Main Features
Multiple Demand Lines
Each SKU can have multiple associated demand lines representing different forecast types: market forecasts, actual orders, tenders, budget, and machine learning forecasts. This allows comparison of different demand indicators for a single SKU to enhance planning accuracy. Demand lines are versioned, allowing planners to track changes over time without overwriting prior values. You can also create a customized, new plan type.
Machine Learning and Benchmarking
PLAIO generates forecasts using advanced ML models trained on historical sales data that automatically capture patterns such as trends, seasonality, and demand shifts. In parallel, it produces a benchmark forecast using a rolling average as a reference point for comparison.
Flexible Forecast Granularity
The system supports flexible levels of forecast granularity, allowing plans to be created and maintained at different dimensions such as product, market, or customer group. These plans can then be aggregated or broken down depending on the configuration.
Performance & Analysis
Forecast KPIs
PLAIO evaluates forecast quality using two complementary metrics:
Error: Quantifies the overall magnitude of inaccuracy
Bias: Identifies systematic tendencies (over-forecasting or under-forecasting)
The metrics are visualized on each demand line, allowing planners to quickly assess reliability and make adjustments.
Forecast Accuracy Tracking
PLAIO measures forecast accuracy by comparing forecasted market quantities to actual sales over a 12-month period using the formula: abs(1 - (Total Absolute Error / Total Sale Quantity)) Γ 100.
Planning Support
On-Demand Updates
Users can trigger on-demand updates at any time, allowing both forecast types to be recalculated on the latest available data. This ensures planners work with current information while retaining control over when new calculations are performed.
Exception Alerts
Standard exception alerts include:
Missing Demand: Shows non-discontinued items without demand forecasts
Forecast Reliability: Shows items with high forecast error and bias
Volume and Value Visibility
Demand Planning supports viewing the demand plan in both quantity (volume) and value. Value is calculated as a derived view based on planned quantities multiplied by the applicable unit price.
Integration & Data Management
Sales and Inventory History
Sales history serves as the basis for statistical forecast calculations, while inventory history is used for informational purposes showing how it has evolved over time.
Make-to-Stock vs. Make-to-Order Support
The system is designed to accommodate both MTS and MTO strategies effectively, which can be defined down to each SKU level.
β
Definitions
If you click on the ... on any column, you can choose your columns. Below is a breakdown of each column and what it represents.
Custom columns can be added if needed.
Column Name | Description |
SKU No | Unique identifier for the Stock Keeping Unit (SKU). |
Name | Full name of the product variation, including format and dosage. |
Demand Segment | SKU + Market ID, to allow for high granularity when planning. |
Product Family | Grouping of products by family or product line. |
Series Type | Forecast type. Market, DemandML, Benchmark, Customer Orders, or custom demand segments. |
Location | Physical location in the supply network. |
Start Date | Items lifecycle startdate. |
Market | Geographic or commercial market where the product is sold (e.g., Spain, Sweden). |