Creating a Machine Learning Model in WMS

To build and use a machine learning model in WMS, follow the structured process below:

Step 1: Create a Scheduled Job

Initiate the process by scheduling a job to prepare the necessary training data.

Step 2: Review the Training Data

Use the AIML Training Data screen to review and validate the dataset generated by the scheduled job.

Step 3: Define the Model

Configure your model using parameters and filters. Assign the model to an AI/ML Training Template, where you can also set a default template for reuse. This can be done in AIML Training Template UI.

Step 4: Run Predictions

Execute the model using the AIML Prediction Run UI to generate predictions. The system will capture and display the results from this run.

Step 5: Deploy and Refine the Model

Deploy the model to begin using it in production. You can also retrain and adjust the model as needed based on results and performance insights.

Available AI/ML Screens in WMS

• AIML Training Data: View the structured training dataset prepared via scheduled job.
• AIML Training Template: Create, configure, and manage training templates; set defaults for ease of reuse.
• AIML Training Date Rule Hdr: Create custom seasonality based date ranged as need for you warehouse needs. This only applies for Market Basket Analysis.
• AIML Models UI: View models created from your training templates, including parameters and status.
• AIML Model’s Training Logs UI: View relevant logging information related to model creation.
• AIML Prediction Run UI: Run models and view detailed prediction results.
• AIML Predictive Dashboard: Visualize recent prediction outcomes (currently supports Order Cycle Time only).