Predict Locations for Cycle Count using AI/ML

To help boost warehouse efficiency and improve the cycle counting process, we have now added the ability in Warehouse Management to analyze your inventory and create cycle count tasks powered by AIML. This new feature is available in the existing AIML UI as Intelligent Cycle Count. This flow is similar to existing AIML metrics available within the WMS.

Introducing Intelligent Cycle Count

To help automate Intelligent Cycle Count, we have introduced a new scheduled job, Generate Prediction Run for Intelligent Cycle Count. This scheduled job will allow you to generate a prediction run and automatically create cycle count tasks based on the output.Refer to the Extract Scheduled Job document for more information on the job parameters.

Another new scheduled job, “Generate Training Data for Intelligent Cycle Count” is introduced that allows you to set the amount of data you would like to be used in your Intelligent Cycle Count predictions.

A new Intelligent Cycle Count Training Template is available in the AI/ML Training Template UI consisting of fields like, Template Name, AIML Metric, Default Flag, Training Filters, AIML Algorithm, Algorithm Parameters, Last Trained Model reference, Create Timestamp, Mod Timestamp, Create user and Mod User.

  • This Intelligent Cycle count data can be viewed from the Metrics drop-down on the AIML Training Data UI.

The Intelligent Cycle Count is added to AI/ML Model UI. On invoking the details button from the Prediction Run UI for the specific prediction run number, the system opens the AI/ML Intelligent Cycle Count sub-screen and displays the output. On invoking the “Predict” button, a new pop-up is displayed consisting of field like Location Type, Area, Allocation Zone, and Replenishment Zone.

Updates to Task Creation Template 

The Task Creation Template UI is now updated to reference the Intelligent Cycle Count predictions to generate tasks with a high probability. You will now be able to check a flag (Use Intelligent Cycle Counting) that enables the AIML based cycle counting for task creation. Once enabled, users will be able to input a from/to threshold/probability field, prediction run number and/or a training template name. 

NOTE: This flag is NOT enabled by default.

Steps to Enable

  1. Go to the Schedule Job module > create a scheduled job to generate the training data for Intelligent Cycle Count.
  2. Navigate to the AI/ML Training Data to verify that the training data contains the correct data.
  3. Create an Intelligent Cycle Count Training Template in the AI/ML Training Template UI.
  4. After creating your Intelligent Cycle Count template with all the mandatory fields completed, click Train Template.
  5. Navigate to the AI/ML Model UI and select the Prediction Run you would like to view via Prediction Run UI.   You can now view the results of your successful Intelligent Cycle Count and create tasks.

Key Resources