https://github.com/lukestorry/inventory-optimisation

Demo for a talk about inventory modelling and strategy optimisation in Python
https://github.com/lukestorry/inventory-optimisation

Last synced: 3 months ago
JSON representation

Demo for a talk about inventory modelling and strategy optimisation in Python

• Host: GitHub
• URL: https://github.com/lukestorry/inventory-optimisation
• Owner: LukeStorry
• Created: 2022-01-17T12:58:55.000Z (over 4 years ago)
• Default Branch: main
• Last Pushed: 2022-01-24T15:30:19.000Z (over 4 years ago)
• Last Synced: 2025-01-16T00:26:06.716Z (over 1 year ago)
• Language: HTML
• Size: 232 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Inventory-Optimisation-Demo

#### Example code for a talk about inventory modelling and strategy optimisation in Python.

Using:

- [simpy](https://simpy.readthedocs.io/en/latest/) for discrete event simulation

- [matplotlib](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.html) for graphs

- [epsilon-greedy algorithm](https://en.wikipedia.org/wiki/Multi-armed_bandit#:~:text=Epsilon%2Dgreedy,-strategy) for input optimisation

- [reveal-js](https://revealjs.com/) framework for the presentation

- [excalidraw](https://excalidraw.com/) for diagrams

>Slides viewable [here](https://lukestorry.github.io/Inventory-Optimisation/slides.html)

---

## Overview

![Discrete Events Diagram](assets/simulation.svg)

---

## Output

#### With only the Purchasing event

![availability diagram - only purchasing (goes up)](assets/availability_1_only_purchasing.png)

#### After adding aged stock clearout

![availability diagram - with stock clearout (goes up and down blockily)](assets/availability_2_with_stock_clearout.png)

#### After adding sales

![availability diagram - with clearout and sales (blocks are smoothed off by sales - realistic diagram)](assets/availability_3_with_clearout_and_sales.png)

#### Result of greedy-epsilon optimisation

![availability diagram - optimised (minimal spending, no out-of-stock)](assets/availability_optimised.png)

#### Comparison of different Epsilon values (reward per iteration)

![epsilon comparison](assets/epsilon_comparison.png)

---

## Setup & running

Install required packages: `pip install -r .\requirements.txt`

Run the scripts with: `python simulation.py` or `python optimisation.py`

---

## Creating your own optimiser

Either use the simple greedy-epsilon-agent-based optimiser as a starting point, or in a new file:

```python

from simulation import PurchaseOrder, Simulation

# Initial inputs

purchase_orders = [PurchaseOrder(0, 20), PurchaseOrder(100, 100)]

for iteration in range(100):

    simulation = Simulation(purchase_orders)

    simulation.run()

    simulation.plot()

    # Customise this calculation depending on requirements

    cost = sum(simulation.availabilities)

    

    # Calculate a new set of more-optimal inputs

    purchase_orders = [] 

```