Oracle Warehouse

Management Cloud

AI/ML User Guide

Oracle Warehouse Management Cloud
AI/ML User Guide

Copyright © 2024, Oracle and/or its affiliates.

Author: Oracle WMS Cloud Product Team

This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected
by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate,
broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering,
disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited.

The information contained herein is subject to change without notice and is not warranted to be error-free. If you find any errors, please report them
to us in writing.

If this is software, software documentation, data (as defined in the Federal Acquisition Regulation), or related documentation that is delivered to the
U.S. Government or anyone licensing it on behalf of the U.S. Government, then the following notice is applicable:

U.S. GOVERNMENT END USERS: Oracle programs (including any operating system, integrated software, any programs embedded, installed, or
activated on delivered hardware, and modifications of such programs) and Oracle computer documentation or other Oracle data delivered to or
accessed by U.S. Government end users are "commercial computer software," "commercial computer software documentation," or "limited rights
data" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, the use, reproduction,
duplication, release, display, disclosure, modification, preparation of derivative works, and/or adaptation of i) Oracle programs (including any
operating system, integrated software, any programs embedded, installed, or activated on delivered hardware, and modifications of such programs),
ii) Oracle computer documentation and/or iii) other Oracle data, is subject to the rights and limitations specified in the license contained in the
applicable contract. The terms governing the U.S. Government's use of Oracle cloud services are defined by the applicable contract for such services.
No other rights are granted to the U.S. Government.

This software or hardware is developed for general use in a variety of information management applications. It is not developed or intended for use
in any inherently dangerous applications, including applications that may create a risk of personal injury. If you use this software or hardware in
dangerous applications, then you shall be responsible to take all appropriate fail-safe, backup, redundancy, and other measures to ensure its safe
use. Oracle Corporation and its affiliates disclaim any liability for any damages caused by use of this software or hardware in dangerous applications.

Oracle®, Java, and MySQL are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.

Intel and Intel Inside are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and are trademarks
or registered trademarks of SPARC International, Inc. AMD, Epyc, and the AMD logo are trademarks or registered trademarks of Advanced Micro
Devices. UNIX is a registered trademark of The Open Group.

This software or hardware and documentation may provide access to or information about content, products, and services from third parties. Oracle
Corporation and its affiliates are not responsible for and expressly disclaim all warranties of any kind with respect to third-party content, products, and
services unless otherwise set forth in an applicable agreement between you and Oracle. Oracle Corporation and its affiliates will not be responsible
for any loss, costs, or damages incurred due to your access to or use of third-party content, products, or services, except as set forth in an applicable
agreement between you and Oracle.
Oracle Warehouse Management Cloud
AI/ML User Guide

Contents
AI/ML User Guide ................................................................................................................. 1
Contents ................................................................................................................................. 1
Get Help in the Applications ..................................................................................................................... 3
Get Support ................................................................................................................................................... 3
Get Training .................................................................................................................................................. 3
Join Our Community .................................................................................................................................. 3
Learn About Accessibility .......................................................................................................................... 3
Share Your Feedback ................................................................................................................................. 3
Oracle WMS Artificial Intelligence (AI) and Machine Learning (ML) ............................. 4
Introduction .................................................................................................................. 4
Artificial Intelligence ................................................................................................... 4
What is Machine Learning? ......................................................................................... 4
Machine Learning Process Flow .............................................................................................................. 5
Types of Algorithms available in WMS .................................................................................................. 5
Random Forest:............................................................................................................................................ 5
Random Forest Parameters ...................................................................................................................... 5
Feed Forward Neural Networks ............................................................................................................... 6
Feed Forward Neural Networks Parameters ......................................................................................... 6
Gradient Boosting ....................................................................................................................................... 6
Gradient Boosting Parameters ................................................................................................................. 6
Working with AIML in Warehouse Management.............................................................. 7
Create A Scheduled Job .............................................................................................. 7
Scheduled Jobs UI ....................................................................................................................................... 7
Order Cycle/Waiting/Processing Time .................................................................................................. 9
Generate Prediction Run for Order Cycle/Waiting/Processing Time ............................................ 9
View Training Data .................................................................................................... 10
Define Model with Parameters/Filters andSelect it in the AI/ML Training
Template................................................................................................................ 11
Predict/Execute Capturing Results in AIML Prediction Run UI ............................ 14
Prediction Run UI ....................................................................................................................................... 14
Order Cycle Time, Order Waiting Time & Order Processing Time ................................................ 15
AIML Models Inquiry ................................................................................................................................ 15
Interpreting Results ....................................................................................................16
AI/ML Predictive Fulfillment Dashboard .........................................................................16
1
Oracle Warehouse Management Cloud
AI/ML User Guide
Oracle Market Basket Analysis ..........................................................................................19
Index .................................................................................................................................... 22
Oracle Market Basket Analysis for Putaway ................................................................... 23

2
Oracle Warehouse Management Cloud
AI/ML User Guide

Get Help
There are a number of ways to learn more about your product and interact with Oracle and other users.

Get Help in the Applications

Access the online help from the user drop-down menu in the Warehouse Management application.

Get Support
You can get support at My Oracle Support. For accessible support, visit Oracle Accessibility Learning and Support.

Get Training
Increase your knowledge of Oracle Cloud by taking courses at Oracle University.

Join Our Community

Use Cloud Customer Connect to get information from industry experts at Oracle and in the partner community. You
can join forums to connect with other customers, post questions, and watch events.

Learn About Accessibility

For information about Oracle's commitment to accessibility, visit the Oracle Accessibility Program. (if videos) Videos
included in this guide are provided as a media alternative for text-based topics also available in this guide.

Share Your Feedback

We welcome your feedback about Oracle Warehouse Managment. If you need clarification, or find an error, you can
direct your questions via a service request to My Oracle Support.

3
Oracle Warehouse Management Cloud
AI/ML User Guide

Oracle WMS Artificial Intelligence (AI)

and Machine Learning (ML)
Introduction
Predicting Order Cycle Time can be a challenge in today’s ever-changing environment. Our
Warehouse Management (WMS) Machine Learning allows you to use historical order data
with optimized machine learning algorithms and techniques to help deliver accurate
predictions.

WMS now has AI/ML capabilities and models supporting the following uses cases:

• Predicting Order Cycle/Waiting/Processing Time

Artificial Intelligence
Artificial Intelligence (AI)-enabled programs can analyze and contextualize data to provide
information or automatically trigger actions without human interference. .Artificial intelligence
can help manufacturing leaders automate their business processes by applying data analytics
and machine learning to applications.

What is Machine Learning?

Machine learning (ML) provides users with the ability to take historical data to predict the
future. By feeding the data into Oracle’s proprietary algorithms, users can create customized
models that are good at predicting certain events. A well-trained machine learning model
provides the ability for users to predict future events, and automatically identify certain
features/groups. To describe a good or bad machine learning model we use the terms
“overfit” and “underfit. In these instances, we must ‘teach’ our model to predict our use cases
accurately.

• Underfit Model: An underfit model fails to learn the current problem from the dataset
and poorly performs well with the resulting predictions.
• Overfit Model. An overfit model learns the training dataset too well. This model will
perform will in testing, however in production will be too precise and thus provide
inaccurate/unreal predictions.
• Good Fit Model. An ideal model will provide prediction within a good range and
reasonable outcomes.

4
Oracle Warehouse Management Cloud
AI/ML User Guide

Machine Learning Process Flow

Train, evaluate
Source and Define the
and tune the
Prepare data model
model

Analyze the
Deploy the
predictions for
model
the model

Types of Algorithms available in WMS

We currently offer three different Machine Learning algorithms in WMS. They all require
different parameters and model generation methodology varies, however they can all be used
to predict Order Cycle Time.

Note: Results will vary based on the model and parameters given.

Random Forest:
Random Forest is useful in helping solve real world problems like Regression and
Classification (see index for explanation of those)

Random Forest Parameters

N_Estimators: is the number of decision trees that the Random Forest model will contain. In
general, if there are more trees, the better the model will learn the data; however, if there are
too many trees, it could extend the processing time considerably. To help minimize the
processing time, we have set the default value of trees to be 100. If a user would like to change
this value, they can input a number between 1 and 200.

5
Oracle Warehouse Management Cloud
AI/ML User Guide

Max_Depth: max_depth represents the depth of each tree in the forest. The deeper a tree is,
the more decision it will generate, and captures more information about the data. We fit each
decision tree with depths ranging from 1 to 20 and plot the training and test errors. If the user
enters a value outside of this range, an error message will appear.

There is no default value for this parameter.

Note: When scoring/deploying the mode, it is necessary to run several iterations. This is
training the model. This leads to optimizing you model as there is variation in results based on
parameters.

Feed Forward Neural Networks

Feed Forward Neural Networks work best when finding solutions for problems that a person
would be good at versus a computer. This can include identifying pictures, texts, and sounds.
They are common in clustering and classification use cases.

Feed Forward Neural Networks Parameters

hidden_layer_sizes: This parameter allows us to set the number of layers and the number of
nodes we wish to have in the Neural Network Classifier.

max_iter: It denotes the number of epochs.

Epoch: An epoch is the total number of iterations of the training data in one full cycle.

Gradient Boosting
Gradient Boosting is a powerful machine learning algorithm that can solve classification and
regression problems.

Gradient Boosting Parameters

max_iter: It denotes the number of epochs.

Max_Depth: max_depth represents the depth of each tree in the forest. The deeper a tree is,
the more decision it will generate, and captures more information about the data. We fit each
decision tree with depths ranging from 1 to 20 and plot the training and test errors. If the user
enters a value outside of this range, an error message will appear.

learning_rate: Learning Rate is the speed at which your model will learn. If this value is too
high, we may pass the optimal solution, conversely, if it is too low, it may take too long to
converge to our optimal solution.
6
Oracle Warehouse Management Cloud
AI/ML User Guide

Working with AIML in Warehouse

Management
This section walks you through the screens you can use in WMS to schedule data for AIML
models, define training models, interpret results, and make adjustments as needed.

The process to create one of these models is defined in the following sections:

• Step One: Create a Scheduled Job.

• Step Two: Viewing the Training Data
• Step Three: Define model with parameters/filters & select it in ai/ml training template
• Step Four: Predict/Execute Capturing Results in AIML Prediction Run UI
• Step Five: Deploy, Retrain Models and Interpret Results

We have introduced the screens:

• AIML Training Data (users view the prepared data from a scheduled job)
• AIML Training Template (users create templates and can set a default)
• AIML Models UI (users view their models built from templates)
• AIML Prediction Run UI (where users view results)
• AIML Predictive Dashboard (users can also view results captured from their
latest prediction here- only for order cycle time

Create A Scheduled Job

Each metric pre-determines the specific algorithms allowed for it. It will also determine the
type of input data it needs and the output it will produce.

We currently support functionality for the following metrics.

• Order Cycle Time

• Order Waiting Time
• Order Picking Time

Scheduled Jobs UI
You will be able to schedule the data creation for AIML Models through the Scheduled Jobs

7
Oracle Warehouse Management Cloud
AI/ML User Guide

UI.

These drop-downs are available in the CRUD pane of the Scheduled Jobs UI. Select ‘Generate
Training Data for ______” in order to create a scheduled job.

Selection of the above job type from the drop down will further expand the Job parameters:

Username - (mandatory field) a valid username/login in WMS should be provided).

From Date:

To Date:

Past Number of Days:

8
Oracle Warehouse Management Cloud
AI/ML User Guide

NOTE:
• Each of the jobs will not be enabled by default. If you attempt to save without any value
for username, you will receive an error message.

The output (model) of the job (different for each metric) will be captured in the AIML Models
UI. One for each job run.

• Each record thus generated is defined by a unique composite key (business

key).
• The job will add, or update records based on the key.

For every run of the scheduler, the job will be logged in the Scheduled Job Detail UI. Logs
should be provided for each job run. If the job fails, logs will be updated accordingly with the
Failed job status.

Order Cycle/Waiting/Processing Time

You can generate Order Cycle/Waiting/Processing time using a scheduled job. This will
predict the number of orders during a user specified time. Users can configure the scheduled
job in Scheduled Jobs UI and will create the data required to build the model.

Term: Description:
Order Cycle Time Order cycle time is the time between the
Order created and the Order shipped

Order Waiting Time Order waiting time is time between the

Order created and the Order allocated

Order Processing Time Order processing time is time between

the Order allocated and the Order
Shipped

Generate Prediction Run for Order Cycle/Waiting/Processing

Time
The above job type is available in the CRUD panes of the Scheduled Job UI.

• Selection of the above job type from the drop down will further expand the Job
parameters:

9
Oracle Warehouse Management Cloud
AI/ML User Guide

o Username
Training Run Number
Orders -
o Wave number - (optional)
o Order Type(s)

The result of the model is available in Prediction Run UI when a user clicks on the
corresponding details button in Prediction Run UI. By selecting the run number users will be
able to see the results of the specific run.

Output results will be stored separately per KPI/Metric and per prediction run and will be
different for each metric.

The output data will also record if it was the result of a "default" template for that KPI/Metric.

Users will then be able to view the Prediction Run Number, type (cycle/wait/picking), order
numbers and handling time.

Note: The detail drill down from the Prediction Run UI will show the above data in a data grid.

View Training Data

When the job runs, it will process the Username and AIML template given, and generates a
model on successful completion of the job. This job will indirectly invoke the “Train” action on
the AIML Training template for the template given in the job parameter. This provides the
ability for the user to generate the model periodically.

The output (model) of the job (different for each metric) will be captured in the AIML Models
UI. One for each job run.

You can use the AIML Training Data screen to view the output of the scheduled job for Order
Cycle/Waiting/Processing Time.

10
Oracle Warehouse Management Cloud
AI/ML User Guide

You have to select the Order Cycle/Waiting/Processing Time metric from the drop-down in
this screen. After selecting this metric, a data grid will display the desired data. This data
captures the output of the job “Generate Training data for Order
Cycle/Waiting/Processing Time.

You can see the following fields between the two metrics:

• Facility/Company
• Shipped Date
• Order Nbr
• Order Count
• Handling Time
• Create User/Mod User
• Create Timestamp/Mod Timestamp

Changing the drop-down will swap in a new data grid screen for that KPI/metric. Since the
data fields for each metric will be different.

• Each such sub-screen has search functionality.

• You will not be able to Create or Edit in this screen, however, you can delete records.
• You have the option for a CSV download.
• You can search/filter for a specified date range (From date/To Date).
Order Cycle/Waiting/Processing Time

The “Order Cycle/Waiting/Processing Time” is sub-screens in the AIML Prediction Run

Details UI. This is where the results will be stored.

The UI will display the Prediction Run Number, Date and Order Cycle/Waiting/Processing
Time.

Define Model with Parameters/Filters andSelect

it in the AI/ML Training Template
The 'AIML Training Template' allows you to configure the specific parameters and other
details required to generate an AIML model for each metric. The AIML model allows you to
manage and train your data for future business predictions.

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

• Template Name
• AIML Metric

11
Oracle Warehouse Management Cloud
AI/ML User Guide

• Date Range
• AIML Algorithm

Note:
• You cannot have a ‘/ ‘ in the template name and you cannot edit the name after saving
it
• You will now see an error message upon saving an incorrect entry in the date range
fields in all of your AIML Training Templates. This allows you to see the correct format
based on which type of date input-type you prefer.

In ths AIML Training Template, you will have the ability to train and generate AI/ML
Models for future predictions. This screen is like the Wave Template UI and will allow
you 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 Cycle Time (regression problem)

• Order Waiting Time (regression problem)
• Order Picking Time (regression problem)

The AIML Training Template UI has the following action buttons:

Train Template

Enabled only for metrics that generate a model (order cycle time)

12
Oracle Warehouse Management Cloud
AI/ML User Guide

Users can save the template. Once it is saved, a user can invoke the action button ‘Train
Template’. Users then will be asked to confirm whether to run the template or not. After
confirming, a model is generated which is then captured in AI/ML Models UI.

The action button ‘Predict’ can only be invoked for certain metrics that do not require an
actual model to be generated. KPI’s that require ‘Predict’ will directly generate the output (not
a model).

You can determine the date range for the data that they will like to generate their AI/ML KPIs
on; however, if a user attempts to build one without sufficient data, the error use ‘Insufficient
Training Data to Generate Model’ will be returned.

You can use CRUD operations on this screen. On editing an already created template, (except
for the Template name) all other fields will be editable. On configuring the required fields with
valid values, you should be able to save the template.

13
Oracle Warehouse Management Cloud
AI/ML User Guide

Predict/Execute Capturing Results in AIML

Prediction Run UI
Prediction Run UI
In the Prediction Run UI, each prediction run is defined by the training run that has been used
and other parameters that will vary by the metric. For Order Cycle Time, it is the order filter
criteria.

Each time a metric invokes the ‘Predict’ button, the results will be stored here. Additionally,
each scheduled job run for the job type 'Generate Prediction Run' will result in an entry here.

The following fields are in the Prediction Run UI:

Field Name: Description:

Prediction Run Nbr System generated number on invoking

Predict action or on running a 'Generate
Prediction Run' job.
Training Run Nbr Reference to the training Run from the AIML
Training Models
Prediction Parameters Reference to the parameters used to execute
the model against new data to generate
predictions

Default flag Copied from training run; there can be

multiple runs for same metric with this
enabled. the latest one is the relevant one.
created, training, completed, failed,
Status
cancelled

There is a drill down detail button available when you select any given record. This shows the
output data for that run. It is specific to the metric of the run that the user selects. The output
will be one of the following KPIs:

▪ Order Cycle time

▪ Order Waiting Time
▪ Order processing Time

You can also use search/filters for Prediction Run Nbr, Traiing Run Nbr, Status and the Default
Flag. Each drill down screen also supports CSV download and appropriate filters as needed.

14
Oracle Warehouse Management Cloud
AI/ML User Guide

Order Cycle Time, Order Waiting Time & Order Processing Time
When you select a record in the AI/ML Models screen for Order Cycle Time, Order Waiting
Time and Order Processing Time, you must click the ‘Predict’ button to carry out the
prediction. The popup will prompt for the filters to be applied on all open orders.

o Orders - List of orders

o Wave number
o Order Type(s)
o Reqd ship date ON Day

Note: If no field is populated, then a user will be shown an error.

• On clicking 'OK', the execution i.e. the prediction for the respective metrics will
commence for the set of orders selected. (On 'Cancel', the action is cancelled.)
• On completing the execution, a record is generated in the Prediction Run(Inquiry) UI
which records the parameters and details used to run that prediction.
• Also, the execution will generate the output i.e. the Order Cycle Time or Order Waiting
Time or Order Processing Time depending on the metric of the record selected for
prediction, viewed in a separate UI screen 'AIML Order Cycle Wait Processing Time

AIML Models Inquiry

Every training template execution, whether manual or scheduled will result in an entry here.
For metrics that produce a model, the model generated will be stored within this screen.

The following fields are supported within this UI:

Field Name Description

Facility User's context facility
Training Run Number System generated number on running a
AIML training template
AIML Training Template name of the training template ran for the
specific run number
Training Start Date Start date of the training data
Training End Date End date of the training data
AIML Algorithm Algorithm for which the model was
generated
Training Filters Copied from template
Algorithm Parameters Copied from template
Default Flag Same as from template
Status Created, training, completed, tailed,
cancelled

15
Oracle Warehouse Management Cloud
AI/ML User Guide

Field Name Description

Message Text • Shows the latest message from
the training run
• Will be a reference number or link
• updated when training completes.
some metrics wont have a model
(like basket analysis)
• This should be copied back to the
template "Last Trained Model"

AIML Model Reference Shows the performance of the model

Performance Metrics Mean absolute error, r^2
Create Timestamp Date and time of the model/record.
Creation
Create User User who generated the model/record
(one who ran the AI/ML training
template)
Metric Metric being training, comes from
template

Action Buttons in AIML Models Inquiry:

▪ Logs
▪ Predict
▪ Prediction Run

Each of the drill down screens supports CSV download and appropriate filters as needed.

Interpreting Results
After successfully creating and viewing your models in the AIML Models UI, it is important to
note that your results may not yet be optimal. It could potentially take several iterations of the
template to generate better results. Oftentimes it may be best to change the type of algorithm
along with the types of parameters. Your results could also vary based on the type of industry
you are in.

AI/ML Predictive Fulfillment Dashboard

The a AI/ML Predictive dashboard This allows you to make well informed decisions based on
past order data to view future Order Cycle Time predictions. This information comes from the
most recent prediction from your model in AIML Prediction Run.
16
Oracle Warehouse Management Cloud
AI/ML User Guide

The predictive dashboard displays the average order cycle time in hours. We have included
various KPIs that gives insight on how your orders have and will move out of the warehouse.
By clicking on the order status, it will further filter and display the order type by status.

Your orders above and below your set target threshold will be shown in a datagrid that
displays the following information:
• Order Number
• Status, Order Type
• Predicted Time
• Expected Shipping Date
• Required Shipping Date

The ratio of orders above and below your target threshold will also be shown in a pie chart
above the data grid.

17
Oracle Warehouse Management Cloud
AI/ML User Guide

Steps to enable:

1. Navigate to the UI and click on the username dropdown.

2. Select Redwood UI Prototype

3. All the data should be there

You can access the AI/ML Predictive Fulfillment Dashboard by clicking the on the username

18
Oracle Warehouse Management Cloud
AI/ML User Guide

Oracle Market Basket Analysis

We have now added the ability in Warehouse Management to analyze your inventory using
Market Basket Analysis. By running this analysis, you will be able to generate a report that
displays frequent items in your warehouse that are commonly ordered together. This
information allows you to see what items may be beneficial to slot together during putaway.

Understanding your customer’s buying patterns can be extremely beneficial to your

warehouse operations. Having the knowledge of what items are typically ordered together,
you can potentially structure your inventory and 0slotting positions more efficiently.

Leveraging Oracle’s AI Machine Learning algorithms, the WMS now has the ability to perform
a Market Basket Analysis on your inventory. This analysis looks at purchasing patterns, and
groups items together based on how likely they are to be ordered together. The analysis will
display a report showing you these “frequent item-sets”, how often they are ordered together
and how probable it is to continue.

In the Scheduled Jobs UI, a new Job Type is now available which allows you to generate
training data for Market Basket Analysis. Follow these steps to generate and view analysis
specific to your data.

1. Create a scheduled job to generate the training data for Market Basket Analysis in the
Scheduled Jobs UI.

Note: You must enter a valid username, or else will be returned an error.

2. Navigate to the AI/ML Training Data to verify that the training data contains the
correct data.
3. Create a Market Basket Analysis Training Template in the AI/ML Training Template
UI
19
Oracle Warehouse Management Cloud
AI/ML User Guide

NOTE: To run the Market Basket Analysis, you must select the “Apirori” algorithm.

4. After creating your Market Basket template with all of the mandatory fields completed,
click Train Template.

5. Navigate to the AI/ML Model UI and select the most recent Market Basket Analysis
entry. In the “Model Reference” field, click on the hyperlink.

6. You can now view the results of your successful Market Basket Analysis. In 23B we have
added the option to filter by Max/Min values using the search option in the Model
Reference UI.

20
Oracle Warehouse Management Cloud
AI/ML User Guide

Note: For 23A release timeframe, when viewing your analysis via the Model Reference
hyperlink, it is recommended to download the results to save as a CSV.

In 23B we have introduced the ability to create season-focused templates (for example, Fall and
Christmas, season putaway rules using targeted date ranges. In order to create these, you
must navigate to the AIML Training Date Rule Hdr.

You can input multiple date ranges in the training date rule and they will all be considered.

Metric Defintions for Market Basket Analysis:

Support – is the percentage (expressed as a decimlal) of transactions that contain all

of the items in an itemset (e.g., printer, paper and staples). If the support is a high
percentage, then the itemset appears frequently. Conversely, if the support is low, the
itemset is considered infrequent.

Confidence – represents the likelihood that two items will appear in a transaction
together. For example, if I have a confidence of .7, then I am 70% certain that if
someone orders a printer, they will also order a pack of paper.

21
Oracle Warehouse Management Cloud
AI/ML User Guide

Lift – represents the likelihood of all items in a rule will be ordered together. This
calculated by dividing the support by the product of the probabilities of the each item
occurring as if they are not associated at all. When lift > 1, it suggests that the presence
of Item A, will increase the likelihood of seeing Item B in an order. A large lift suggests
a strong relationship between two or more products.

Index

Classification: Classification is the process of determining what class an object or data point
belongs in based on data points. A machine learning model will group similar data points
together in order to categorize future predictions.

Regression: Regression in statistics is the process of identifying the relationship between a

dependent and one (or more) independent variables.

Tuple: A tuple is a set of ordered objects.

22
Oracle Warehouse Management Cloud
AI/ML User Guide

Oracle Market Basket Analysis for

Putaway
We have added the ability for a user to use Market Basket Analysis in putaway via a new
Putaway Method “Market Basket Analysis Radial”. This new Putaway Method applies to both
active and reserve locations and honors the existing “Radial” Putaway Method. This leverages
the “Primary Item” as the starting point which is then radiated out from to find locations within
the same Area/Bay. The “Primary Associated Item” is the item that a user should be needing
to conduct putaway on.

Example use case:

I am an office supply store warehouse supervisor conducting putaway. The store runs a
promotion for the past several months for every printer sold, the customer also gets 10 packs
of paper for free. After running our Market Basket Analysis for the past several months, we
notice that paper is strongly associated with printers, but printers aren’t as strongly associated
with paper.

This could be due to the promotion, because each time a printer is ordered, paper is as well;
however, whenever someone orders paper, this does not necessarily mean they are buying a
printer as well.

So, now that I am conducting putaway, I can see that I would like to put paper next to my
printers, but this does not necessarily mean I need to slot my printers next to my paper (paper
could also exist in multiple locations). The MBA Putaway Method will allow you to find
locations within the same Area/Bay that you can slot your Primary Associated Item in.

Example MBA Output:

23
Oracle Warehouse Management Cloud
AI/ML User Guide

The following steps describe the workflow:

Create Training Create Scheduled

Template Job

Create New Putaway Perform Putaway

Method based on Based on Associated
MBA Radial Items

1. In the Oracle Market Basket Analysis section, we showed you the steps to create
a Market Basket Analysis Report. The chart above describes the workflow for
creating a run and then incorporating into your putaway.
2. Once you have identified which run and associated template you would like to
use, navigate to the AIML Training Template UI, and check the checkbox called
“Use for Market Basket Analysis based Putaway”. This will determine what
template is the sole template for running your Market Basket Analysis based
Putaway on.

Note: You can only use one template for putaway at a time. If you have a template marked
and then mark another, the first one will be automatically unselected.

24
Oracle Warehouse Management Cloud
AI/ML User Guide

The Putaway Method Priorities screen allows you to create new putaway methods and
configure priority, location types, and what increment the system should try to find a location
for your items.

3. Next, navigate to your Putaway Method Priorities screen and create a new Putaway Method
based on your warehouse needs. For the Putaway Method field, select “Market Basket Radial”.
Navigate to the Putaway Method Priorities screen and create a new putaway method. For the
Putaway Method field, select “Market Basket Radial”.

After creating your new putaway method, you will now have the ability to assign this to a
Putaway Priority. The above screen shows that:

When an IBLPN of Putaway Type “RRMBA” is scanned for putaway, the WMS will look for all
active locations to putaway to. Out of all of these locations, the WMS will identify where your
“Primary Item” is and then identify empty locations to put your Primary Associated Item in.
25
Oracle Warehouse Management Cloud
AI/ML User Guide

You can also review more information about the Radial Putaway Method here:

https://docs.oracle.com/en/cloud/saas/warehouse-management/21a/

26