Forecasting, Allocation, Replenishment, Merchandise Financial Planning and Pricing

View the deck about Forecasting, Allocation, Replenishment, Merchandise Financial Planning, and Pricing presented to The Container Store team.

/live_board/v2/downloads?uid=b9d367cd-6a57-4b10-a9b8-9424ddadabf7&redirect=true

Platform for Modern Retail

July 7, 2022

Safe harbor statement

The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions.

The development, release, timing, and pricing of any features or functionality described for Oracle’s products may change and remains at the sole discretion of Oracle Corporation.

Build a blueprint for agility

Get closer to your Customers

Create space for innovation

Historical view of the Retail landscape (pre-Platform)

Data

Architecture

Reporting

Planning

Process

Data

Architecture

Reporting

Forecasting

Process

Data

Architecture

Reporting

Merchandising

Process

Data

Architecture

Reporting

Supply Chain

Process

Data

Architecture

Reporting

Customer

Process

Data

Architecture

Reporting

Finance

Process

1. Processes and Data are locked in Application Silos together.

2. Data created is captive to the Process, and elements needed in multiple apps were duplicated (ex. Daily $ales)

3. Interdependency of Processes are totally dependent on Data conversion necessary between the App silos.

Daily $ales Daily $ales Daily $ales Daily $ales Daily $ales Daily $ales

How does Oracle define the Platform for Modern Retail?

Data

Architecture

Reporting

Planning

Process

Data

Architecture

Reporting

Forecasting

Process

Data

Architecture

Reporting

Merchandising

Process

Data

Architecture

Reporting

Supply Chain

Process

Data

Architecture

Reporting

Customer

Process

Data

Architecture

Reporting

Finance

Process

Step 1: Remove silos and liberate Processes from the Data

How does Oracle define the Platform for Modern Retail?

Data

Architecture

Reporting

Planning

Process

Data

Architecture

Reporting

Forecasting

Process

Data

Architecture

Reporting

Merchandising

Process

Data

Architecture

Reporting

Supply Chain

Process

Data

Architecture

Reporting

Customer

Process

Data

Architecture

Reporting

Finance

Process

Step 2: Unify the Retail Data Model

Step 3: Embed AI & ML across the Architecture

Step 4: Comprehensive Suite of Cloud Services

Retail Cloud Services

Merchandising

Inventory

Operations

Brand

Compliance

Planning

Store

Operations

Order

Broker

Analytics

Order

Mgmt.

Supply

Chain

Customer

Engagement

Retail Data Store

Retail AI Foundation

Forecasting

Engine

Affinity

Analysis

Profile

Science

Customer

Decision Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction

Advanced

Clustering

Customer

Segmentation

Data

Process

The Platform for Modern Retail

Retail Home BUSINESS KPI’S ADMINISTRATIVE MONITORING NOTIFICATIONSSERVICE PORTAL

Oracle Cloud Infrastructure / Next-Gen Architecture STORAGENETWORKING COMPUTE CONTAINER ORCHESTRATIONCONTAINERS

Retail AI Foundation

Forecasting

Engine

Affinity

Analysis

Profile

Science

Customer

Decision Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction

Advanced

Clustering

Customer

Segmentation

Retail Cloud Services

Merchandising

Inventory

Operations Brand

Compliance Planning

Store

Operations

Order

Broker

Analytics

Order

Management

Supply Chain Customer

Engagement Retail

Extensibility

Low-code App

Development

REST Data

Services

Cloud

Extensions

Retail Data Store AUTONOMOUS DATA WAREHOUSE DATA MANAGEMENT RETAIL SCHEMA EXTENSIONSPACKAGED RETAIL SCHEMAS

Retail Reference

Model

Business

Process Flows

Glossary

Technical

Models

The Platform for Modern Retail

Retail Home BUSINESS KPI’S ADMINISTRATIVE MONITORING NOTIFICATIONSSERVICE PORTAL

Oracle Retail Home -

a single access point or portal

to simplify users’ interactions

with the applications and data

that are most relevant to their

roles and personas, to better

empower them to anticipate

informed actions, and to

inspire engagement.

Retail Home is intended first to

provide timely and role-specific

high-level insights, and to

enable selectively drilling into

relevant applications for more

details.

The Platform for Modern Retail

Retail Cloud Services

Merchandising

Inventory

Operations Brand

Compliance Planning

Store

Operations

Order

Broker

Analytics

Order

Management

Supply Chain Customer

Engagement

42+ Oracle Retail Cloud Services (a few highlighted below): • Merchandising Foundation

• Supplier Evaluation (ESG)

• Brand Compliance

• Pricing

• Invoice Matching

• Allocation

• Enterprise Inventory

• Merchandising Insights

• Demand Forecasting

• Merchandise Financial Planning

• Assortment & Item Planning

• Assortment & Space Optimization

• Allocation & Replenishment

• Inventory Optimization

• Offer Optimization

• Promotion & Markdown Optimization

• Xstore Point of Service Cloud

• Store Inventory Operations

• Order Brokering and Order Management

• Customer Engagement (CE) Foundation & Segmentation

• CE - Campaign & Deal Mgmt. (coupons, promos & offers)

• CE - Loyalty & Awards (entitlements)

• CE - Gift Card Mgmt.

• Customer Insights

The Platform for Modern Retail

Retail AI Foundation

Forecasting

Engine

Affinity

Analysis

Profile

Science

Customer

Decision Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction

Advanced

Clustering

Customer

Segmentation

Like the RDS, the AI Foundation’s subscription model is based on consumption of

storage (per Terabyte) and compute (per Core), providing for flexibility and scale.

Drawing upon the data consolidation and extensibility of the Oracle Retail Data Store, the Oracle

Retail AI Foundation is a central analytical retail data lake-house in the Oracle Cloud, sourced

primarily from Oracle Retail’s Cloud Services, in which statistical algorithms inform its descriptive,

predictive and prescriptive analytics.

Highly-configurable, the AI Foundation can be tuned to reflect a retailer’s unique business model

and objectives, and in the spirit of machine learning is constantly improving, regardless of the

level of user assistance.

The Platform for Modern Retail

Retail Data Store AUTONOMOUS DATA WAREHOUSE DATA MANAGEMENT RETAIL SCHEMA EXTENSIONSPACKAGED RETAIL SCHEMAS

Packaged Oracle Retail Data Schemas

Merch Enterprise

Inventory

Private

Label

Compliance

Supplier

Evaluation Store

POS

Customer

Engagement

Order

Management

PO’s InventoryItemsSuppliers Sales CustomersOrders

Schema Extensions

Custom

Schema

Other

3rd Party

??? ???

The Platform for Modern Retail

Retail Reference

Model

Business

Process Flows

Glossary

Technical

Models

The Oracle Retail Reference Model (RRM) is a collection of detailed implementation information for our customers and partners, including glossary, business process and technical models based on Retail best practices.

Now available via Retail Home, the RRM helps retailers • Create A Baseline for Differentiation • Reduce TCO • Decrease Time to Value • Blueprint For Future Growth

A competitive differentiator: no other software provider has reference documentation with the breadth or depth of content as the Oracle Retail.

The Platform for Modern Retail

Retail Reference

Model

Business

Process Flows

Glossary

Technical

Models

Retailers can Leverage the Oracle Retail Reference Model:

✓ Early – to help users understand the Oracle Retail solution capabilities,

even before environments are provisioned.

✓ Often – to leverage the RRM to support conversations across all levels of the retailer's organization,

in every stage of the project, starting with discovery to detailed design.

✓ Consistently – to support business process re-engineering discussions and to create a common

approach across the business to drive user adoption.

RRM recent translations are available in French and Spanish: MOS patch ID 34069869.

RRM Brazilian Portuguese translation available soon.

Oracle Retail Documentation: https://docs.oracle.com/en/industries/retail/

To request an RRL demo or access free RRL training:

https://www.oracle.com/industries/retail/products/reference-library/

Email: retailprocess_ww@oracle.com to request an overview for your team.

https://docs.oracle.com/en/industries/retail/index.html https://www.oracle.com/industries/retail/products/reference-library/ mailto:retailprocess_ww@oracle.com

The Platform for Modern Retail

Retail Extensibility

Low-code App

Development

REST Data

Services

Cloud

Extensions

Using APEX, developers can quickly develop and deploy

compelling apps that solve real problems and provide

immediate value.

Oracle Application Express (APEX) is a low-code

development platform that enables you to build

scalable, secure enterprise apps, with world-

class features, that can be deployed anywhere.

REST (Representational State Transfer) is an architectural

standard for API’s (Application Programming Interfaces).

Oracle REST Data Services enables bridging data in

Oracle Retail Data Store and Oracle Retail AI

Foundation with systems outside of the Oracle Cloud.

The Platform for Modern Retail

Retail Extensibility

Low-code App

Development

REST Data

Services

Cloud

Extensions

An extension is a plug-in to the platform that extends the functionality of a

Cloud Service or the Platform itself.

It is designed to

improve the fit of a

cloud service to a

specific segment

or vertical, provide

competitive

advantage and

ultimately improve

the value a

customer is able

to extract from

their investments

in Oracle Retail.

The Platform for Modern Retail

Retail Home BUSINESS KPI’S ADMINISTRATIVE MONITORING NOTIFICATIONSSERVICE PORTAL

Oracle Cloud Infrastructure / Next-Gen Architecture STORAGENETWORKING COMPUTE CONTAINER ORCHESTRATIONCONTAINERS

Retail AI Foundation

Forecasting

Engine

Affinity

Analysis

Profile

Science

Customer

Decision Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction

Advanced

Clustering

Customer

Segmentation

Retail Cloud Services

Merchandising

Inventory

Operations Brand

Compliance Planning

Store

Operations

Order

Broker

Analytics

Order

Management

Supply Chain Customer

Engagement Retail

Extensibility

Low-code App

Development

REST Data

Services

Cloud

Extensions

Retail Data Store AUTONOMOUS DATA WAREHOUSE DATA MANAGEMENT RETAIL SCHEMA EXTENSIONSPACKAGED RETAIL SCHEMAS

Retail Reference

Model

Business

Process Flows

Glossary

Technical

Models

Oracle Retail Analytics & Planning

Affinity

Analysis

Profiling

Analytics

Customer

Decision

Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction &

Binning

Advanced

Clustering

Customer

Segmentation

Merchandise

Financial

Planning

Assortment

Planning

Assortment &

Space

Optimization

Promotion &

Markdown

Optimization

Offer

Optimization

Replenishment &

Inventory

Optimization

Retail Insights & Data Visualizations

Demand

Forecasting

Forecasting Engine

Optimize assortments by spatial constraints

Automate optimal regular pricing

Merchandise Operations

Merchandise Decisions

Merchandise Execution

Initial allocation & launch

Replenish & Optimize Inventory

Determine In-Season Promotions & Optimal

Offers

Maximize margin & sell- through markdown

strategies

Performance understanding &

insight

Define customer assortment &

sales plan

Define channel & merchandise financial plan

SERIES Platform for Modern Retail

Statistical Forecasting

Mei Planner

Alex Buyer

Beth Inventory

Analyst

Business Scenario Demonstration Forecasting, Replenishment and Allocation

Name

Scott Erpelding

Oracle Retail

July 7, 2022

Oracle Retail Analytics & Planning

Affinity

Analysis

Profiling

Analytics

Customer

Decision

Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction &

Binning

Advanced

Clustering

Customer

Segmentation

Merchandise

Financial

Planning

Assortment

Planning

Assortment &

Space

Optimization

Promotion &

Markdown

Optimization

Offer

Optimization

Replenishment &

Inventory

Optimization

Retail Insights & Data Visualizations

Demand

Forecasting

Forecasting Engine

Story: Ori can analyze and break large volumes of complex data into actionable insights, so you can better understand customer behavior and market trends. With Ori, retailers can create data models to derive insights and in-turn build prescriptive or predictive decision engines. This can help retailers with processes like demand forecasting allowing them to make better data-driven decisions

AI ASSISTANT

Ori

How is Ori measured:

• Model accuracy • Data governance • Forecast accuracy • Staff efficiency & Automation • Quick time to value • Partnership • Processing time • Constant Evolution of Innovation • Cost of staff errors • Supplier returns

Demand/Replenishment Analyst are responsible for determining the consumer demand in an omnichannel retail environment. The main duty of the Demand Analyst is to build and maintain a sales forecast, using data and insights to predict how much of a specific product or service customers will want to purchase during a defined time period. This method of predictive analytics helps retailers understand how much stock to have on hand at a given time

Demand / Replenishment Analyst

Alex

How is Alex measured: • Forecast Accuracy • Days of Stock • Out of Stock • GMROI • Inventory Carrying Cost • Spoilage • Staff efficiency • Cost of staff errors

Forecasting, Replenishment & Allocation Process Flow

Understand Consumer Demand

Gather Statics (Fact Data)

Determine Orders & Distribution

Execute Orders & Distribution

Manage Orders

Forecasting, Replenishment & Allocation Process Flow

Gather Statics (Fact Data)

Determine Orders & Distribution

Execute Orders & Distribution

Manage Orders

Understand Consumer Demand

Forecasting

Understanding Consumer Demand

Oracle’s Approach to Retail Demand Forecasting Empowering Commerce with the Forecast-driven Enterprise

• Drive sales, margin and customer satisfaction with a Single View of Demand

• Increase forecast accuracy for the entire product lifecycle

• Maximize the value of your data with best fit science that combines artificial intelligence, machine learning and decision science purposed for retail

• Maximize the agility of your business with extensible science, workflows and operations. (Gurobi, R, TensorFlow, Keras, scikit-learn, MXNet, NetworkX, Java, Python, PL/SQL)

• Maximize the productivity of your team with exception- driven processes paired with experience-inspired UI

Model Training Period Forecast Horizon

Trends & Seasonality

Slow Seller- Intermittent Demand Demand

Machine Learning & Time Series Forecasting

Forecast HorizonModel Training Period

Prepare Reference

Data

Estimate Demand

Parameters

Forecast & Evaluate

Model Performance

Select Best Fit Model

Refit over entire history

& Generate Forecast

Automated Analytical Engine

✓ Trends & Seasonality ✓ Out of Stocks ✓ Events ✓ Promotions ✓ Outliers

Trends & Seasonality ✓ Out of Stocks

Events Promotions

Automated New Item Modeling

• Attribute based similarity scoring

• Flexible weighting & eligible product controls

• Alerts guide user process to review/approve new item additions

• Automatic recommendations & approval by location

New item added

Attributes loaded

Auto or Manual

Approval

Auto-score ‘like’ products -Use attributes/weights -Recommend by location

Alert User Review/Adjust /Manual Approval

-Apply model choice -Build initial forecast

RDF Capability Evolution

Capability v13 v16 v19 v21

Workflow

Exception Based Workflow ● ●

Responsive Scorecards & Dashboards ● ●

What-if Analysis ● ● ●

New Item Automation ● ●

Jet UI Client (Look and Feel) ● ● ●

Forecast Calculation

Escalation & Pooling ● ●

Anomaly Detection & Correction ● ●

Model Transparency ● ●

Demand Transference ● ●

Extensibility ● ●

Flexible Source Level Groupings ● ● ●

Markdowns ● ●

Price Elasticity ● ● ●

Promotion Effects ● ● ● ●

Promotion ● ● ● ●

Price Discounts ● ● ● ●

● New Introduction

● Significant Enhancement

● Same from Previous Version

On Prem

RDF Algorithm Evolution

RDF Algorithms v13 v16 v19 v21

Regression Trees ● Decision Trees ● Random Forrest ● Neural Networks (Returns) ● Gradient Boosting (Returns) ● ML Cluster Generation for Forecast Model Training ● ● Ridge Regression ● ● Forecast Ensemble (Blended Forecast) ● ● TensorFlow Extensibility ● ● Lost Sales ● ● ● ● Simple Exponential Smoothing ● ● ● ● Holt Exponential Smoothing ● ● ● ● Additive Winters: Triple Exponential Smoothing ● ● ● ● Multiplicative Winters: Triple Exponential Smoothing ● ● ● ● Croston’s Intermittent Demand ● ● ● ● Simulation Based Parameter Optimization ● ● ● ● Average ● ● ● ●

● New Introduction

● Significant Enhancement

● Same from Previous Version

On Prem

Forecasting – Preparing Reference Data

Scenario Number Scenario Description

FRA 1 Removing anomalies from sales history

FRA 2 Auto selecting the best fit forecasting algorithm

FRA 3 Forecasting at multiple levels of the product and location hierarchies (to the store/SKU level granularity)

FRA 4 Approaches for forecasting shorter lifecycle SKUs

FRA 5 Adding new stores to the network

Demo

Forecasting – Review and Interaction

Scenario Number Scenario Description

FRA 6 Managing by exception with configurable metrics and dashboards

FRA 7 Understanding forecast components – base, seasonal and promotional sales

FRA 8 Forecasting SKUs sold both standalone and in pre-packs or kits

FRA 9 Planning for demand transference

FRA 10 Approving the forecast and interfacing to replenishment

FRA 11 “What if” forecasting and versioning

FRA 12 Running the forecast on demand and in scheduled batches

FRA 13 Measuring forecast accuracy with configurable metrics and user dashboards

Demand Transference

Demo

Forecasting, Replenishment & Allocation Process Flow

Gather Statics (Fact Data)

Determine Orders & Distribution

Execute Orders & Distribution

Manage Orders

Understand Consumer Demand

Replenishment & Allocation

Oracle Retail Lifecycle Inventory Planning High level business process

Classification

Location / Product Groups

Strategy & Rule Definitions

Set Lifecycle Strategy for Class to Sub-

Class

Set Up

Planning Calendars /

Lifecycle Phases

Detailed Allocation and Replenishment Planning

Automated Generation of the Time-Phased Inventory Plan, based on Class Strategy

AI & Machine Learning Inputs

Merchandise Operations Management / Warehouse Management

Customer Segmentation

Advanced Clustering

Attribute Extraction

Analyze Performance

In Season Adjustments

MFP OTB

Size Profile Optimization

CDT/DT

100

200

300

400

500

600

0 2 4 6 8 10 12 14 16 18

S K U S a le s

TAE

Assortment elas city

SKU sales

Small / Unique Assortment Large / Similar Assortment

Demand Forecast (Profile Curves,

Rate of Sale, etc)

AP Buy Plan

Rule-based Parameters Management Modeling the lifecycle

Season 1

2 3 4 5 6 7 8 9 10 11 12 1 2 3

EOLDIS MAT

EOLMATDIS

EOLMAT

Collection 1

Collection 2

MATNOS

Season 2

Year X Year X+1

Today

S e

a so

n s

/ C

o ll

e ct

io n

MAT Maturity

DIS

EOL

Discovery

End Of Life

• Lifecycle Parameters Management • The system behaves in a different way acccording to the lifecycle phase

• Allocation & Replenishment parameters are time variable

IA Initial Allocation

MATDISIA

Replenishment Methods + Dynamic Lifecycle Replenishment methods for any lifecycle phase

End of LifeInitial Alloc. MaturityDiscovery

This just is a lifecycle example. Lifecycles can have any number of phases and any meaning

Safety Time

Statistical SS

Min of active above

End of Life

% of Total Season Sales

Fixed (from assortment)

Safety Time

Min of active above

Fixed (from assortment)

Satefy Time

Max of active above

Allocation / Replenishment Item Lifecycle

End of LifeInitial Alloc. MaturityDiscovery

Initial Allocation

Discovery

Maturity

End of Life

PUSH – 70%

A ll

o ca

ti o

PUSH – 30%

Lifecycle

Initial Allocation

Maturity

PUSH – 30%

A ll

o ca

ti o

PUSH – 20%PULL – 50%

PUSH

• Pre-Launch Push • Post-Launch Push

MIXED

• Pre-Launch Push • Mid-Life Pull e.g. Sell 1, Send 1 • End of Life Push

Allocation & Replenishment Pushing

MCD MCD

MCD

DSP

PSH

PSH PSH

PSHPSH PSH PSH PSH

Store 1 Prio High

Store 2 Prio High

Store 3 Prio Medium

Store 4 Prio Medium

Store 5 Prio Medium

Need Allocation

R e

o rd

e r U

p -To

-Le v

e l

CO Customer Order

DSP Display

MCQ Minimum Credible

Qty

TS Target Stock

Allocation & Replenishment Rationing

MCQ

Store 1 Prio High

MCQ

DSP

Store 2 Prio High

Store 3 Prio Medium

Store 4 Prio Medium

Store 5 Prio Medium

TS TS

CO Customer Order

DSP Display

MCQ Minimum Credible

Qty

TS Target Stock

Ta rg

e t S

to ck

u b

la ye

Need Allocation

Reorder Up-To- Level

This allocation maximizes the service

level and thus probability to sell

Allocation & Replenishment Supply Chain Network & Planning Calendars

Supply Chain Network

• Maximum flexibility

• Links and lead times

• Time-phased

Replenishment Calendar

• Valid days for generating a replenishment proposal

Assortment Calendar

• SKU-Store validity

• Start-End sales calendar

• Interfaced, internally managed, hybrid

Lifecycle Calendar

• Start-End replenishment validity

• Phase transition rules

Distribution Center

Warehouse

Suppliers

Production

Stores

Distribution Center

Strategy-based Parameters Management What is a strategy?

• Strategies are assigned following a human rationale strategy-based approach

Dynamic SKU-Loc Filter

E.g. Class = Elfa Store Type = All Stores

All. & Repl. Parameters

E.g. Min Qty = 1 Max Qty = 2

Strategy 1

SKU -Loc

Dynamic SKU-Loc Filter

E.g. Class = Elfa Store Type = A Volume (High Volume Stores)

All. & Repl. Parameters

E.g. Min Qty = 2, Max Qty = 3

Strategy 2

SKU -Loc

Valid Dates

E.g. 7/10/22 – 8/10/22

Valid Dates

E.g. 7/10/22 – 8/10/22

Strategy-based Parameters Management What is a strategy?

• Strategies are assigned following a human rationale strategy-based approach

Strategy 1

SKU -Loc

Valid Dates

Priority

Dynamic SKU-Loc Filter

E.g. Class = Elfa Store Type = All Stores

All. & Repl. Parameters

E.g. Min Qty = 1 Max Qty = 2

SKU -Loc

Strategy 2

SKU -Loc

Valid Dates

Priority

Dynamic SKU-Loc Filter

E.g. Class = Elfa Store Type = A Volume (High Volume Stores)

All. & Repl. Parameters

E.g. Min Qty = 2, Max Qty = 3

SKU -Loc

Max Priority

E.g. 7/10/22 – 8/10/22 E.g. 7/10/22 – 8/10/22

Replenishment

Scenario Number Scenario Description

FRA 14 DC to DC inventory balancing to optimize DC to store replenishment

FRA 15 Elfa peak season purchase order management

FRA 16 Purchase order & shipment permutations

FRA 17 Seeding the replenishment process with a demand forecast

FRA 18 Tailoring replenishment strategies for store/SKU segmentations

FRA 19 Balancing ordering cost and inventory carrying cost (ICC)

Demo

Replenishment continued

Scenario Number Scenario Description

FRA 20 Gaining visibility to and control over inventory optimization levers

FRA 21 SOM's and VOM's. Managing vendor and store inventory ordering/ replenishment requirements.

FRA 22 Automating the basics and generating suggested PO’s (netting & time- phasing)

FRA 23 Scheduled and manual replenishment runs with exception alerts

FRA 24 Managing SKU inventory replenishment cradle-to-grave

FRA 25 Growing the supply chain network with minimal pain

Demo

Replenishment continued

Scenario Number Scenario Description

FRA 26 Measuring replenishment effectiveness

FRA 27 Automatic and Manual Purchase Orders

FRA 28 Campaign Monitoring

FRA 29 Elfa Shipped PO Quantity Changes

FRA 30 Elfa Shipped PO Cost Changes

FRA 31 Kit & Prepack SKU Types

Demo

Allocation

Scenario Number Scenario Description

FRA 32 Efficiently supporting allocation in a replenishment-centric environment

FRA 33 Seeding pre-season allocation with forecast and history

FRA 34 Grouping stores for differentiated allocation strategies

FRA 35 Allocating from an Advanced Shipping Notice (ASN)

FRA 36 Allocating from DC backstock using methods that consider on-hand and already allocated

FRA 37 Constrained Supply Allocations

FRA 38 Pushing out inventory before end of season

FRA 39 Scenario Planning & Measuring Allocation Effectiveness

Forecasting, Replenishment & Allocation Process Flow

Gather Statics (Fact Data)

Determine Orders & Distribution

Execute Orders & Distribution

Manage Orders

Understand Consumer Demand

Merchandise Financial Planning

Oracle Retail Merchandise Financial Planning Cloud Service

Highlights of the key capabilities

Core capabilities include:

• Full Pre and In Season Merchandise Financial Planning including Direct, Stores and Wholesale

• Built-in Location Planning to support department plans for each Store

• Plan by exception through Integrated Dashboards and Exception-driven planning

• Smarter planning through embedded forecasting and automation

• Optimize inventories by planning your customers buying, fulfillment and return journeys

• Simulate the full impact of different planning strategies with scenario based planning

• From planning to execution – A common user experience on all devices

Alex Buyer

Beth Inventory

Analyst

Jane Key Item Planner

Mei Planner

Merchandise Financial Plans are Created Assortment Plans

Created

Item Plans Created/Maintained

Allocation & Replenishment

Execution

Collaborative & Data-Driven Planning

Mei Planner

Merchandise Financial Targets are Created

Jill Planning Manager

Alex Buyer

Allocation & Replenishment

Execution

Assortment & Item Plans Created /

Maintained

Beth Inventory

Analyst

Seed Plans Plan Sales & Markdowns

Plan Inventory &

Receipts Review Margin

Submit Plan for Approval

Reconcile to Targets

Collaborative Merchandise Financial Planning

Price Optimization

Understanding Consumer Demand

Oracle Retail Analytics & Planning

Affinity

Analysis

Profiling

Analytics

Customer

Decision

Trees

Demand

Transference

Innovation

Workbench

Attribute

Extraction &

Binning

Advanced

Clustering

Customer

Segmentation

Merchandise

Financial

Planning

Assortment

Planning

Assortment &

Space

Optimization

Promotion &

Markdown

Optimization

Offer

Optimization

Replenishment &

Inventory

Optimization

Retail Insights & Data Visualizations

Demand

Forecasting

Forecasting Engine

The Roadmap: Delivering a Retail Platform to Accelerate Innovation

AGILITY

• Dynamic item role analysis & recommendations

• Streamline pricing recommendations across AI Foundation & Merchandising Cloud Services

CUSTOMER

• Regular Price Optimization

• Realtime customer behavior and transaction level data

• Fulfilment modeling (Customer journey based fulfilment forecast)

INNOVATION

• Customer Value measurement and course of action recommendation

• Oracle Autonomous Data Warehouse adoption for high performance analytics

AGILITY

• Dynamic pricing

CUSTOMER

• Connected planning, execution and consumer automation

• Promotion & offer optimization with personalization expansion

• Fulfilment and Order optimization and modeling

INNOVATION

• Citizen data scientist common library

• Strategy modelling

AGILITY

• Rules-based regular pricing

• Regular price optimization as a service

• Price execution

CUSTOMER

• In-Season Promotion & Markdown Optimization

• Customer Segment Offer Optimization

• Data Visualizations & Reporting

INNOVATION

• Machine Learning in forecast estimation

• Advanced promotional modeling

• AI models to support returns forecasting

• Open source framework to support building a retailer’s secret sauce

Current Capabilities Next Future

Supports merchandising &

marketing objectives,

reinforcing brand image &

maximizing margin

Maximize accuracy using machine

learning & optimization with a

modern AI Foundation

Holistic modeling to drive

incremental value to the lifecycle

While adhering to key business rules

Automated Analytical Price Optimization

Pricing

Scenario Number Scenario Description

Pricing 1 Strategy and Resource Roadmap- Today, TCS pricing is managed by 2 people and manual processes. There is limited visibility to the need for or impact of price changes

Pricing 2 Elasticity Modeling- We need to determine price elasticities (at the item, category level) to aid predictive modeling.

Business Constraints & Objectives

Demand Sensing

Statistically forecasted demand

Lifts and demand estimates Price elasticity

Pricing strategy Business rules Target goals

Lifecycle eligibility

Price Recommendations & Estimated Impact

Optimization

Demand Sensing With Time Series & Machine Learning Techniques

Demand Sensing

Trends Seasonality

Causal Factors Holidays Events

Promotional Events Campaigns

Marketing Vehicles

Decomposing effects of promotions and price changes

Campaign: Back to School

30% off offer

Coupon Offers

Clearance: 60% off

100

150

200

250

1 2 3 4 5 6 7 8 9 10 11 12 13

Base Campaign lift Offer lift Coupon effect Markdown effect

Interaction effects

P ro

m o

ti o

n E

ff e

ct s

P ro

m o

ti o

n C

a le

n d

a r

Price Elasticity Advanced Modeling for Price Optimization

% of Change in Quantity Demand

% of Change in Price

Style- Color

Sku Sku Sku Sku

Style

Style- Color

Sku Sku Sku Sku

Style

Subclass

Attribute 1 Attribute 2

Price elasticities are modeled at multiple levels of the hierarchy.

Pricing

Scenario Number Scenario Description

Pricing 4 What-if scenarios & simulations- We would like to be able to control select SKU prices and view suggestions for other SKUs in the same category/collection in relation to that SKU.

Pricing 5 Pricing Optimization- ability to recommend price changes.

Pricing 7 Customer Loyalty Based/ Offer Pricing

Pricing 8 What-if scenarios & simulations- optimize and estimate impact of promotions & offers.

RECOMMEND NEW ITEM REGULAR PRICE

Carried-over ItemNew Item All models are trained to read new items and place them into the right bucket for pricing

Carried-over Items are always at most up-to-date pricing recommendation when changes happened

Automates current manual price setting

Intelligent data processing -- efficient and fast

5% revenue improvement from optimized prices

*unconstrained

Expected Optimization Benefits

RECOMMEND PROMOTION, MARKDOWN & OFFERS

Offer Optimization

Promotion & Markdown Optimization In-season price recommendations to drive margin, revenue and sell- through

Determine customer segments with the highest probability of redemption

Sell-Thru Targets

Time Constraints

Promotion Rules

Markdown Rules

Product Groups & Associations

Min/Max Discount # of Markdowns

Min/Max Discount Marketing Vehicles

All Items Promoted/Markdown All Items Priced the Same

Business Constraints & Objectives

Promotion and Markdown Optimization

Demand Shaping- Promotion/ Markdown & Offer Optimization

Product Introduction $54

Promotion 10% Off

Promotion 15% Off

Markdown 15% Off

Markdown 25% Off

Markdown 50% Off

Targeted Offer Highlighting

Relevant 10% Off Promotion

Targeted Offer Special 25%

Offer for “Trend Setter”

Targeted Offer Optimization

Wk 1 Wk 3 Wk 5 Wk 7 Wk 9 Wk 11 Wk 13 Wk 15 Wk 19Wk 17 Wk 21 Wk 23

Pricing

Scenario Number Scenario Description

Pricing 6 Market Basket Analysis

FAST ACCURATE Scalable & automated to drive exception management

AI/ML Models of choices Transparent and rigorous error measurement

- Optimization with inflation adjustment and elasticity

Highly agile and compatible with various business focuses

FLEXIBLE

SOLUTION DIFFERENTIATORS, LEVERAGING AI

The Value of One

Transforming Inventory Precision

$ 1

Net Revenue $ Power of One Net Revenue $s

$1.020

$1.010

$1.000

$1 Billion retailer, AUR $36, 30% markdown revenue

-1 +1

$ B

ill io

n s

Customers typically see anywhere from

2 – 4 day improvement

➢ Robust Analytical & Machine Learning Techniques ➢ Optimization ➢ Automation ➢ Scalability ➢ Integration

+ $2,500,000 One time benefit for 1 day reduction in inventory *unconstrained

Supports merchandising &

marketing objectives,

reinforcing brand image &

maximizing margin

Maximize accuracy using machine

learning & optimization with a

modern AI Foundation

Holistic modeling to drive

incremental value to the lifecycle

While adhering to key business rules

Automated Analytical Price Optimization

Cape Union

Why Oracle for Forecasting, Financial Planning, Allocation, Replenishment & Pricing

✓ Agility- Utilize transaction level, customer data to drive insights at aggregate levels to drive buying, planning, moving and selling decisions. Create self-funding projects with the AI Foundation capabilities to infuse across the enterprise.

✓ Advanced Forecasting- Sense & shape demand utilizing advanced, automated forecasting engine to drive the end to end decisioning process.

✓ Automation- Utilize default strategies across solutions along with advanced analytics to drive an exception management process.

✓ What-if scenario modeling & simulation- Empower users with the ability to understand changes to parameters and estimated impacts.

✓ Localization- Localize pricing decisions across channels, price zones, customer segments and locations across the product lifecycle. Utilize demand signals to effectively place inventory to meet seasonal peaks and valleys.

Q&A

Forecasting, Allocation, Replenishment, Merchandise Financial Planning and Pricing

Item Type: pdf