preventative maintenance of robots in automotive industry

Post on 09-Jan-2017

499 Views

Category:

Technology

2 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Hadoop Summit 2016 Preventive maintenance of Robots

in Automotive industry

Ari Flink, Amit Kumar

2© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

• Intro • IoT evolution, Big Data in IoT• Cisco Cloud Platform

• Case Study• Preventive maintenance of Robots in Automotive industry• Adaptive, self-learning next-gen Predictive maintenance platform

Agenda

3© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Ari fun fact: Kemi, Finland

4© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

About Ari

Solutions Architect at Cisco Cloud • Architect service deployments on Cisco’s cloud platform (BDaaS,

DBaaS, BSS)

Previously Operations Architect at WebEx, eBay, Excite@Home• Ensure operational readiness for complex distributed services• HA, DR,, config, deployment, monitoring, event correlation

5© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

What I love doing: Bikram Yoga @ 105 F

6© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Amit: my other passion

Big Data Architect at Cisco Cloud Platform & Services Group• Big Data Solutions for clients and infra needs using Hadoop, Cassandra• Analytics platform design• Data Center infra software abstraction : Firewall as a Service, Networking as a

Service.

Previously Symantec/Verisign, HCL-US, BoA• Distributed Systems design and implementation• Hadoop based solutions for large data sets

7© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Cisco Cloud Platform

Global platform deployed across

Cisco and SP Partners

API-driven, elastic experience for

developers,based on open

standards

Cisco-architected and operated for rapid application development and

deployment

8© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Audience?

• Big Data ?• Robotics / car manufacturer?• IoT ?

9© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

IoT evolution

10© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

The Four Eras of Compute

1960 1980 2005 2015

Mainframe

x86 Linux Web VMs iPhone

PC + Web Cloud + MobileIoT + Analytics + Automation (ML)

Cloud Containers

Enterprise Consumer IoT ( Machines )

11© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Why preventative maintenance for robots

12© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

How much does unplanned downtime cost a car manufacturer?

$20k per minute

How much can a single incident can cost?

$2 million

13© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Million dollar question

Which robot will fail next?

How can we predict robot failure?

14© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Keep the assembly line moving

15© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Why does a single robot failure matter?

16© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Zero Downtime• Cisco and Fanuc have created a Zero Downtime Solution (ZDT)

that analyzes data from robots to detect potential problems that could lead to a failure.

• ZDT is currently used in production with over 6,000 robots at automotive plants globally. GM alone has deployed ZDT in 27 factories in 5 countries analyzing over 5,000 of robots

• ZDT has successfully detected over 45 cases of potential failure across 26 production plants over the past year and saved already customers  $40 million

18© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Platform for Preventative Maintenance

19© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Overview

$2 million outage avoided !

Telemetrycollected

Notify robot manufacturer

and plant

PlantDataCollector

Cisco Cloud

Partswarehouse

Car plant

Scheduledmaintenance

20© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Cisco Cloud

Automotive manufacturer APlant 2

Plant 1

Plant Data Collector

Case study: Data Flow

Cisco IoT Platform

Plant 3

Cisco IoT Platform

Cisco BDaaS

ZDT application

Reporting

Analytics

Car manufacturers

Robot manufacturer

Automotive manufacturer BPlant 2

Plant 1

ZDT Data Collector

Cisco IoT Platform

Plant 3

Notifications

21© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Cisco Cloud

Car PlantCar Plant

Batch Layer

Cisco Cloud: High Level Arch Framework

Speed Layer

Serving LayerMasterdataset Batch

view Batch view

Real-time view

App

Car Plants

Batchprocessing

Real-timeview

Real-timeprocessing

DataIngestLayer

Datastream

22© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Cisco Cloud

Batch

Case study: ZDT Cisco Cloud Pilot details

Real-time

Serving

Masterdata

Computeddata

HBase

Ingest

Cisco IoT

Kafka

Flume

Spark Streaming

Batch processing:Pig, Hive

Impala

Impa

la S

QL

schema

Data

API

23© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

UI

Hadoop

Multi-tenancy

User Interfaces

API

SQL(Impala)

HDFS

Customer Portals

Mobile Devices PD BI

24© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Next Gen Platform: “Predictive” Maintenance

25© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Why Predictive? Car Production facilities operate at high volume

Unexpected downtime creates considerable losses There is a need to be informed of a potential robot, controller or process problem

before unexpected downtime occurs

Early detection is key in the following scenarios Mechanical failures Process control failures System issues: Controller Maintenance reminders

Not-too-early and not-too-late detection is “key” Too early is expensive in the long run Too late is detrimental as well Finding the sweet spot is key to the most “optimal solution”

26© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Sweet Spot: “Not-too-early” and “not-too-late” either

Time

Met

ric

Sweet spot

Too early Too late

27© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Preventative Maintenance

Unscheduled outage avoided

Torque out of range

Notify robot manufacturer

and plant

28© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Predictive analytics: Increased ROI

Repor

t• What

happened

Analyze

• Why did it happen

Monitor

• What is happening now

Predict• What

might happen

Increasing ROI and Complexity

29© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Data Modeling details

Initial Dataset

Run/Evaluate Models

Gather Data

Define Problem

Validation Dataset

Test Model

Select Model

Test Dataset

Apply Model Run Prediction

30© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Stream Processing Layer

HDFS

Data Ingest Layer

Predictive Analytics: High level architecture

Learning Layer

Action Layer

Rawdataset

Processeddataset

KafkaCisco

IoTPlatform

Near “real-time” (micro-batch) processing

( Spark )

Machine Learning ( Spark ML )

HDFSKnowledge

Base

Operational Dashboard platform

( custom built / Sensu

customized )

31© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Stream Processing Layer

HDFS

Data Ingest Layer

Predictive Analytics: High level architecture

Learning Layer

Action Layer

Rawdataset

Processeddataset

Near “real-time” (micro-batch) processing

( Spark )

Machine Learning ( Spark ML )

HDFS

Knowledge Base

32© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

File Formats: Avro vs Parquet vs ORC Avro is row-based

storage format, optimized for scans of all fields in a row for each query

Parquet is column-based, best used when dataset has many columns and only a few columns are worked on

ORC is column-based as well

33© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Spark based Predictive platform on Hadoop

Data Integration ( Kafka, Sqoop, Flume )

Storage for any type of data

Filesystem(HDFS)

Online NoSQL(HBase)

Workload Management ( YARN )

Machine Learning(Spark, Mahout)

Stream Processing(Spark)

34© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Stream Processing LayerData Ingest Layer

Predictive Analytics: High level architecture

Learning Layer

Action Layer

Operational Dashboard platform

( custom built / Sensu

customized )

35© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Action Layer

Predictive Analytics: Action layer

Event store

Event consumer API based event Topic

Consumer for email

Events from the

Predictive platform

Dashboard middle-tier

API for Ad-hoc queries

Consumer for PagerDuty

Custom built / Sensu

customized

36© 2015 Cisco and/or its affiliates. All rights reserved. Cisco Confidential

Recap

Unscheduled outage avoided: Savings $40 million

PRESENTATIONEND

top related