Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
© 2017 SAP SE. All rights reserved.
SAP Predictive Maintenance & Services
Alan Southall, SVP Engineering, Head of PdMS | S...
© 2017 SAP SE. All rights reserved.
Disclaimer
This presentation outlines our general product direction and should not be ...
© 2017 SAP SE. All rights reserved.
Agenda
SAP & Internet of Things
Connecting Things to Business Processes
Predictive Mai...
© 2017 SAP SE. All rights reserved.
Agenda
SAP & Internet of Things
Connecting Things to Business Processes
Predictive Mai...
© 2017 SAP SE. All rights reserved.
Delivering Outstanding Results to Customers and Stakeholders
Customers
87%
of Forbes
G...
© 2017 SAP SE. All rights reserved.
SAP Leonardo empowers the LIVE business
Connecting Things to Business Processes
Busine...
© 2017 SAP SE. All rights reserved.
SAP Leonardo Portfolio
© 2017 SAP SE. All rights reserved.
Agenda
SAP & Internet of Things
Connecting Things to Business Processes
Predictive Mai...
© 2017 SAP SE. All rights reserved.
Connected Asset Lifecycle
SAP IoT and PdMS delivers tangible business outcome
Failurer...
© 2017 SAP SE. All rights reserved.
Connected Asset Lifecycle
SAP IoT and PdMS delivers tangible business outcome
Failurer...
© 2017 SAP SE. All rights reserved.
Connected Asset Lifecycle
SAP IoT and PdMS delivers tangible business outcome
Fleet Le...
© 2017 SAP SE. All rights reserved.
Business Value for SAP PdMS
Overall cost reduction in maintenance efforts
~1% machines...
© 2017 SAP SE. All rights reserved.
Business Value for SAP PdMS
Overall cost reduction in maintenance efforts
Health of As...
© 2017 SAP SE. All rights reserved.
Edge, Connectivity
and Storage
Key Challenges
Rare Events | Data Quality and Varsity |...
© 2017 SAP SE. All rights reserved.
Edge, Connectivity
and Storage
Key Challenges
Rare Events | Data Quality and Varsity |...
© 2017 SAP SE. All rights reserved.
PDMS Machine Learning Engine Overview
Usable for any asset type and manufacturer
Data
...
© 2017 SAP SE. All rights reserved.
General Approach
• Learn the normal behavior
• Principal Component Analysis (PCA) rota...
© 2017 SAP SE. All rights reserved.
Anomaly Detection
With Distance Based Failure
Rank Battery
1 128
2 348
3 133
4 144
5 0...
© 2017 SAP SE. All rights reserved.
Agenda
SAP & Internet of Things
Connecting Things to Business Processes
Predictive Mai...
© 2017 SAP SE. All rights reserved.
Customer Example
Compressor Manufacturer
Company
One of the largestproviders of compre...
© 2017 SAP SE. All rights reserved.
Customer Example
GEA Separators
Company
GEA is one of the largest suppliersof process ...
© 2017 SAP SE. All rights reserved.
Customer Example
Train Operator
Company
The company owns and operatesa fleet of around...
© 2017 SAP SE. All rights reserved.
SAP Predictive
Maintenanceand
Services
Thank you.
Upcoming SlideShare
Loading in …5
×

Alan Southall, SVP of Engineering, Head of IoT Predictive Maintenance, SAP

660 views

Published on

MIT Enterprise Forum of Cambridge Connected Things 2017 Keynote Speaker: Alan Southall, SVP of Engineering, Head of IoT Predictive Maintenance, SAP

Published in: Business
  • Be the first to comment

  • Be the first to like this

Alan Southall, SVP of Engineering, Head of IoT Predictive Maintenance, SAP

  1. 1. © 2017 SAP SE. All rights reserved. SAP Predictive Maintenance & Services Alan Southall, SVP Engineering, Head of PdMS | SAP IoT | March 2017
  2. 2. © 2017 SAP SE. All rights reserved. Disclaimer This presentation outlines our general product direction and should not be relied on in making a purchase decision. This presentation is not subject to your license agreement or any other agreement with SAP. SAP has no obligation to pursue any course of business outlined in this presentation or to develop or release any functionality mentioned in this presentation. This presentation and SAP's strategy and possible future developments are subject to change and may be changed by SAP at any time for any reason without notice. 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. SAP assumes no responsibility for errors or omissions in this document, except if such damages were caused by SAP intentionally or grossly negligent.
  3. 3. © 2017 SAP SE. All rights reserved. Agenda SAP & Internet of Things Connecting Things to Business Processes Predictive Maintenance & Services Combining IT/OT data to optimize maintenance Customer Stories Real life applications of predictive maintenance
  4. 4. © 2017 SAP SE. All rights reserved. Agenda SAP & Internet of Things Connecting Things to Business Processes Predictive Maintenance & Services Combining IT/OT data to optimize maintenance Customer Stories Real life applications of predictive maintenance
  5. 5. © 2017 SAP SE. All rights reserved. Delivering Outstanding Results to Customers and Stakeholders Customers 87% of Forbes Global 2000 98% of the 100 most valued brands Financials €14.87B (+6%) software and software- related services revenue €5.5 B (+4%) software and cloud revenue 100% of Forbes top sustainability companies 80%+ are SME companies €17.5 B (+4%) Total revenue Solutions 25 Industries 11 Lines of business Employees 74,406 employees EMEA: 33,340 Americas: 22,071 APJ: 18,995 79% Employee Engagement Index SAP HANA 5,800 SAP HANA customers 1,800 startups 8,500 trained partners 120+ nationalities worldwide 70% Business Health Culture Index Source: SAP Corporate Fact Sheet 1/2015; SAP Integrated Report 3/2015
  6. 6. © 2017 SAP SE. All rights reserved. SAP Leonardo empowers the LIVE business Connecting Things to Business Processes Business Processes SAP Leonardo Foundation SAP Cloud Platform SAP Leonardo Applications Things Next Level of Experience Sources Of DataIntegration | Business Partners | Networks Machine LearningBlockchain
  7. 7. © 2017 SAP SE. All rights reserved. SAP Leonardo Portfolio
  8. 8. © 2017 SAP SE. All rights reserved. Agenda SAP & Internet of Things Connecting Things to Business Processes Predictive Maintenance & Services Combining IT/OT data to optimize maintenance Customer Stories Real life applications of predictive maintenance
  9. 9. © 2017 SAP SE. All rights reserved. Connected Asset Lifecycle SAP IoT and PdMS delivers tangible business outcome Failurerate Burn-in "infant mortality" Wear-outNormal life Asset lifetime Emerging Issues Detection (EID) Early identify, monitoring and managementof emergingasset issues using exploration, root cause and warranty analytics PredictiveMaintenanceandService (AHCC& VA) Holistic managementof asset health and dynamic optimizationof maintenance schedules and resources based on health scores, anomaly detection and spectral analysis AssetInvestmentOptimization and Simulation Analyze remaining useful life of assets to optimallyplan for new investments based on business needs, asset health and risk of failure. SAP ERP, S4HANA, CRM, C4C
  10. 10. © 2017 SAP SE. All rights reserved. Connected Asset Lifecycle SAP IoT and PdMS delivers tangible business outcome Failurerate Burn-in "infant mortality" Wear-outNormal life Asset lifetime Emerging Issues Detection (EID) Early identify, monitoring and managementof emergingasset issues using exploration, root cause and warranty analytics AssetInvestmentOptimization and Simulation Analyze remaining useful life of assets to optimallyplan for new investments based on business needs, asset health and risk of failure. SAP ERP, S4HANA, CRM, C4C PredictiveMaintenanceandService (AHCC& VA) Holistic managementof asset health and dynamic optimizationof maintenance schedules and resources based on health scores, anomaly detection and spectral analysis
  11. 11. © 2017 SAP SE. All rights reserved. Connected Asset Lifecycle SAP IoT and PdMS delivers tangible business outcome Fleet Level View Single Asset Details
  12. 12. © 2017 SAP SE. All rights reserved. Business Value for SAP PdMS Overall cost reduction in maintenance efforts ~1% machines with down time Convert unplanned maintenance to planned maintenance to avoid down- time and improved equipment effectiveness NumberofAssets Health of Asset / Maintenance Need
  13. 13. © 2017 SAP SE. All rights reserved. Business Value for SAP PdMS Overall cost reduction in maintenance efforts Health of Asset / Maintenance Need Dynamically optimize the entire maintenance schedule in order to reduce the overall maintenance costs and reduce components on stock NumberofAssets Standard maintenance interval for all assets the same Optimized maintenance interval per asset 6 services executed 4 services really needed 4 weeks 4 weeks4 weeks 4 weeks 4 weeks 4 weeks 5 weeks 5 weeks 5 weeks9 weeks 6 weeks 6 weeks 6 weeks 6 weeks 4 weeks 4 weeks4 weeks 4 weeks 4 weeks 4 weeks
  14. 14. © 2017 SAP SE. All rights reserved. Edge, Connectivity and Storage Key Challenges Rare Events | Data Quality and Varsity | Data Variety, Fusion and Volume v Dynamically optimize maintenance and service activities with prescriptive analytics v Integration into EAM, PM, MRS and AIN v Condition Monitoring v Onboarding v Device management v Security v Connectivity v Data ingestion v Big Data infrastructure v Avoid unplanned down-time v Improved equipment effectiveness v Reduce overall maintenance costs v Reduce components on stock v Data Preparation v Support data fusion process, i.e. sensor data combined with business information v Operationalized data fusion services v KFR Engine v Machine Learning Engine v Anomaly Detection v Ensemble Learning v Model Base Engine v FEA Engine IT/OT Convergence PdMS Derived Signal Management PdMS AHCC, DMM & Integration PdMS Business Outcome
  15. 15. © 2017 SAP SE. All rights reserved. Edge, Connectivity and Storage Key Challenges Rare Events | Data Quality and Varsity | Data Variety, Fusion and Volume v Dynamically optimize maintenance and service activities with prescriptive analytics v Integration into EAM, PM, MRS and AIN v Condition Monitoring v Onboarding v Device management v Security v Connectivity v Data ingestion v Big Data infrastructure v Avoid unplanned down-time v Improved equipment effectiveness v Reduce overall maintenance costs v Reduce components on stock v Data Preparation v Support data fusion process, i.e. sensor data combined with business information v Operationalized data fusion services v KFR Engine v Machine Learning Engine v AnomalyDetection v Ensemble Learning v Model Base Engine v FEA Engine IT/OT Convergence PdMS Derived Signal Management PdMS AHCC, DMM & Integration PdMS Business Outcome
  16. 16. © 2017 SAP SE. All rights reserved. PDMS Machine Learning Engine Overview Usable for any asset type and manufacturer Data Data Preparation, Fusion and Feature Selection Reinforcement using user feedback Health Scores & Alerts Create Work Activities Continuous learning & application to new data Continuous learning & application to new data Continuous learning & application to new data Failure Prediction using automatic ensemble learning on known failures New Algorithm Using extensibility Anomaly Detection using unsupervisedlearning without labeled data
  17. 17. © 2017 SAP SE. All rights reserved. General Approach • Learn the normal behavior • Principal Component Analysis (PCA) rotates the coordinate system to explain a major part of the variation of the data by the first few new coordinates • Detect deviation from normal • We apply PCA coordinates to search for multivariate anomalies using an adjusted sum of squares as scoring function • Choose a threshold at which a data point is considered an anomaly • An alert is being raised which has to be validate by a domain expert Anomaly Detection With Principal Component Analysis PCA
  18. 18. © 2017 SAP SE. All rights reserved. Anomaly Detection With Distance Based Failure Rank Battery 1 128 2 348 3 133 4 144 5 008 6 181 7 366 8 051 9 336 10 536 … 371 103 372 135 373 281 374 463 375 096 376 109 377 086 The algorithm is trained to inspect the data for you
  19. 19. © 2017 SAP SE. All rights reserved. Agenda SAP & Internet of Things Connecting Things to Business Processes Predictive Maintenance & Services Combining IT/OT data to optimize maintenance Customer Stories Real life applications of predictive maintenance
  20. 20. © 2017 SAP SE. All rights reserved. Customer Example Compressor Manufacturer Company One of the largestproviders of compressed airsystems and compressed air consultingservices. Situation:Changed the business model from selling compressors to sellingcompressed air Solution • Compressors equippedwith sensors • SAP PredictiveMaintenanceand Service solution • SAP HANA software • SAP CRM applicationfor use in service on SAP HANA IT and OT connectivity Asset health control center Fault pattern recognition Machine health prediction Create maintenance or service order Execute order on mobile device % 0011001 1101001 Visual supportSchedule order OrderStatus Non-SAP applications SAP S/4HANA C4C / CRM Process Innovation Benefits • IoT as an enabler for the new business model • Improved availability of compressor stations • Move from unplanned to planned maintenance
  21. 21. © 2017 SAP SE. All rights reserved. Customer Example GEA Separators Company GEA is one of the largest suppliersof process technology for the food industry and for a wide range of other industries. In 2015, GEA generated consolidatedrevenues in excess of about EUR 4.6 billion. Situation:Need for IoT solution in order to extend service business and as differentiatorto their competition. Solution • SAP PredictiveMaintenanceand Service solution • SAP PredictiveAnalyticssoftware • SAP CRM Process Innovation Benefits • Company: Ability to offer new higher margin service business models with lower service costs. • Its customer: Improved equipmentuptime and guidance for optimizedmaintenanceschedules. • Improved transparencyfor machine availabilityand usage pattern. • Remote monitoring and analysis of remote equipment • Optimizedspare parts exchange timelinesbased on maintenancecosts and costs due to materialdeterioration causing lower productionthroughputs • Classificationand patternrecognition basedon historic sensor data and error codes
  22. 22. © 2017 SAP SE. All rights reserved. Customer Example Train Operator Company The company owns and operatesa fleet of around 2.000 electro-trains,2.000 locomotivesand 30.000 coaches and wagons. Situation:40% of maintenanceeffort is for corrective maintenance. Solution • Data fusion betweenIT and OT data • Multidimensionalassets description • Remote train diagnostics • Engineeringrules and predictivemodels • Indicators-basedplanning • Dynamic optimizationof maintenanceschedules Process Innovation Benefits • Higher asset availability leading to higher passenger satisfaction • Less effort for corrective maintenance
  23. 23. © 2017 SAP SE. All rights reserved. SAP Predictive Maintenanceand Services Thank you.

×