Predictive Analytics: Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform
 

Predictive Analytics: Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform

on

  • 1,451 views

Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform

Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform

Statistics

Views

Total Views
1,451
Views on SlideShare
1,421
Embed Views
30

Actions

Likes
0
Downloads
47
Comments
0

2 Embeds 30

http://www.content-loop.com 25
https://twitter.com 5

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment
  • Talking Points: S Harris <br /> <br /> Industrial machines have always issued early warnings, but in an inconsistent way and in a language that people could not understand. The advent of networked machines with embedded sensors and advanced analytics tools has changed that reality. For the first time in history, remotely distributed machines across the globe – from MRIs to wind turbines to aircraft engines – can be monitored in real time, unlocking the language of machines and opening tremendous benefits.
  • Add Examples to this slide by Industry
  • Predictive Model <br /> Benchmark Performance <br /> Predict Efficiency <br /> Online Dashboard <br /> Dynamically monitor and control deviations
  • Expected ROI/ROI examples
  • 18

Predictive Analytics: Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform Predictive Analytics: Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform Presentation Transcript

  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.2 Use the mobile app to complete a session survey 1. Access “My schedule” 2. Click on this session 3. Go to “Rate & review” If the session is not on your schedule, just find it via the session scheduler, click on this session and then go to “Rate & review”. Thank you for providing your feedback, which helps us enhance content for future events. Session BB3061 Speaker James Redlinger, Steffin Harris Please give me your feedback
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. Predictive Analytics Extending asset management framework for multi-industry applications leveraging HP’s Big Data HAVEn platform James Redlinger – HP Global Alliance CTO Steffin Harris – North American Big Data Leader, Capgemini June, 2014
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.4 Welcome to HP HAVEn Approaching Big Data with an intelligent software architecture Social Media IT/OT ImagesAudioVideo Transactional dataMobile Search engineEmail Texts Catalog massive volumes of distributed data Hadoop/ HDFS Process and index all information Autonomy IDOL Analyze at extreme scale in real time Vertica Collect and unify machine data Enterprise security Powering HP software and your apps nApps Documents HAVEn
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.5 Capgemini nApps in Action Commercial Insurance Risk Analytics (CIRA) Predictive Asset Management (PAM) Operating Efficiency Piecemeal Data Unified Information Operate at frontiers of efficiency & productivity Claim Cost Reduction Generic Approach Identify the Higher Risks Reducing the cost to settle claims Managing Risk Account Level Management Individual Risk Level Management Mitigate risk & improve compliance Customer Experience One Size Fits All Able to Reward The Good Risks Improved customer satisfaction Cost Management Missed Opportunities Lower Loss Ratios Increase profitability & improve use of capital
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.6 Webinar Speaker Introduction Practice: NA BIM Level: Principal Office: South San Francisco “Capgemini is helping to revolutionize the Big Data Industry by developing industry specific partner enabled solutions to solve our customers most challenging problems - HP Discover Barcelona 2013 Consulting Competencies  Big Data Strategy – Technologies, Platforms, Analytics  Project Mgmt, Governance, IT Strategy, Performance Mgmt  EDW Architecture, BI/DW Strategy, Design, Implementation, Run  HP Vertica, SAP, Oracle, Teradata, Hadoop, BI/BW/BOBJ, OBIEE, SSAS/RS,  ERP Design, Implementation & Delivery  Energy, Retail Consumer Products, High-Tech, Manufacturing, Telco. Business Experience / Education  20 years of Management Consulting Services  Global CTO Board Member  Big data platforms, Analytics, Services  Master Data Management, Knowledge Management, KPI Development  Business Intelligence (BI) Strategy, Enterprise Performance Management, Sales Performance Management, Infrastructure & IT Portfolio Optimization, KPI Development  Business Intelligence Solution Services (Arch, Technology & Process)  Education: Bachelor of Science, Airway Science, Masters, Information Systems Management. ”
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.7 Who is interested in this solution? What should my strategy be to ensure that we take full advantage of Big Data Analytics and remain relevant in the marketplace. How do I become the leader in the world of everything being connected? CEO We are helping client to improve reliability and availability of plant and equipment whilst attempting to reduce operational cost by using predictive asset analytics. COO Understanding the value drivers, real profitability through enterprise performance analytics is changing business strategies the focus is on revenue leakage management, working capital, expenses and controlling analysis and optimization of Assets. Client is talking Innovation, Big Data, Market is Talking Internet of Things – How can I be more effective in enabling the enterprise to support an IOT strategy? CFO CIO We are using detailed analysis of activities locations and schedules to optimize workforce, supply chains and logistics – achieving substantial performance improvements. COO Risk is a complex area, analytics (and increasing the use of big data) is helping our client to quantify and manage those risks in a proactive manner. Risk Manager
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.8 Big Data & Analytics Industry Outlook (II/IOT) Asset Intensive Industry Revolution powered by Industrial Internet, Internet of Things Source: GE’s Jeff Immelt’s vision on IoT
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.9 Predictive Asset Analytics Advanced Analytics to monitor, predict and optimize asset performance in an increasingly connected world Industrial Internet, Internet of Things Data Insights from Asset Analytics End Users Asset Register Sensors/ Telemetry – usage, operations settings, logs, events / alarms, … Asset Management Guidelines Performance Benchmarks Failure/ Warranty Claims Asset Service History Field/Technical Inspection Notes Asset & Services Catalog Asset Financials Social Media / Call Center Data Engineering, R&D Manufacturing Technical Services Quality Logistics, Sourcing Field Services Finance & HR Suppliers & Customers 3rd Party Service Providers  Reliability Metrics  Event Sequence  Plant Optimization. Optimize Equipment Reliability  Performance Patterns  Metrics/ KPIs & Drivers  Alerts & Control.  Benchmarking  Segmentation/Clustering  Business Rules.  Root Cause Analysis  Drivers & Time to failure  Failure Prediction.  Warranty & Claims Mgmt  Parts/ Field optimization  Fraud Detection. Operations Control Performance Management Service Optimization Predictive Asset Maintenance
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.10 Maintenance – The Maturity Model Reactive [fix when something fails] Preventative [OEM prescriptions, observations by technicians] Condition Based [rule based monitoring at real time to track anomalies] Predictive [analyze usage patterns, trends in behavior, maintenance history to predict failure well ahead of time] Business Drivers Prevent Unplanned Outages Improved MTBF, MTTR, MTTF Reduced Maintenance Improved budgeting & forecasting Improved decision making Minimize repeat repairs Unplanned Outages cost The Oil & Gas Industry $100B/Annually Unplanned Outages in The Utilities Industry can Cost the business $55K/Hr
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.11 Solution Overview – Asset Analytics Monitor, predict and optimize asset performance Joint Solution Developed with HP  Reliability Metrics  Event Sequence  Plant Optimization.  Benchmarking  Segmentation & Business Rules.  Usage Pattern  Health Monitoring  Alarms & Control.  Failure Prediction  Root Cause Analysis.  Predictive warranty  Parts optimization  Fraud Detection. Optimize Equipment Reliability Performance Management Operations Control Predictive Asset Maintenance Service Optimization Scale Hadoop/ HDFS Source Autonomy IDOL Speed Vertica Secure Enterprise Security Powering HP Software + your apps nApps Social media Video Audio Email Texts Mobile Transactional data Documents IT/OT Search engine Images HAVEn
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.12 Performance Monitoring & Control Actionable insights for Energy Management  The Client, a major Energy & Utility equipment manufacturer, launched a strategic program to design eco-friendly energy utilization solutions that define new global standards.  As part of the program, smart energy meters and environment sensors were installed in key areas of selected facilities. This data was made available for analysis  The Mandate: Analyze energy utilization and efficiency in a facility and extract insights to support actionable energy management strategies. Overview  Quality and Variations in data • Statistical analysis to test quality of data from the meters/sensors • Measures and reports to flag and alert on any deviations from expected standards  Energy Consumption Analysis • Analyze patterns and trends in energy consumption – by location, time, day, etc. • Relationships between energy consumption and facility use and facility comfort levels (light, heat, ventilation).  Data volume and quality was a limiting factor for much of the analytics  The data was from very different types of facilities and did not support comparative analysis or benchmarking  These data issues resulted in the project being designed as a proof of concept on “what is possible” using advanced analytics. Challenge/Issue  Used limited data to showcase the potential of analytics in energy management Client Testimonial  “…testify that the team was able, in three month’s time, to deliver six analytical frameworks on top of our data, in spite of the data’s relatively poor quality. As a conclusion, this project was a success and demonstrated the team’s skills”. Benefits  Exploratory cross tabulations including frequency counts, histograms etc.  Alerts the energy manager to actively identify deviations  ‘What if’ simulation for key Decisions. Solution
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.13 Predictive Asset Analytics Industry Solutions  Metrics/ KPI Definition by Department  Reports, Dashboards & Trends. Warranty Management for Indian Auto Manufacturer  Prediction of warranty claims and cost by vehicle & part. Warranty Claims Forecasting for US Auto Manufacturer  Prediction of out-of-Stock, Over- Stock Conditions  Market Product Penetration & Customer Satisfaction. Retail /Consumer Products  Analytics to understand drivers of energy consumption and derive control parameters. Energy Management for E&U Solution Provider  Analyze installed asset base and map to VAS to drive revenue. Service Opportunity Identification for E&U Equipment Manufacturer  Predict Network Demand  Failure prediction  Workforce Management. Telecom Network Analytics Performance Monitoring & Control for HVAC Equipment
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.14 Performance Monitoring & Control Chillers Plant Operational Efficiency Challenge/Issue  Large data storage (~1TB for medium size builders), unable to analyze in near real time.  Unable to know usage of the equipment & performance behavior.  What are the variables that controls equipment performance.  What will be control limits on various usage scenario KPI Description kwPerTon It is a measure of Chiller Efficiency – it is measured as utilization of kw energy to cool one ton of air inside the building at 40F. Lesser the better Chiller Supply Load Overrun (%) It is a measure to understand the gap between the chiller cooling load supply and building cooling requirement . • If it is positive => excess chiller cooling load supply => Energy saving opportunity • If it is negative => less chiller load supply => Customer dissatisfaction due to not meeting the right ambient temperature inside the building
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.15 Demo
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.16 Use cases by Industry by Asset Value Chain Industry Asset Manufacturing Asset Service Provider Asset Owner Energy & Utilities •Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Identification & Prediction • Labor Planning & Scheduling • Asset Reliability Analysis, RCM • Consumption Theft Modeling Oil, Gas & Petrochemical •Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Identification & Prediction • Labor Planning & Scheduling • Asset Reliability Analysis, RCM • Consumption Theft Modeling Heavy Engineering & Construction •Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Reliability Analysis, RCM Aerospace & Defense • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Reliability Analysis, RCM Heavy Vehicles & surface Transportation • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Reliability Analysis, RCM
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.17 Use cases by Industry by Asset Value Chain cont’d Industry Asset Manufacturing Asset Service Provider Asset Owner Communication Service Provider • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Usage Analysis, Tracking Household Assets – Electrical/Electron ic/Mechanical • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Usage Analysis, Tracking Light Vehicles • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Usage Analysis, Tracking Medical Devices • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Usage Analysis, Tracking Building HVAC Assets • Asset Reliability Analysis • Maintenance & Warranty Limits • Asset Failure Demand Prediction • Service Planning & Scheduling • Asset Usage Analysis, Tracking
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.18 Big Data challenges that are often overlooked 1. Did you just call my baby ugly? 2. That’s not what we used before! 3. It’s just commodity hardware, right? 4. Is your solution sustainable?
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.19 Wrap Up/Q&A For Additional Information Contact: James Redlinger HP Global Alliance CTO , James.redlinger@hp.com Steffin Harris NA Big Data Leader, Capgemini Steffin.Harris@capgemini.com @SHarrisSFO
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. Thank you
  • © Copyright 2014 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice.