Sample data & analysis results

Download sample datasets and their analysis results to understand the data format and see what insights the inventory optimizer provides.

๐Ÿš€ Hybrid forecasting system

Advanced ML-powered demand forecasting with automatic method selection

๐Ÿง  SARIMAX

For continuous demand patterns with seasonality and trends

๐Ÿ“Š Croston's Method

For intermittent demand with sporadic sales patterns

โšก Smart fallback

Rolling mean for edge cases and insufficient data

Try the hybrid forecasting system by uploading your data with exogenous variables

Showing 9 of 9 dataset groups

Basic Small Dataset

Learning Dataset166 total records

A clean, small dataset perfect for learning and understanding the inventory optimization process

๐Ÿ“ฅ Input data

27 records

Original inventory data with typical business patterns

๐Ÿ“ค Analysis results

Main Results
27 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
136 records
๐Ÿ“Š Metrics (click to learn more):

Holiday Season Dataset

Seasonal Dataset117 total records

Seasonal dataset showing how inventory optimization handles holiday demand spikes and seasonal patterns

๐Ÿ“ฅ Input data

18 records

Seasonal inventory data with holiday demand patterns

๐Ÿ“ค Analysis results

Main Results
18 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
2 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
97 records
๐Ÿ“Š Metrics (click to learn more):

Sample Input Dataset

Reference Dataset165 total records

Standard reference dataset used as the baseline for inventory optimization analysis

๐Ÿ“ฅ Input data

27 records

Standard reference data showing the required input format

๐Ÿ“ค Analysis results

Main Results
27 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
135 records
๐Ÿ“Š Metrics (click to learn more):

Intermittent Demand Scenario

Forecasting Scenario1065 total records

Test scenario with 50-70% zero sales days and sporadic demand spikes - perfect for testing Croston's method

๐Ÿ“ฅ Input data

186 records

Synthetic intermittent demand data with exogenous variables (holidays, weekends, promotions)

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

Seasonal Demand Scenario

Forecasting Scenario1065 total records

Test scenario with strong weekly and monthly seasonality patterns - ideal for testing SARIMAX

๐Ÿ“ฅ Input data

186 records

Synthetic seasonal demand data with weekly/monthly patterns and exogenous variables

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

Trending Demand Scenario

Forecasting Scenario1065 total records

Test scenario with linear growth/decline trends and seasonal overlay - ideal for testing SARIMAX trend detection

๐Ÿ“ฅ Input data

186 records

Synthetic trending demand data with linear growth patterns and exogenous variables

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

Promotional Demand Scenario

Forecasting Scenario1065 total records

Test scenario with random promotional events causing 3-5x demand spikes - perfect for testing SARIMAX with exogenous variables

๐Ÿ“ฅ Input data

186 records

Synthetic promotional demand data with random 2-3 day events and price discounts

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

Stockout Demand Scenario

Forecasting Scenario1065 total records

Test scenario with inventory stockout periods causing censored demand - ideal for testing SARIMAX with stockout indicators

๐Ÿ“ฅ Input data

186 records

Synthetic stockout demand data with inventory = 0 periods and censored sales

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

Mixed Demand Scenario

Forecasting Scenario1065 total records

Comprehensive test scenario combining all demand patterns - showcases automatic method selection

๐Ÿ“ฅ Input data

186 records

Synthetic mixed demand data with various patterns and exogenous variables

๐Ÿ“ค Analysis results

Main Results
531 records
๐Ÿ“Š Metrics (click to learn more):
Store Turnover
3 records
๐Ÿ“Š Metrics (click to learn more):
Merged Metrics
531 records
๐Ÿ“Š Metrics (click to learn more):

๐Ÿงช Forecasting test scenarios

Test the hybrid forecasting system with these specialized scenarios designed to showcase different demand patterns and forecasting methods:

๐Ÿ“Š Intermittent demand

50-70% zero sales days with sporadic spikes

Expected Method: Croston's Method
Use Case: Spare parts, seasonal items

๐ŸŒŠ Seasonal demand

Strong weekly and monthly patterns

Expected Method: SARIMAX
Use Case: Fashion, holiday items

๐Ÿ“ˆ Trending demand

Linear growth/decline with seasonality

Expected Method: SARIMAX
Use Case: Growing products, declining items

๐ŸŽฏ Promotional demand

Random 2-3 day events with 3-5x demand

Expected Method: SARIMAX
Use Case: Marketing campaigns, sales events

โš ๏ธ Stockout demand

Inventory = 0 for 3-7 days (censored demand)

Expected Method: SARIMAX
Use Case: Supply chain issues, high-demand items

๐ŸŽฒ Mixed demand

Combination of all patterns

Expected Method: Multiple methods
Use Case: Real-world scenarios

๐Ÿ“ฅ Download forecast scenarios

These scenarios are available in the dataset groups. Each includes synthetic data with exogenous variables (holidays, weekends, promotions, stockouts) to test the hybrid forecasting system.

Note: Look for "Forecasting Scenario" category in the dataset groups. Legacy datasets (Learning, Seasonal, Reference) use rolling mean forecasting.

Required Data Format

Your CSV file must contain the following columns with the specified data types. Exogenous variables improve forecasting accuracy:

Column NameData TypeRequiredDescription
dateDateRequiredTransaction date (YYYY-MM-DD)
storeStringRequiredStore identifier
skuStringRequiredProduct identifier
sales_quantityNumberRequiredQuantity sold
inventory_levelNumberRequiredCurrent inventory level
unit_costNumberRequiredCost per unit
selling_priceNumberRequiredSelling price per unit
lead_time_daysNumberRequiredLead time in days
supplier_idStringRequiredSupplier identifier
categoryStringRequiredProduct category
ordered_dateDateOptionalDate when order was placed
ordered_quantityNumberOptionalQuantity ordered
order_statusStringOptionalStatus of the order
moqNumberOptionalMinimum order quantity required by supplier
is_holidayBooleanOptionalWhether the date is a holiday (improves SARIMAX accuracy)
is_weekendBooleanOptionalWhether the date is a weekend (improves SARIMAX accuracy)
is_promoBooleanOptionalWhether there is an active promotion (improves SARIMAX accuracy)
is_stockoutBooleanOptionalWhether inventory is zero (improves SARIMAX accuracy)
price_discountNumberOptionalDiscount percentage (0.0-1.0) during promotions

How to Use Dataset Groups

1

Download dataset group

Download all files for a dataset group to get both input data and analysis results together.

2

Compare input vs output

See how the original input data transforms into actionable inventory insights and recommendations.

3

Format your data

Use the input data as a template to format your own inventory data for analysis.

Ready to Optimize Your Inventory?

Upload your formatted data and start optimizing your inventory management today.

Need help? Contact us for assistance.