1. Online Help
  2. MBA Radial Putaway

MBA Radial Putaway

Market Basket Analysis (MBA) Radial Putaway in Oracle WMS Cloud helps decide where to store items using specific search patterns. Setting the radius and increment correctly keeps things running smoothly and prevents slowdowns or system crashes.

  • Radius determines how far the system searches from a central point. A bigger radius covers more ground but can slow things down.
  • Increment sets the step size for each search, affecting how quickly the system finds an available spot.

The Radial Putaway method for Market Basket Analysis in WMS uses the Primary Item's location as a starting point to identify candidate locations nearby. If multiple locations are available, they are prioritized by location sequence. If no suitable location is found, the system moves to the next Primary Item location.

Performance Considerations

A max SKU evaluation limit field allows control over performance impact. To keep things running smoothly:

  • Set the radius to less than 20 to speed up searches and use space efficiently.
  • Keep the increment at 10 or below to avoid unnecessary expansion.

For example, if you set both the radius and increment to 20, the system will check 20 bays to the left and right of the center, then expand by 20 bays at each step. This setup keeps the search efficient without overloading the system.

Note: Avoid entering values higher than 20 in these fields, as larger numbers can slow down performance.

Putaway Prioritization

Putaway prioritization follows a strict order: confidence (to three decimals), then support, then lift. The values for these three concepts can be viewed by following the ‘Model Reference’ hyperlink on a Completed AI/ML model for Market Basket Analysis.

Additionally, the AI/ML Training Template that generated such a model must have the flag ‘Use for Market Basket Analysis based Putaway’ active. This ensures the model is correctly configured for putaway decisions.

  • If all values are equal, either item is acceptable.
  • If no items meet the minimum thresholds, the system falls back to the secondary putaway rule.

Data Persistence & Search Filtering

Market Basket Analysis data is stored, with each run having a unique Model Reference Number, so users can always access past results using hyperlinks.

Filtering options for support, confidence, and lift allow users to input one, two, or all three criteria in the search field. The system provides results accordingly, even if only a minimum or maximum value is given.