Seeing SEMLA through the eyes of Software Engineers ... · Predictive Maintenance Load Forecasting...

PUBLIC

Tina Yang, SAP

Month 05, 2019

Seeing SEMLA throughthe eyes of Software Engineers & Architects

2PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ

SAP

98,659+Employees(3/31/2019)

180+Countries

18,800+Partners

€24.74BRevenue(FY2018)

25Industries

€3.6B+R&D spend

Source: SAP Global Corporate Affairs, April 24, 2019

https://www.sap.com/documents/2017/04/4666ecdd-b67c-0010-82c7-eda71af511fa.html

3PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ Source: SAP Global Corporate Affairs, April 24, 2019

437,000+Customers

77%of the world’s

transaction

revenue

92%of the Forbes

Global 2000

80%SMEs

78%of the

world’s food

82%of the world’s

medical devices

https://www.sap.com/documents/2017/04/4666ecdd-b67c-0010-82c7-eda71af511fa.html


30%Women in

management

by 2022

#1 Software Company

in the Dow Jones

Sustainability

Indices

298,000Employee hours

volunteered in

2017

1.4MYouth trained

in STEM

100%Renewable energy

in all data centers

and facilities

0Net emissions

by 2025


1960s – 1980s 1990s – 2000s 2000s – 2010s 2010s – 2020s


Intelligent Technologies will drives a next-generation value economy

Impact of Machine Learning

“Early evidence suggests that AI can deliver real value to serious adopters and can be a powerful force for disruption.

Early adopters are already creating competitive advantages, and the gap with the laggards looks set to grow”

- McKinsey Global Institute

Human Tasks

Will be Automated

by 2025

60%*

Accuracy in Voice &

Video Recognition

by 2020

99%Image Recognition

Accuracy Today(better than human

accuracy 95% )

97%$3.5 Trillion Annual

Value Created in the

Enterprise

$3.5T

Source: The Intelligent Enterprise Brochure

https://www.sap.com/documents/2018/06/1ea231a7-077d-0010-87a3-c30de2ffd8ff.html


Our IntelligentEnterprise works as one to automate, anticipate, and invent.


Manufacturing& Supply Chain

It connects every line of business through an Intelligent Suite of integrated applications


a Digital Platform to orchestrate data and integrate processes


and Intelligent Technologies to detect patterns, predict results, augment decisions,

And turn actions into outcomes.


Digital Platform

Intelligent Apps

SAP

Machine LearningBusiness Outcomes

Increase revenue

Re-imagine processes

Enhance productivity

Satisfy customers

Enabling innovations

SAP Machine Learning enables the Intelligent Enterprise

SAP Data

Conversational

Experience77% of the world’s

transaction revenue

touches an SAP system

25 industries

The world’s largest

business network


Intelligent Suite: Deliver intelligence across value chains

Out-of-the-box integration leveraging SAP

Cloud Platform, the SAP Analytics Cloud

solution, and a common data foundation with

SAP HANA and SAP Data Hub

Best-in-class UX with

consistent experience

across the entire portfolio

Easy to extend, allowing

customers and partners to

customize solutions quickly

Intelligence embedded

in the applications making

the workflows smarter

Modular, making

it easy to consume and

cost-effective to operate.


The business problems SAP customers are solving

Fraud + Risk Finance + HR

▪ Fraud and Abuse Detection

▪ Claim Analysis

▪ Collection and Delinquency

▪ Credit Scoring

▪ Operational Risk Modeling

▪ Crime Threat

▪ Revenue and Loss Analysis

▪ Cash Flow and Forecasting

▪ Budgeting Simulation

▪ Profitability + Margin Analysis

▪ Financial Risk Modeling

▪ Employee Retention Modeling

▪ Succession Planning

Operations

▪ Predictive Maintenance

▪ Load Forecasting

▪ Inventory/Demand Optimization

▪ Product Recommendation

▪ Manufacturing Process Opt.

▪ Quality Management

▪ Yield Management

25 Industries

Sales + Marketing

▪ Churn Reduction

▪ Customer Acquisition

▪ Lead Scoring

▪ Product Recommendation

▪ Campaign Optimization

▪ Customer Segmentation

▪ Next Best Offer/Action


Examples of business questions and related algorithms

Who will churn, commit

fraud, or buy next

week/next month?

How much will the monthly

revenue be or what is the

number of churners next year?

Classification

Forecasting

What are the groups of

customers with similar

behavior or profile?

How are the customers

and products related

to each other?

Segmentation

Link Analysis

How many products will

a customer buy next

month/next quarter?

What is the best offer or

recommended action for a

customer or internet user?

Regression

Recommendations

18© 2018 SAP SE or an SAP affiliate company. All rights reserved.

Build the Intelligent Enterprise

Machine Learning Roadmap Excerpt

Learning

RecommenderJob

Analyzer

Employee Self

Service Bot Total Workforce

Insights

Resume Matching

Manager &

Administrator Self

Service Bot

Career Planning

"People like me"

Knowledge

Bots

Payroll Fraud

Detection

SAP Fieldglass Live

Insights

Job Matching for

Candidates

Support and

Productivity BotsJob Seeker

Resume Ranking

Intelligent Customer

Experience Suite

Lead

Intelligence

Customer

Retention

Ticket

Intelligence

Product & Offer

Recommendation

Influencer Map &

Deal Finder

Multi-Touch

Customer Attribution

Contextual

Merchandizing

Self-Writing Expense

Computer Vision

Receipts

Anomaly

Detection

AI Expense

Approvals

Invoice

Digitization

AI Invoice

Processing

Itinerary

Capture

Chatbot

BookingsRisk Impact

Predictions

Automated

Duty of Care

Proactive Assistant

Semantic

Contract Repository

Item

Recommendation

Self-Service

Contracts

Attribute

Normalization

Semantic Search

Item

Normalization

Sourcing

OptimizationSourcing

Recommendation

Job Matching

Timesheet

Anomaly

DetectionProgram Office

Guidance

Job

Normalization

Statement of

Work Builder

Contract

ConsumptionSAP Tax Compliance

Smart Automation

Payment Block – Cash

Discount at Risk

Smart Alerts for Real Spend

and P&L AnalysisDemand-Driven

Replenishment Adjustment

Stock in

Transit

Sales Performance

PredictionCash

Application

Predictive Engineering

InsightsPredictive & Prescriptive

Maintenance

Demand

Sensing

Predictive Overall

Equipment

EffectivenessPredictive Quality

Management

Smart Worker

Enablement on Shop

Floor

Supply Chain

Segmentation

Advanced Forecast

AccuracyManufacturing &

Supply Chain

Chatbot

Bookings


Intelligent Technologies: SAP Leonardo

Applications that deliver intelligence

within core business process (such as

intelligent ERP, intelligent HR)

Innovation services combining design-thinking

and industry accelerators to help ensure

customers derive value from innovative new

technologies quickly and with reduced risk

A toolbox of intelligent technologies (IoT,

AI/ML, and analytics), microservices, and data

management tools that will be available over

SAP Cloud Platform to deliver intelligence out

of the box as well as through co-innovation

Universal analytics and SAP Digital Boardroom

solution connecting the enterprise for the CXO


SAP HANA – The only in-memory Machine Learning on live data

In-database predictive and machine learning capabilities

Strategic data platform for all SAP applications

Machine learning libraries allowing in-memory and co-located transactional and analytics processing

Native ML Function Libraries− Predictive Analysis Library (PAL) 90+ algorithms, covering classification,

regression, clustering, association analysis, time series forecasting, link analysis, recommender systems, outlier detection, statistical and pre-processing functions

− Automated Predictive Library (APL) and others.

ML extensibility− R integration and TensorFlow integration

− Streaming analytics embedded machine learning

− Application Function Library (AFL) SDK embedding custom C++ functions

− SAP application-specific function libraries for optimization and demand forecasting in SAP Supply Chain and SAP Retail applications

Algorithms

and Data

Push machine

learning close

to data

Algorithms

designed to run

in-memory

Parallel processing

for fastest

predictions,

forecasts, …

First-class ML environment for data stored in HANA, for on premise and cloud


SAP HANA Predictive Analysis Library (PAL) – Functional Overview

Algorithm overview by category

Classification Analysis

▪ CART, C4.5 and CHAID

Decision Tree Analysis

▪ K Nearest Neighbor

▪ Logistic Regression Elastic Net

▪ Back-Propagation (Neural Network)

▪ Naïve Bayes,

▪ Support Vector Machine

▪ Random Decision Trees

▪ Hybrid Gradient Boosting Tree (HGBT)4

Gradient Boosting Decision Tree (GBDT)*

▪ Linear Discriminant Analysis (LDA)*

▪ Confusion Matrix, Area Under Curve

▪ Conditional Random Field 4

Regression

▪ Multiple Linear Regression Elastic Net

▪ Polynomial, Exponential, Bi-Variate

Geometric, Bi-Variate Logarithmic

Regression

▪ Generalized Linear Model (GLM)*

▪ Cox Proportional Hazards Model*

▪ Random Decision Trees

▪ Hybrid Gradient Boosting Tree (HGBT) 4

Gradient Boosting Decision Tree (GBDT)*

Association Analysis

▪ Apriori, Apriori Lite

▪ FP-Growth

▪ KORD – Top K Rule Discovery

▪ Sequential Pattern Mining*

Probability Distribution▪ Distribution Fit/ Weibull analysis

▪ Cumulative Distribution Function

▪ Quantile Function

▪ Kaplan-Meier Survival Analysis

Outlier Detection▪ Inter-Quartile Range Test (Tukey’s Test)

▪ Variance Test

▪ Anomaly Detection

▪ Grubbs Outlier Test

Recommender Systems▪ Factorized Polynomial Regression Models**

▪ Alternating least squares****

▪ Field-aware Factorization Machines (FFM) ****

Link Prediction▪ Common Neighbors, Jaccard’s Coefficient,

Adamic/Adar, Katzβ

PageRank ****

* New HANA 2 SPS 00 | ** New HANA 2 SPS 01 |

Statistical Functions

▪ Mean, Median, Variance, Standard

Deviation, Kurtosis, Skewness

▪ Weighted Scores Table, ABC Analysis

▪ Covariance Matrix

▪ Pearson Correlations Matrix

▪ Chi-squared Tests: Quality of Fit,

Test of Independence

▪ F-test (variance equal test)

▪ Data Summary*

▪ Correlation Function*

▪ ANOVA**, One-sample Median Test**, T

Test**, Wilcox Signed Rank Test**

▪ Kernel Density Estimation 4,

▪ Entropy 4

Data Preparation

▪ Sampling, Binning, Scaling, Partitioning,

Discretize 4

▪ Substitute Missing Values,

Missing Value Handling 4

▪ Principal Component Analysis (PCA)/PCA

Projection

▪ TSNE 4

▪ Factor Analysis***

▪ Multi dimensional scaling***

Cluster Analysis

▪ DBSCAN, K-Means/Accelerated K-

Means**, K-Medoid Clustering,

K-Medians, GEO DBSCAN 4

▪ Kohonen Self Organized Maps

▪ Agglomerate Hierarchical

▪ Affinity Propagation

▪ Latent Dirichlet Allocation (LDA)

▪ Gaussian Mixture Model (GMM)

▪ Cluster Assignment

Time Series Analysis

▪ Single/Double/ Brown/Triple Exp.

Smoothing

▪ Forecast Smoothing

▪ Auto – ARIMA/Seasonal ARIMA

▪ Croston Method

▪ Forecast Accuracy Measure

▪ Linear Regression with Damped Trend

and Seasonal Adjust

▪ Test for White Noise, Trend, Seasonality

▪ Fast Fourier Transform (FFT)*

▪ Hierarchical Forecasting ****

▪ Change Point Detection 4

*** New HANA 2 SPS 02 | **** New HANA 2 SPS 03 4new HANA 2 SPS04


SAP HANA ML – Automated Predictive Library (APL) Native In-Database Automated Predictive Analytics

SAP HANA embeds the SAP Predictive Analytics

automated analytics engine (formerly KXEN)

Automated Predictive Library (APL)*

– Addresses key scenarios like automated Classification,

Regression or Time Series Forecasting (and more)

– Automation is based on concepts of “Structural Risk

Minimization” and covers analysis steps of automated

variable selection, data preparation, variable encoding,

missing value handling, outlier handling, binning and

banding, model testing and best model selection

Automation is the key to broad and fast adoption

– Quick and easy to leverage for non-expert Data Scientist and

to consume in applications built on HANA

– The APL provides simple procedure functions for developers

to Create, Train, Apply, Deploy and Query predictive models

SAP HANA Platform

Automated Predictive Library (APL)

Classification

Regression

Cluster

analysis

Time series

forecasting

Association

analysis

Recommendation

Link analysis

*SAP HANA 2 SPS04, available with full-use Predictive-option & HANA Enterprise license


SAP HANA ML – External Machine Learning Integration

Leverage open source machine learning with SAP HANA

External Machine Learning Integration covers

R Integration with SAP HANA

– Connect and interoperate with the SAP HANA database

from R Studio

– R script-code to be processed as part of the overall query

execution plan from SAP HANA

Python driver for SAP HANA

– Full support for the SAP HANA network protocol

– Leverage SAP HANA predictive & machine learning

capabilities from Python development environment

TensorFlow Integration with SAP HANA

– Easily extend deep learning from SAP HANA

– Retain the familiar database development environment Active

Model(s)

ODBC data

SAP HANA Platform

External Machine Learning Integration

R IntegrationTensorFlow

Integration

R-Serve Server

R-Processing

data + R-script result EML call prediction

result

Python Notebook


SAP Leonardo Machine Learning Foundation

enabling customers and partners to build the intelligent enterprise

Data Scientist

Enterprise System Developer

End-User

Image Speech Text

SAP Leonardo ML Foundation

Functional Services

Core Capabilities

Training APIs Consumption APIs

SAP Cloud Platform

Applications


SAP HANA Cloud Service & SAP Data Intelligence

Source: SAP HANA Cloud Service

https://saphanacloudservices.com/

27© 2018 SAP SE or an SAP affiliate company. All rights reserved.

SAP Data Intelligence

https://events.sap.com/sapandasug/en/session/45758

Combine SAP Leonardo ML, SAP Data Hub, and the best of python and R open source


Why is delivering Intelligent Applications difficult?


ML Application – Data and Model relationship challenge

Challenge: The customer needs more security on ROI from implementation of ML use cases.

Business metrics

A business metric should give a customer a feeling of the benefits promised by the

implementation of the ML scenario.

Technical metrics

Technical metrics serve another purpose: they should ensure that the system is technically

suitable for the implementation of the corresponding ML scenario. It especially concerns

the amount and quality of training data in the system which must be sufficient to

achieve appropriate performance of the ML model.


SAP’s Product Standards

Operations & Support

Performance Security Software Lifecycle

Accessibility Business Configuration

Functional Correctness

Globalization Licensing

UX Consistency

▪ SAP software can be used by everyone, including people with disabilities

▪ Configuration content as part of the product

▪ Completeness

▪ Correctness

▪ Translation

▪ Functional localization

▪ Internationalization

▪ To meet global needs

▪ Approval process for all use & distribution of commercial, open source and other third party software and services

▪ Business processes and system landscape operations

▪ Supportability

▪ Information Lifecycle

▪ Good Performance

▪ Scalability

▪ State-of-the-art security concept

▪ Security vulnerabilities

▪ Security legal requirements

▪ SAP strategy security topics

▪ Secure & reliable development

▪ Shipment, installation / deployment

▪ Technical configuration upgrade / update

▪ Un-installation / service termination

▪ Consistent SAP Fiori UX across products and technologies

▪ Design aspects like theming, terminology, icon usage, action placement

▪ Quality Attributes for product development

▪ Derived from ISO 25010 software quality model * * Successor to ISO 9126


ML Application – Operation and Lifecycle Challenges

Model adjustment / extensionmodel cannot be adjusted or extended easily,

if allowed, how will the lifecycle of extension

be managed

Model Deploymentallow multiple models (A/B testing)

history of active models

Phased OperationTrain and Deploy

Model Monitoring and

Degradation

Training and Retraining

Model Testingperformance can only be assessed on

productive data

Emergency Fallback &

Change of scopeML Applications


Statistician

Software Engineer

Data ScientistSoftware Architect

Operational Researcher

Industry

Academia

Extract Scale ConsumeConnect

(Domain Driven) Data Pattern

Dev/SE Processes

New

Testing Strategies…

…

(Data-Driven) Design Pattern

Quality Attributes

Contact information:

Tina Yang

Development Architect

SAP

Thank you.

© 2019 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of

SAP SE or an SAP affiliate company.

The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its

distributors contain proprietary software components of other software vendors. National product specifications may vary.

These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or

warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.

The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty

statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional

warranty.

In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or

any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,

and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and

functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason

without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or

functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ

materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they

should not be relied upon in making purchasing decisions.

SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered

trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names

mentioned are the trademarks of their respective companies.

See www.sap.com/copyright for additional trademark information and notices.

www.sap.com/contactsap

Follow us

https://www.sap.com/copyright

https://www.sap.com/registration/contact.html

https://www.linkedin.com/company/sap

https://www.youtube.com/user/SAP

https://twitter.com/sap

https://www.facebook.com/SAP

www.sap.com/germany/contactsap

© 2019 SAP SE oder ein SAP-Konzernunternehmen. Alle Rechte vorbehalten.

Weitergabe und Vervielfältigung dieser Publikation oder von Teilen daraus sind, zu welchem Zweck und in welcher Form auch immer,

ohne die ausdrückliche schriftliche Genehmigung durch SAP SE oder ein SAP-Konzernunternehmen nicht gestattet.

In dieser Publikation enthaltene Informationen können ohne vorherige Ankündigung geändert werden. Die von SAP SE oder deren

Vertriebsfirmen angebotenen Softwareprodukte können Softwarekomponenten auch anderer Softwarehersteller enthalten. Produkte

können länderspezifische Unterschiede aufweisen.

Die vorliegenden Unterlagen werden von der SAP SE oder einem SAP-Konzernunternehmen bereitgestellt und dienen ausschließlich

zu Informationszwecken. Die SAP SE oder ihre Konzernunternehmen übernehmen keinerlei Haftung oder Gewährleistung für Fehler

oder Unvollständigkeiten in dieser Publikation. Die SAP SE oder ein SAP-Konzernunternehmen steht lediglich für Produkte und

Dienstleistungen nach der Maßgabe ein, die in der Vereinbarung über die jeweiligen Produkte und Dienstleistungen ausdrücklich

geregelt ist. Keine der hierin enthaltenen Informationen ist als zusätzliche Garantie zu interpretieren.

Insbesondere sind die SAP SE oder ihre Konzernunternehmen in keiner Weise verpflichtet, in dieser Publikation oder einer

zugehörigen Präsentation dargestellte Geschäftsabläufe zu verfolgen oder hierin wiedergegebene Funktionen zu entwickeln oder zu

veröffentlichen. Diese Publikation oder eine zugehörige Präsentation, die Strategie und etwaige künftige Entwicklungen, Produkte

und/oder Plattformen der SAP SE oder ihrer Konzernunternehmen können von der SAP SE oder ihren Konzernunternehmen jederzeit

und ohne Angabe von Gründen unangekündigt geändert werden. Die in dieser Publikation enthaltenen Informationen stellen keine

Zusage, kein Versprechen und keine rechtliche Verpflichtung zur Lieferung von Material, Code oder Funktionen dar. Sämtliche

vorausschauenden Aussagen unterliegen unterschiedlichen Risiken und Unsicherheiten, durch die die tatsächlichen Ergebnisse von

den Erwartungen abweichen können. Dem Leser wird empfohlen, diesen vorausschauenden Aussagen kein übertriebenes Vertrauen

zu schenken und sich bei Kaufentscheidungen nicht auf sie zu stützen.

SAP und andere in diesem Dokument erwähnte Produkte und Dienstleistungen von SAP sowie die dazugehörigen Logos sind Marken

oder eingetragene Marken der SAP SE (oder von einem SAP-Konzernunternehmen) in Deutschland und verschiedenen anderen

Ländern weltweit. Alle anderen Namen von Produkten und Dienstleistungen sind Marken der jeweiligen Firmen.

Zusätzliche Informationen zur Marke und Vermerke finden Sie auf der Seite www.sap.com/corporate/de/legal/copyright.html.

SAP folgen auf

https://www.sap.com/germany/registration/contact.html

https://www.sap.com/corporate/de/legal/copyright.html

https://www.linkedin.com/company/sap

https://www.youtube.com/user/SAP

https://twitter.com/sap

https://www.facebook.com/SAP

Seeing SEMLA through the eyes of Software Engineers ... · Predictive Maintenance Load Forecasting...

Documents

Transcript of Seeing SEMLA through the eyes of Software Engineers ... · Predictive Maintenance Load Forecasting...