Inventory Management for Retail — Periodic Review Policy
Implement inventory management rules based on a periodic review policy to reduce the number of store replenishments
Objective
Design inventory management rules that minimize the number of replenishments and meet store demand.
Introduction
For most retailers, inventory management systems take a fixed, rule-based approach to forecast and replenishment order management.
The objective is to build a replenishment policy to minimize ordering, holding and shortage costs.
In a previous article, we built a simulation model based on a continuous review inventory policy, assuming a normal demand distribution.
However, this kind of policy can be inefficient when you handle a large portfolio of items with different replenishment cycle lengths.
In this article, we will improve this model and implement a periodic review policy with Python to limit the number of replenishments.
💌 New articles straight in your inbox for free: Newsletter
📘 Your complete guide for Supply Chain Analytics: Analytics Cheat Sheet
SUMMARY
I. Scenario
1. Problem Statement
As an Inventory Manager of a mid-size retail chain, you are in charge of setting the replenishment quantity in the ERP.
2. Limits of the continuous review policy
What is the impact a continuous review policy on the number of replenishments when you handle several hundred SKUs?
II. Periodic Review Policy: Order-Up-To-Level (R, S)
1. Introduction of the Inventory Policy
2. Definition of the Safety Stock
3. How do you define k?
III. Example of Replenishment Policies
IV. Conclusion
1. Simulation Model with ChatGPT - "The Supply Chain Analyst"
2. Explore other Inventory Rules
3. Machine Learning for Retail Sales Forecasting
I. Scenario
1. Inventory Management for Retail
As an Inventory Manager of a mid-size retail chain, you set the replenishment quantity in the ERP.
Because your warehouse operational manager is complaining about the order frequencies, you start to challenge the replenishment rules implemented in the ERP, especially for the fast runners.
Previously, we have implemented several inventory rules based on continuous review policies.
At each time t, we check the inventory on hand (IOH):
Example for 1 SKU
We need 27 replenishment orders per year.
Question
How frequently do you need to order if you manage 2,294 SKUs?
2. Limits of the continuous review policies
You can find the source code with dummy data in my Github repository: Link
My portfolio for other Supply Chain related articles: Samir Saci
We suppose that we only have 365 days of sales
- 10 stores (STORE_1, … STORE_10)
- 1 product family (FOOD_GROCERY)
- 2,294 unique SKU (STORE ID + PRODUCT FAMILY + ITEM CODE)
Simulation 1
Let’s implement the continuous review policy for 1 SKU
- SKU: STORE_1-FOOD_GROCERY_009
- Safety stock designed with k = 3
(3 times the standard deviation of the demand distribution) - Order Quantity: Q = 3 x Average_Annual_Sales
Comments
18 replenishments in the first 100 days
Simulation 2
What if we now have 10 SKUs to manage?
Comments
54 replenishments in the first 100 days
Simulation 3
What if we have 100 SKUs to manage?
Comments
935 replenishments in the first 100 days with less than 10 days without orders
Simulation 4
What if we have 2,294 SKUs to manage?
Comments
19,267 replenishments in the first 100 days (less than 10 days without orders)
With this method, the number of replenishments is exploding when you have a large portfolio.
Solution
We will build a replenishment policy using periodic reviews.
💡 Follow me on Medium for more articles related to 🏭 Supply Chain Analytics, 🌳 Sustainability and 🕜 Productivity.
II. Periodic Review Policy: Order-Up-To-Level (R, S)
Introduction of the Inventory Policy
To solve this issue of replenishment frequency, let us introduce a periodic review policy (s, Q)
- At each period R, the inventory on hand (IOH) level will be reviewed
- Stores will order to reach an inventory level of S
To simplify the comprehension, let’s introduce some notations:
2. Definition of the Order-Up-To-Level S
How to calculate the level S:
Level S is calculated to set the inventory to cover the need during the review period and the replenishment lead time.
3. How do you define k?
The safety stock level will directly impact your performance metrics
- You fix a target for any of the two metrics (e.g: I want the Cycle Service Level to be 95%)
- You calculate k to reach this target based on the distribution of your demand (Normal, Poisson)
- You fix your reorder point
Code
💡 Follow me on Medium for more articles related to 🏭 Supply Chain Analytics, 🌳 Sustainability and 🕜 Productivity.
III. Example of replenishment policies
Example with SKU: STORE_1-FOOD_GROCERY_009
Review Period: R = 10 days
k = 1
Comments
10 replenishments vs. 18 replenishments in the first 100 days with 0 stock-outs
How many replenishments if you handle the full portfolio?
Comments
13,986 replenishments vs. 19,267 replenishments in the first 100 days with only 10 days of replenishments
💡 Follow me on Medium for more articles related to 🏭 Supply Chain Analytics, 🌳 Sustainability and 🕜 Productivity.
IV. Conclusion
Simulation Model with ChatGPT — “The Supply Chain Analyst”
Large Language Models like GPT can support the analysis of your inventory management rules and interact with users.
What is the optimal rule to minimize my ordering costs?
I introduce a custom GPT designed to automate Supply Chain Analytics tasks in this article.
This initial prototype is a proof of concept.
OpenAI’s GPTs can be used with advanced analytics models in Python scripts to create interactive analytics products.
We can imagine users uploading their sales and interacting with the agent to understand how to set an optimal rule.
For more details,
Explore other Inventory Rules
After implementing a periodic review policy, you will reduce your operating costs by
- Limiting the time windows of replenishments
This will optimize your inventory management, order preparation and transportation management resources. - Reducing the number of replenishments
This will increase the quantity per replenishment and provide opportunities for shipment consolidations for stores delivered simultaneously.
For more details about inventory management rules,
Machine Learning for Retail Sales Forecasting
Discover the power of Machine Learning for retail sales forecasting combined with Inventory Management Rules.
Based on the last Makridakis Forecasting Competitions feedback, machine learning models can reduce forecasting errors by 20% to 60% compared to benchmark statistical models.
A complete end-to-end approach would be to
- Combine the inventory rules discovered in this article with a powerful ML-based forecasting model
- Find the optimal set of parameters to reduce future stock-outs and reduce inventory costs
For more information,
About Me
Let’s connect on Linkedin and Twitter, I am a Supply Chain Engineer using data analytics to improve logistics operations and reduce costs.
If you are interested in Data Analytics and Supply Chain, have a look at my website
References
- Supply Chain Science, Wallace J. Hopp
- “Inventory Management for Retail — Stochastic Demand”, Towards Data Science, Samir Saci
- “Inventory Management for Retail — Deterministic Demand”, Towards Data Science, Samir Saci