Seeing SEMLA through the eyes of Software Engineers ... · Predictive Maintenance Load Forecasting...
Transcript of 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
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
4PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
5PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
1960s – 1980s 1990s – 2000s 2000s – 2010s 2010s – 2020s
6PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
7PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
8PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Our IntelligentEnterprise works as one to automate, anticipate, and invent.
9PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Manufacturing& Supply Chain
It connects every line of business through an Intelligent Suite of integrated applications
10PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
a Digital Platform to orchestrate data and integrate processes
11PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
and Intelligent Technologies to detect patterns, predict results, augment decisions,
And turn actions into outcomes.
12PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
13PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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.
14PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
15PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
20PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
21PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
22PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
23PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
24PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
25PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
26PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
SAP HANA Cloud Service & SAP Data Intelligence
Source: SAP HANA Cloud Service
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
28PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Why is delivering Intelligent Applications difficult?
29PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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.
30PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
31PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
32PUBLIC© 2019 SAP SE or an SAP affiliate company. All rights reserved. ǀ
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
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