Dark Mode
Inventory Optimisation Dataset
Retail & Consumer Behavior
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This synthetic dataset is designed for practising inventory management and demand forecasting within a retail context. It provides realistic daily data across multiple stores and products, encompassing key attributes such as sales figures, inventory levels, pricing, and external influencing factors like weather conditions, promotional activities, and public holidays. The dataset is particularly well-suited for machine learning tasks, including demand forecasting, dynamic pricing, and inventory optimisation. It allows data scientists to explore various time series forecasting techniques and analyse the impact of external variables on retail sales performance.
Columns
- Date: Daily records spanning from 1st January 2022 to 1st January 2024.
- Store ID: Unique identifiers for individual retail stores, with 5 distinct values.
- Product ID: Unique identifiers for specific products, comprising 20 distinct values.
- Category: Classifies products into categories such as Furniture, Toys, and other miscellaneous groups, with 5 unique categories.
- Region: Indicates the geographical area of the store, featuring 4 unique regions including East and South.
- Inventory Level: Represents the stock available at the start of each day, with values typically ranging from 50 to 500 units.
- Units Sold: The number of units sold during the day, generally ranging from 0 to 499 units.
- Units Ordered: The quantity of units ordered, with values typically between 20 and 200.
- Demand Forecast: Predicted demand based on past trends, with values ranging from approximately -9.99 to 519.
- Price: The price of the product, ranging from 10 to 100.
- Discount: The discount applied, with values between 0 and 20.
- Weather Condition: Daily weather conditions that can affect sales, including 4 unique conditions such as Sunny and Rainy.
- Holiday/Promotion: Binary indicators (0 or 1) signifying the presence of a holiday or a promotion.
- Competitor Pricing: Pricing data from competitors, typically ranging from approximately 5.03 to 105.
- Seasonality: A seasonal indicator, with 4 unique values like Spring and Summer.
Distribution
The dataset is formatted as a CSV file named
retail_store_inventory.csv and has a file size of 6.19 MB. It contains over 73,000 rows of daily data. All 15 columns are valid, with no reported missing or mismatched values.Usage
This dataset is ideal for:
- Time Series Demand Forecasting: Building models to predict daily product demand across various stores using historical sales and inventory data.
- Inventory Optimisation: Analysing sales trends to minimise stockouts and reduce overstock situations.
- Dynamic Pricing: Developing pricing strategies influenced by demand, competitor pricing, and discount levels to maximise revenue.
- Exploratory Data Analysis (EDA): Conducting initial analysis of sales trends, visualising data, and identifying patterns.
- Pricing Analysis: Investigating the impact of discounts and competitor pricing on sales performance.
Coverage
The dataset provides daily records from 1st January 2022 to 1st January 2024. Its geographical scope includes multiple regions, such as East and South, corresponding to various store locations. Product coverage spans categories like Electronics, Clothing, Groceries, Furniture, and Toys.
License
CC0: Public Domain
Who Can Use It
This dataset is primarily intended for data scientists and analysts who are involved in:
- Developing and assessing machine learning models for forecasting.
- Optimising retail operations and improving supply chain performance.
- Investigating how external factors, such as weather and holidays, influence sales.
Dataset Name Suggestions
- Retail Inventory Forecasting
- Store Sales Demand Data
- Daily Retail Demand
- Inventory Optimisation Dataset
- Retail Supply Chain Analytics
Attributes
Original Data Source: Inventory Optimisation Dataset
QUALITY