Filip Kaznowski,

Business Development Director, Oracle Consulting CEE

Amadeusz Dańko

BI Consultant, Accelerated Autonomous Practice, Oracle Consulting EMEA

“Autonomous Database & Analytics”Nowe podejście, czyli czy można w 5 dni skutecznie pomóc działom biznesowym?

Oracle Consulting

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.

Statements in this presentation relating to Oracle’s future plans, expectations, beliefs, intentions and prospects are “forward-looking statements” and are subject to material risks and uncertainties. A detailed discussion of these factors and other risks that affect our business is contained in Oracle’s Securities and Exchange Commission (SEC) filings, including our most recent reports on Form 10-K and Form 10-Q under the heading “Risk Factors.” These filings are available on the SEC’s website or on Oracle’s website at http://www.oracle.com/investor. All information in this presentation is current as of September 2019 and Oracle undertakes no duty to update any statement in light of new information or future events.

Safe Harbor

Expotential growth

33

Courtesy of Ray Kurzweil and Kurzweil Technologies, Inc.

Calculations/SecPer 1000 USD

Unicorns (>1 bln USD market cap companies)

What can you expect from Oracle Consulting?

Expert knowledgeGuidance that’ll help you Accelerate with Autonomous

Customer centricWe listen to your needs and help get the most out of your Data

Innovative & Rapid approachWe offer real business value in only 5 days!

Consulting IP Library

7

Enterprise Analytics -Finance

Talent Review

Telco Churn PredictionBank Loan Analysis

Sales Pipeline

Tax and Welfare

Enterprise Analytics -HCM

Enterprise Analytics -SCM

Resource Mining Extraction Control

Energy Controlling

POS Analytics

IoT Analytics on Elevators

Budget Analysis

Sales Compensation

Asset Liability Management reports

Banking Channel Analysis

Spend Analysis

Patient Visit Analysis

Electricity Load Profile

HR Attrition

Sales Prediction

Field Service

Airline Sales Analysis

Service Cloud Incidents

Urban Transport Organization Datamart

IoT Analytics on Elevators

DV Plugins

DV Sample Project Extended

Results & Next Steps

Present findings & results of the Use Case and define a Roadmap for Expansion:-

- Cloud Services / OAC

- Data Integration / ETL

- Users

- Data

- Functionality

What does an ADW Rapid Start look like?

Benefits of adopting ADWCMake your Analytics AutonomousStart your cloud journey

Use Case Definition

Solution Design

Provisioning

Guide you on deciding a quick

win that maximizes

value of ADWCS & agree

deliverables for a Use Case.

Evaluate how to implement the

solution including any

applicable connections

and agree data source(s) for

Use Case.

Set up ADW instance & show how to connect

data load option,

performance dashboard and

Data Visualization

Desktop

Load Data Build Reports

Migrate your data into the ADW environment.

Design & build reports, dashboards or insights to deliver real business value

with a permanent solution

Day 1 Days 2-4 Day 5

Sample ATP themes / use cases

Build new or

move existing APEX

application

Create an FAQ Chatbot

Create a Mobile App

Design & set up a New ATP to capture IoT

data

L&S an existing database to ATP

Every business has FAQ tool or application, either for Employees

or Customers. Modernize by loading to ATP and improve

responsiveness with Machine Learning

A modern looking and responding mobile app for

an existing or new standalone workflowie Invitation tracking for internal event,

customer/employee survey, …

Several dashboardsto allow for smart

real-time monitoring Sensor IoT Prognostics, …

To show the performance / security / maintenance gains

ADW Rapid Start project delivery reference

Oracle ML in Action – Bus TerminalEnergy consumption prediction for a bus station

Oracle Consulting Advanced Analytics

Problem

• Energy consumption as key cost

• Undefined consumption drivers

• Inability to spot key consumption drivers

• Buying more than needed - inability to predict usage and choose best energy supply

• Inability to be „greener” thorough better optimized power consumption

Bus terminal – Project Overview

Action

• Determine the power consumption drivers

• Spot core consumption drivers

• Predict future consumption with ML techniques

Bus terminal – Objective

15

Bus terminal – Project Overview

In June 2019, Oracle started a 10-days Advanced Analytics Project to identify potential improvements in

prediction, integrating additional internal and external data.

Scope of the ProjectBus Station

2016 2017 2018 20202019

Jan-Dec 2018

Data extraction Time Horizon Project Team

Cloud & Consulting Service

Bus Station

External and Internal Data Sources used for the Project

Energy Consumption

PREDICTION

Weather Data

INTERNAL

• External Temperature (hourly)

• External Humidity (hourly)

Other Business info

impacting Energy Consum.

• Bus arriving in the station

• Bus departing from the station

Weather Data

EXTERNAL

• Air temperature

• Air pressure

• Clouds

• ...

Prediction Model

Energy Consumption

ACTUAL

Paid data

• Electrical Energy Consumption. (hourly)

Energy consumption profile related to each bus during stop at the station

Energy related to

Passengers ARRIVING

Energy related to

Passengers DEPARTING

Arrival Departure

Time

Electric

Energy

t0 t1 t2 t3 t4 t4

Bus Terminal

15 min

~1 h

Advanc. (min) % passeng.

60 10%

45 20%

30 50%

20 90%

10 100%

5 100%

0 100%

For DEPARTING passengers, the following distribution

has been defined for advance minutes vs departing time

Energy related to

Charging Buses

Charging Time

Selected algorithms

Supervised Learning Functions

Unsupervised Learning Functions

2) Selected algorithms for REGRESSION:

1. Generalized Linear Model (GLM)

2. Support Vector Machines (SVM)

3. Neural Network (NN)

1) Selected algorithm for ATTRIBUTE IMPORTNACE:

1. Minimum Description Length (MDL)

Validation approach

Our Oracle Analytics platform provides specific features to support Data preparation and Model Validation.

Validation and comparison of the algorithms has been performed on a random subset of data calculating three standard indicators.

Actual Consumption historical Data

30%

70%

Training

Test

1) Root Mean Squared Error

Actual Values Prediction

2) Mean Absolute Error

Actual Values Prediction

3) Mean Absolute Percentage Error

Actual Values

Prediction

ML process flow

Data preparation(combining)

Data split(training/test)

Creating & populating ML settings table

Building machinelearning model

Checking model settings and

attributes

Testing on test data, getting RMSE, MAE,

MAPE.

Aggregating the results for analysis

21

Machine Learning

Preliminary analysis using Analytics Cloud

Energy Consumption (by Day)

Peaks in summer period(probably for holidays, air conditioning)

Correlation

Correlation close to 1• Electric Power and Arriving Passengers positively correlated• Electric Power and External Temperature positively correlated

Electric Power and Arriving Passengers tend to increase/decrease together

Attribute Importance and Model Validation

Attributes filtering

Algorithm execution

Attribute Importance Output Model Validation

NN Prediction error % lower than the other

most of the time

Regression model results for Electrical Energy - Prediction vs Actual

Month: January 2018

Hourly basis

Daily basis

Customer Success

• The customer uses our model regularly, feeding it with new data

• Very precise estimations of energy demand

• Key consumption drivers determined

• Adequate energy amounts purchased

• All ML processes within one database - reducing cost, increasing stability and accelerating results

The value of Oracle Consulting Rapid Start

Rapid insight into your data providing business value,driven by your requirements

Flexible scope to adapt to your needs

Repeatable & proven approach across International

A vast library of Consulting IP Assets and years of expertisewith Oracle Products

100%

Dziękujemy za uwagę

Accelerate Autonomous Practice

Oracle Consulting