose any supported Python version. 3.13 is the default below.

uv python install 3.13
uv venv --python 3.13
source .venv/bin/activate
uv sync --extra dev

To work on another supported version, replace 3.13 with 3.10, 3.11, or 3.12.

Run the test suite with:

uv run pytest

Quickstart

This example optimizes a mean-forecast safety stock factor on a backtest window and applies the learned policy on the forecast horizon.

from replenishment import (
    generate_standard_simulation_rows,
    optimize_point_forecast_policy_and_simulate_actuals,
    plot_replenishment_decisions,
    replenishment_decision_rows_to_dataframe,
    split_standard_simulation_rows,
    standard_simulation_rows_to_dataframe,
)

rows = generate_standard_simulation_rows(
    n_unique_ids=1,
    periods=18,
    start_date="2031-01-01",
    frequency_days=30,
    forecast_start_period=10,
    history_mean=52,
    history_std=6,
    forecast_mean=48,
    forecast_std=5,
    lead_time=2,
    initial_on_hand=30,
    current_stock=30,
    seed=7,
)
backtest_rows, eval_rows = split_standard_simulation_rows(rows)

optimized, _, decision_rows = optimize_point_forecast_policy_and_simulate_actuals(
    backtest_rows,
    eval_rows,
    candidate_factors=[0.8, 0.9, 1.0],
)

rows_df = standard_simulation_rows_to_dataframe(rows, library="pandas")
decision_df = replenishment_decision_rows_to_dataframe(decision_rows, library="pandas")

example_id = decision_df["unique_id"].iloc[0]
plot_replenishment_decisions(
    rows_df,
    decision_df,
    unique_id=example_id,
    title="Mean forecast + safety stock (optimized)",
    decision_style="line",
)

Example Workflows

Mean Forecast + Safety Stock

Optimize a point-forecast safety stock factor on historical rows, then inspect the resulting replenishment decisions on the forecast horizon.

k*RMSE + Forecast-Level Buffering

Use k*RMSE for the base safety stock and optionally increase it when lead-time forecast quantities rise above a baseline. The README plot below uses three articles with progressively steeper ramps so the extra safety stock is visible directly on the replenishment timeline.

Percentile Target Optimization

Optimize the percentile target per article on backtest rows, then simulate and visualize the chosen forecast target on the evaluation horizon.

Notebooks

Runnable walkthroughs live in notebooks/:

• notebooks/mean_forecast_safety_stock_example.ipynb: mean forecast safety stock optimization
• notebooks/k_rmse_safety_stock_optimization_example.ipynb: k*RMSE optimization and forecast-level buffering
• notebooks/percentile_optimization_example.ipynb: percentile-target optimization
• notebooks/mean_forecast_policy_variants_example.ipynb: compare policy variants
• notebooks/generated_data_example.ipynb: synthetic data generation
• notebooks/stock_replenishment_example.ipynb: table-oriented data loading workflow

Development and Contribution

Contributions are welcome. If you open a PR, keep it easy to review:

• add or update tests for behavior changes
• update notebooks, README examples, and plot assets when public behavior changes
• include before/after plots when you change plotting behavior
• keep PRs focused rather than bundling unrelated refactors

See CONTRIBUTING.md for setup, test, and pull request guidance.

Good Areas for Collaboration

If you want to make the project more useful and more credible to other teams, these are high-leverage contribution areas:

• richer policy variants, such as minimum order quantities, case-pack constraints, and promotion-aware demand handling
• benchmark datasets and scenario packs for retail seasonality, long lead times, and stockout-heavy stress tests
• more notebook walkthroughs that compare service-level objectives and business tradeoffs
• packaging and docs polish, including API docs, changelogs, and release notes
• plotting improvements for aggregate dashboards and multi-SKU diagnostics

Project Links

• Homepage: https://github.com/janrth/replenishment
• Issues: https://github.com/janrth/replenishment/issues
• Changelog: CHANGELOG.md
• License: LICENSE