AI/ML Predictive Fulfillment Dashboard

The ability to be proactive is a crucial component for a state-of-the-art warehousing operation and best-in-class customer service. Oracle Warehouse Managment now gives you just that. With the introduction of our new Predictive Fulfillment Dashboard, we can give you a glimpse into the future and better insight into your fulfillment operations.

Utilizing our new AI/ML algorithms, this intuitive and easy-to-use dashboard allows you to view predictions for your Order Cycle Time. Basically how long it will take you to pick, pack and ship an order. Using these predictions, we can help you to identify orders that might otherwise be delayed, miss expected service levels and shipping windows, or potentially create a bottleneck in the warehouse.

We have included various charts and KPIs that give different views on what orders are predicted to be above or below your expected cycle time threshold. You also have drilling capabilities. By clicking on the order status, it will further filter and display the order types within that status.

To give you even more detailed information, the orders above your target threshold will be shown in a data grid that displays the following information:

Order Number
Status, Order Type
Predicted Time
Expected Shipping Date
Required Shipping Date

Using this detailed information in other WMS UI’s, you now have the information to do things like order detail inquires, wave releases, wave inquires, review allocations, re-prioritze task statuses and verify outbound load information.

Predictive Analytics

Predicting order cycle time and order shipping volume can be a challenge in today’s ever-changing logistics environment. Warehouse Management machine learning allows you to use historical order data with optimized machine learning algorithms and techniques to help deliver accurate predictions.

You can schedule the data creation for AIML Models in the Scheduled Jobs UI. You can also run AIML Training Templates using a scheduled job as well.

Generate Prediction Run for Order Cycle/Waiting/Processing Time

You can configure and generate a Prediction Run for Order Cycle/Waiting/Processing time in the Scheduled Jobs UI. This will predict the number of orders during a user-specified time.

The Generate Prediction Run job type is available in the create pane -> Job Type drop-down in the Scheduled Jobs UI.

Selecting the Generate Prediction Run job type from the drop-down further expands the following Job parameters:

Username
Training Run Number
Orders
Wave number - (optional)
Order Type(s)
Required Ship Date on Day

When you click on the corresponding details button in the AI/ML Prediction Run UI, the results of the model are available. By selecting the run number, you can see the results of the specific run. You can view the Prediction Run Number, type (cycle/wait/picking), order count, and handling time.

Generate Training Data in Scheduled Jobs UI

You can generate training data for Order Cycles in the Scheduled Jobs UI. This will build the data to predict the number of orders during a user-specified time.

The Generate Training Data job type is available in the create pane -> Job Type drop-down in the Scheduled Jobs UI.

AI/ML Training Template

A new UI AIML Training Template now allows you to configure specific parameters and other details required to generate a AIML model for each metric.

The following fields are mandatory for the creation of an ML Model:

Template Name
AIML Metric
Algorithm
Algorithm Parameter

In the AI/ML Training Template screen, you can train and generate AI/ML Models for future predictions. This screen is similar to the Wave Template UI and allows users to configure the specific parameters and other details required to generate an AIML model for each metric.

The following Key Performance Indicators (KPIs) are available to create in this screen.

Order Shipping Volume (time series problem)
Order Cycle Time (regression problem)
Order Waiting Time (regression problem)
Order Picking Time (regression problem)