improved operational insights by it/ot convergence · security remote data sync spatial text...
TRANSCRIPT
SAP CUSTOMER
June, 2018
Improved operational insights by IT/OT convergence: How to empower business users via self-service Analytics
and Machine Learning
Stojan Maleschlijski , Dimitrios Lyras
Platform & Data Management,
SAP Global Centre of Excellence
David Buckingham, former CEO i2C, AIMIA
WE NEED TO FIND IT, EXTRACT IT, REFINE IT, DISTRIBUTE IT
AND MONETIZE IT.
3CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
The Oil and Gas industry
generates value from a mere 1%
of all the data it creates.
4CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
The “data tsunami” and the “digital storm”
Volumes of data are growing
at a rate of 40% per year and
will increase 50 times by
2020. https://www.networkworld.com/article/3241848/data-center/data-tsunami-to-absorb-20-of-world-electricity.html * World Economic Forum, January 2016 : What is the future of the internet?
In the next two to five years, the
“digital storm” will impact asset-
intensive and manufacturing
industries.
5CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
If Data is fuel in a digital world, Analytics is the engine
In leading organizations, executives agree that…
Decision making is
distributed across
the organization
Decisions can be
mapped directly to
company strategies
Decisions are
data driven
Decisions are
made in real time
75% 78% 62% 62%
Source: “Leaders 2020 The Next-Generation Executive: How Strong Leadership Pays off in the Digital Economy,” Oxford Economics, 2016
6CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Use cases
7CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Where are maintenance and service heading?Organizations are maturing their maintenance strategies
Companies are moving from a
reactive to a proactive approach
to maintenance.
An opportunity is available for
organizations to leverage machine
data for better business insights.
Condition-
basedReactive Preventive Predictive
Wait until a
machine fails
and then
Undertake
Maintenance.
Perform
maintenance at
regular intervals,
based on
observations of
Abnormalities.
Continuously
observe the status
of assets and react
to predefined
conditions and
events.
Apply advanced
analytics and operational
and business data to
help determine the
condition of specific
equipment and predict
when to perform
maintenance.
8CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Holistic framework for resolving production issues
TR
EA
TM
EN
TD
IAG
NO
SIS
TR
IAG
E
• Inform the user of a critical event prior to
happening
• Raise early warning signal
• Provide criticality information
• Help identify causes for abnormal situation
• Highlight abnormal process parameters
• Show historical trends
• Show past critical events
• Recommend actions to resolve issue
• Collect operator feedback on the
effectiveness of actions taken
• Update criticality status
I would like to have a
“Ronald” on every plant…
9CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Predictive Maintenance: From reactive to predictive
© 2017 SAP SE or an SAP affiliate company. All rights reserved.
What is the expected output of a plant inWhat is the predict lifespan of Remaining the next month? What is the expectedcomponents? What is the run-to-failure Useful Forecasting process time? How do machine failuresand time-to-failure rate? Lifetime and other factors affect it?
Which machines or components Key What are the main influencers ofAnomaliesshow anomalies compared to influencers machine failures?
healthy machines?
Extraction of
InsightsWhat are the overall key figuresWhat are the trends: historical andof a data set? Provide simple
Descriptive emerging, sudden step changes,summaries about the sample Trendsstatistics unusual numeric values thatand about the observations
impact the business?that have been made.
What are the correlations in the data?Predict the occurrence of failure of FailureRelationships
Which failure events occur together,machines or components using Prediction
which lead to a warranty claim?sensor data.
10CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Trend Analysis
Predictive Maintenance: From reactive to predictive
11CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Relationships Analysis
Detection of high/low correlated sensor data
© 2017 SAP SE or an SAP affiliate company. All rights reserved.
Predictive Maintenance: From reactive to predictive
12CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Identifying Patterns in Signals
Challenges:
▪ Identify all historical latency trends where:
▪ A “w-shaped” trend is observed as a result of delayed
crushing operation followed by application of corrective
actions
Automate the post-processing via manual/auto triggered
procedures
Graphically depict results to get valuable insights
13CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Leverage information from maintenance data
Challenges:
▪ Analyze maintenance data to obtain actionable
insights
▪ Estimate adherence to the execution of
maintenance activities and relate to production
errors
▪ Identify most frequently performed preventive /
corrective actions
▪ Establish items inventory and improve fix-rate via
identifying most often replaced assemblies
▪ Use unstructured textual information to identify
most frequently / most common machine failure
causes
14CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
DATABASE MANAGEMENT
Web Server JavaScript
Graphic Modeler
Data Virtualization ELT & Replication
Columnar OLTP+OLAP
Multi-Core & Parallelization
Advanced Compression
Multi-tenancy Multi-Tier Storage
Graph Predictive Search
DataQuality
SeriesData
R, Python, TensorFlow Integration
Hadoop & Spark Integration
Streaming Analytics
Application Lifecycle Management
High Availability &Disaster Recovery
OpennessDataModeling
Admin &Security
Remote Data Sync
Spatial
Text Analytics
Fiori UX
APPLICATION DEVELOPMENT DATA INTEGRATION & QUALITYADVANCED ANALYTICAL PROCESSING
OLTP + OLAP ONE Open Platform ONE Copy of the Data
SAP Data Hub ( Analytics Workflow Orchestration & Data Pipelining)
SAP HANA Data Management Suite (HDMS) + SAP Analytics Cloud
15CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Technology
Data Science
Domain
Knowledge
Why are machine learning and self-service analytics important?
• Enable SMEs to investigate data
• Allow self-service data preparation
• Provide ad-hoc analysis
• Break dependency from IT - “Analytics
for all”
• Allow the development and
operationalization of ML models
16CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
One piece of the story….
Analysis
Data Scientist
Exploration(flexibility)
Production(stability)
17CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Generating value from data – an end-to-end process
Data Preparation
Automation of data acquisition and
transformation to create input data
sets for data science use cases
1
2
3
4
Access Storage Preparation Visualization Analysis Productize Operate
Value
Self-Service Data Access
Integrate and transform data from flat
files, relational data bases, historians,
cloud storage, or Hadoop for
consumption in statistical models
Self-Service Advanced Analytical Modeling
Utilize SAP automated modelling APL, HANA Predictive
Analytics Library, R, HANA Text, Graph, or Geospatial
engines to generate new business insight
Business Process Integration
Integrate results with existing business
processes or develop new business
applications using BI, HANA SQL,
and/or HTML5 (SAP UI5)
18CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
“Don’t worry about him. He works in the paper industry, and he’s not happy that I’ve gone paperless.”
19CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
DEMO
Detection of quality deterioration in paper mills using
SAP’s self-service analytics platform
20CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
PRODUCTION SUPERVISOROPERATOR
MAINTENANCE ENGINEER
RAW MATERIAL MILL PRESS END PRODUCT
• Low evenness
• Low bonding strength
• Color fluctuations
• Wrong glossiness grade
Are there any maintenance actions needed?
Is there imminent thread for the production?
Can I prevent break-downs?
Is my production running smoothly?
Are there any alarming trends?
Are there any error events?
What are the reasons for trends?
Are there any correlations in my
data?
Can I predict quality problems?
21CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
System overviewV
ISU
AL
IZA
TIO
N
OPERATOR
Process monitor Self-service analytics
PRODUCTION
SUPERVISOR
MAINTENANCE
ENGINEER
(PDMS)
Assets
Import data
Prediction
Predictive Analysis Library
Business Data
(e.g. maintenance)
Part ordering, personnel
scheduling, etc.
SY
ST
EM
S
SAP MII,
Process historians
(e.g. OSIsoft PI)
22CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Key take-away messages
Combination of OTand IT data isbeneficial for theperformance of MLmodels. The platform allows
trend detection,optimization androot-cause analysisas well asautomated modelcreation.
Self-service datapreparation andself-serviceanalytics isprovided to SMEsusing the SAPplatform.
Model deploymentandoperationalizationindependent of IT.
All users across theorganization areinterested in“Analytics for all”.
23CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Thank you!
24CUSTOMER© 2018 SAP SE or an SAP affiliate company. All rights reserved. ǀ
Customer
Legal Disclaimer
The information in this presentation is confidential and proprietary to SAP and may not be disclosed without the
permission of SAP. This presentation is not subject to your license agreement or any other service or subscription
agreement with SAP. SAP has 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's strategy and possible future developments, products and or platforms directions and functionality are all
subject to change and may be changed by SAP 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. This document
is provided without a warranty of any kind, either express or implied, including but not limited to, the implied warranties of
merchantability, fitness for a particular purpose, or non-infringement. This document is for informational purposes and
may not be incorporated into a contract. SAP assumes no responsibility for errors or omissions in this document, except
if such damages were caused by SAP´s willful misconduct or gross negligence.
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,
which speak only as of their dates, and they should not be relied upon in making purchasing decisions.