Challenges of Integrating Ubiquitous Industrial ...princetonacm.acm.org/tcfpro/presentations/Challenges of IIoT in Clo… · Collection, analysis and usage of data generated in industrial

© 2017 WIPRO LTD | WWW.WIPRO.COM | CONFIDENTIAL1

Speaker: Vikas S. Shah

Challenges of Integrating Ubiquitous Challenges of Integrating Ubiquitous Challenges of Integrating Ubiquitous Challenges of Integrating Ubiquitous

Industrial InternetIndustrial InternetIndustrial InternetIndustrial Internet----ofofofof----Thing (IIoT) in Thing (IIoT) in Thing (IIoT) in Thing (IIoT) in

Enablement of Multiple Enablement of Multiple Enablement of Multiple Enablement of Multiple CloudsCloudsCloudsClouds


Section#Section#Section#Section# SectionsSectionsSectionsSections

1111 OverviewOverviewOverviewOverview

2222 Primary ConcernsPrimary ConcernsPrimary ConcernsPrimary Concerns of Industrial Internetof Industrial Internetof Industrial Internetof Industrial Internet----ofofofof----Things (Things (Things (Things (IIoTIIoTIIoTIIoT))))

3333 Ubiquitous Ubiquitous Ubiquitous Ubiquitous IIoTIIoTIIoTIIoT Framework in Heterogeneous CloudsFramework in Heterogeneous CloudsFramework in Heterogeneous CloudsFramework in Heterogeneous Clouds

4444 Cognitive Agents of Cognitive Agents of Cognitive Agents of Cognitive Agents of IIoTIIoTIIoTIIoT EcosystemEcosystemEcosystemEcosystem

5555 Continuous Integration,Continuous Integration,Continuous Integration,Continuous Integration, Evaluation, and Evaluation, and Evaluation, and Evaluation, and ImprovementsImprovementsImprovementsImprovements

6666 Observations and TrendsObservations and TrendsObservations and TrendsObservations and Trends

7777 ConclusionsConclusionsConclusionsConclusions


Overview Industrial Internet-of-Things and Heterogeneous Clouds


Wipro Ltd. �� Connected Enterprise Services

Wipro L tdtdtdtd

• 173000+ workforce 173000+ workforce 173000+ workforce 173000+ workforce serving clients in 175+ cities across 6 continents

• 55+ dedicated emerging technologies 55+ dedicated emerging technologies 55+ dedicated emerging technologies 55+ dedicated emerging technologies ‘Center of Excellence’

• Trusted partnerTrusted partnerTrusted partnerTrusted partner of choice for global businesses

Connected Connected Connected Connected Enterprise ServicesEnterprise ServicesEnterprise ServicesEnterprise Services

• Connected customer experience Connected customer experience Connected customer experience Connected customer experience and integrated enterprise

•Elevating the brand and connecting to the communities where the business operates

•Integration competency center

• 200+ customers 200+ customers 200+ customers 200+ customers and $900M+ revenue

Vikas ShahVikas ShahVikas ShahVikas Shah

• 20+ Years 20+ Years 20+ Years 20+ Years of IT experiences in diversified technologies and verticals

• 22+ Publications and 6 Patents pending22+ Publications and 6 Patents pending22+ Publications and 6 Patents pending22+ Publications and 6 Patents pending in digital technologies and real-time enterprises

•CES architecture consultingconsultingconsultingconsulting to pre-sales enablement and workshops


Lifecycle of Industrial Internet-of-Things

Installing and Configuring

Sensing

Integrating and Communicating

Operating and Monitoring

Analyzing and Decision making

Commissioning, Automating, and Acting


IIoT – Smart Connected Ecosystem Evolution

Manufacturing Governance and

Planning

Product and Services - Business

Systems

IIoT Enabled Connected

Operations

IIoT Enabled Assets: Sensors,

Controls, and other Devices

IIoT Enabled

Ecosystem


IIoT Solutioning and Product Paradigms

Functional

•Connectivity – devices, data transports, event processing, and security

•Cloud integration – Private, Public, and Hybrid. IaaS, SaaS, and PaaS

•Application development – IDE, Workflows, Social, Mobile, Search, and Controls

•Big data analytics

•Automation

Operational

•Different levels of maturity and standards

•Diversified data types

•Heterogeneous infrastructure

•Time binding operations

•Environmental impact

Executional

•Asset visibility, reliability, and predictive maintenance

•Utilization visibility and benchmarking

•Traceability and serialization

•Flexible operations

•Environmental health and safety


IIoT Expectations and Presence

Industry

Verticals

Predictive &

Prescriptive

Manufacturing Healthcare Retail Agriculture Wearable Utility

Unified view of the devices participation to provide decision making capabilities during each

phase of IIoT lifecycle

Continuous improvement to provide predictive + prescriptive event derivation

Capitalize on a new revenue stream while ensuring higher service quality

Analytics

Continuously analyzing real-time equipment sensor data through deep learning algorithms

Cognitively utilize internal and external data sources to co-operate decision making process

Evolve analytics based on new conditions including environment and economic data

Monitoring

Collection, analysis and usage of data generated in industrial operations and throughout the

entire product lifecycle

Ensure performance and availability utilizing real-time sensor data for condition monitoring

Granular visibility of the relevant data for sensors, equipment, plants, and infrastructure

Automation

Automating events across the IIoT ecosystem of devices in publish/subscribe model

Act upon in conjunction of all types of SLAs and their inter/intra-relationship

Eliminate manual interventions and faults implementing intelligently switching to alternate path

or variations

Workforce

MGMTFleet

Energy

Efficiency

Smart

BuildingSecurity


CB Insight View of IIoT Transformation


Primary Concerns Integrating IIoT in Growing Paradigms of Heterogeneous Clouds


Challenges of IIoT in Heterogeneous Clouds (HCs)

� Different resource constraints Different resource constraints Different resource constraints Different resource constraints of IIoT devices to be recognized at the edge of HC (Example: processing strength and integrity)

� Today’s HCs formulated are HCs formulated are HCs formulated are HCs formulated are dynamic dynamic dynamic dynamic as network conditions can change rapidly

� Large spread on device capabilities to be device capabilities to be device capabilities to be device capabilities to be recognizedrecognizedrecognizedrecognized, additionally there are sources sources sources sources and and and and sinks of information sinks of information sinks of information sinks of information across IIoT ecosystem

� HCs and mashup of IIoTIIoTIIoTIIoT applicability applicability applicability applicability are largely unstructuredare largely unstructuredare largely unstructuredare largely unstructured. They can vary rapidly based on the published APIs, specific vertical, and their overlap

� Associations of edge users and their roles to Associations of edge users and their roles to Associations of edge users and their roles to Associations of edge users and their roles to IIoTIIoTIIoTIIoT in context of Hin context of Hin context of Hin context of HC that facilitates publishing custom IIoT APIs by the edge users (or participants of HCs)

� Monitor Monitor Monitor Monitor socio economical concerns socio economical concerns socio economical concerns socio economical concerns and impacts across the IIoT Cloud and respective events


Key Influencing Factors

• IIoT channels and

integration between the

verticals

• High performing IIoT

devices are deployed and

advanced

• HC vendor landscape is

being reshaped and

increased integrity among

them

• IIoT application

development to leverage

‘Fit to purpose’.

• Rich set of analytics,

inbuilt intelligence, and

automation techniques

FactorsFactorsFactorsFactors


Illustration: Usecase of IIoT in HC

Globalization: EndGlobalization: EndGlobalization: EndGlobalization: End----totototo----end end end end SupplySupplySupplySupply----chain of Shoes’ chain of Shoes’ chain of Shoes’ chain of Shoes’ ManufacturingManufacturingManufacturingManufacturing

� Raw material distribution

� Raw material types: Leather, wood, adhesive, color, packaging material,

color, nails, and fitting & decorative items

� Suppliers: India and most of the south Asian countries

� Manufacturing plant locations: China, Mexico, and other parts of

South America

� Transportation: Ship mode, Airplane mode, and Ground mode.

� Distribution: USA

� Retail: All across the world including cities of developed countries

� Recycling of recalled products and wastage: Africa and most of the

underdeveloped countries


Smart Manufacturing Plant Events

Event 1 (E1): Order Manufacturing of 10,000 shoes of type “X” [Initiation Event from SupplierCloud to Vendor Cloud]

EEEE1111----AAAA: AssignAssignAssignAssign CompanyCompanyCompanyCompany “C”“C”“C”“C” inininin ChinaChinaChinaChina [Conditions[Conditions[Conditions[Conditions VendorVendorVendorVendor CloudCloudCloudCloud]]]]

� E1-A-1: Transition to Company “C” Cloud to set Manufacturing Initiation

� E1-A-2: Identify start and completion dates

� E1-A-3: Set configuration, controls, and status of devices for manufacturing product [EventHierarchy “E2”]

� E1-A-4: Run manufacturing process from 8:00AM to 5:00PM [Supplier Cloud to VendorCloud Collaborative Event]

� E1-A-5: Track government (China) regulations to operate devices [Event Hierarchy “E3”:eGov of China Cloud to Vendor Cloud event Hierarchy]

� E1-A-6: Monitor devices for accuracy and fault based on standards defined by the vendors[Conditions: to Vendor Cloud in correlation to Supplier Cloud]

� E1-A-7: Identify fault [Event Hierarchy “E4” for Fault type {F1, F9, F12}]

� E1-A-8: Completion of the manufacturing process [Vendor Cloud Event]

� E1-A-9: Confirmation of completion [Vendor cloud to Supplier Cloud Event]

� E1-A-10: Assign to Ship Company “S” [Event Hierarchy “E5”: Supplier Cloud to ShippingCompany Cloud Collaboration Event]


Variation due to Diversified Conditions

E1E1E1E1----BBBB: Assign to Company “M” in Mexico: Assign to Company “M” in Mexico: Assign to Company “M” in Mexico: Assign to Company “M” in Mexico

� E1-B-5: Track government (Mexico) regulations to operate devices [Event Hierarchy “E3”]

� E1-B-10: Assign Airline “A” [Event Hierarchy “E6”]

E1E1E1E1----AAAA----1111 E1E1E1E1----AAAA----2222 E1E1E1E1----AAAA----3333 E1E1E1E1----AAAA----4444 E1E1E1E1----AAAA----5555 E1E1E1E1----AAAA----6666 E1E1E1E1----AAAA----7777 E1E1E1E1----AAAA----8888 E1E1E1E1----AAAA----9999E1E1E1E1----AAAA----

10101010

E1E1E1E1----BBBB----1111 E1E1E1E1----BBBB----2222 E1E1E1E1----BBBB----3333 E1E1E1E1----BBBB----4444 E1E1E1E1----BBBB----5555 E1E1E1E1----BBBB----6666 E1E1E1E1----BBBB----7777 E1E1E1E1----BBBB----8888 E1E1E1E1----BBBB----9999E1E1E1E1----BBBB----

10101010

Event (E1)

E1-A (Vendor Cloud in China)

E1-A (Vendor Cloud in Mexico)

Condition?Condition?Condition?Condition?

E2 E3 E4

E5

Supplied Cloud Supplied Cloud Supplied Cloud Supplied Cloud

to Shipping to Shipping to Shipping to Shipping

Company CloudCompany CloudCompany CloudCompany Cloud

Supplied Cloud Supplied Cloud Supplied Cloud Supplied Cloud

to Shipping to Shipping to Shipping to Shipping

Company CloudCompany CloudCompany CloudCompany Cloud

E6


Selection of Condition in Diversification

Selection between Selection between Selection between Selection between Event 1Event 1Event 1Event 1----A A A A and 1and 1and 1and 1----B depends on the numerous B depends on the numerous B depends on the numerous B depends on the numerous factors as described below. Which one will take higher precedence? factors as described below. Which one will take higher precedence? factors as described below. Which one will take higher precedence? factors as described below. Which one will take higher precedence?

� environmental condition associated with China and Mexico (from eGovt Cloud associated with devices defined in icGMM),

� contractual agreements of maximum limit set (from Siebel Cloud on contracts),

� condition of the machines and devices (from provider’s private cloud),

� cost of manufacturing based on present currency value (from supplier’s cloud),

� present labor conditions of different vendors (public cloud), and

� transportation related environmental impact (environment agencies’ cloud)


IIoT integration under the influence of HC

Types of IssuesTypes of IssuesTypes of IssuesTypes of Issues

• Advancements in connectivity and connectivity and connectivity and connectivity and

convergenceconvergenceconvergenceconvergence due to potentials of newly

introduced IIoT devices

• Upcoming regulatory and legality regulatory and legality regulatory and legality regulatory and legality of

IIoT technologies in assertion of

globalization (Industry 4.0 and P2413)

• Desired level of intelligence level of intelligence level of intelligence level of intelligence and

accuracies for specific industry vertical

and goals

• OOOOperational perational perational perational agility agility agility agility when introducing

new paradigms of either existing Cloud

or new Cloud environment

• Requirement of upgradesRequirement of upgradesRequirement of upgradesRequirement of upgrades and changes

required due to changing dynamics of

IIoT ecosystem including device level

agreements

Approaching ResolutionApproaching ResolutionApproaching ResolutionApproaching Resolution

• Analyzing operational operational operational operational requirementsrequirementsrequirementsrequirements,

including need for flexibility and

expandability between IT and OT

• Close integration Close integration Close integration Close integration between IIOT, HC, and

Edge Users

• Agreement between suppliers and end

users; and between plant engineers,

process engineers, and data scientists

when developing and implementing developing and implementing developing and implementing developing and implementing

solutionssolutionssolutionssolutions

• Consideration of how the IIoT solution

will be maintained and fine-tuned over

time under the influence of HCsinfluence of HCsinfluence of HCsinfluence of HCs

• Ability to adapt to new business logic

• Addressing cybersecurity cybersecurity cybersecurity cybersecurity including

secure and robust interfaces

• Clear definition of the SLAs between

IIoT and HC and integrity with actual integrity with actual integrity with actual integrity with actual

business level KPIsbusiness level KPIsbusiness level KPIsbusiness level KPIs


Ubiquitous IIoT Framework in Heterogeneous Clouds

Components and their Extensibilities


Factors to Consider

� How IIoTIIoTIIoTIIoT applicationsapplicationsapplicationsapplications areareareare architected,architected,architected,architected, deployed,deployed,deployed,deployed, andandandand managedmanagedmanagedmanaged in

the vicinity of the heterogeneous clouds?

� What aspects are required to be modeled for maintaining the

integrityintegrityintegrityintegrity betweenbetweenbetweenbetween IIoTIIoTIIoTIIoT,,,, devices,devices,devices,devices, andandandand CloudCloudCloudCloud edgeedgeedgeedge usersusersusersusers?

� How to represent the IIoTIIoTIIoTIIoT operationaloperationaloperationaloperational correlationscorrelationscorrelationscorrelations withwithwithwith

heterogeneousheterogeneousheterogeneousheterogeneous cloudscloudscloudsclouds and decision capabilities in presence of

distributed intelligence?

� How to combine distributed IIoTIIoTIIoTIIoT applicationapplicationapplicationapplication logiclogiclogiclogic totototo generategenerategenerategenerate

coherentcoherentcoherentcoherent servicesservicesservicesservices at the edge?

� The levellevellevellevel ofofofof manualmanualmanualmanual interventionsinterventionsinterventionsinterventions necessarynecessarynecessarynecessary to perform functions

of IIoT?


Ubiquitous IIoT Framework for Heterogeneous Clouds


Layered Approach

� HeterogeneousHeterogeneousHeterogeneousHeterogeneous CloudCloudCloudCloud EnvironmentEnvironmentEnvironmentEnvironment ManagerManagerManagerManager (HCEM)(HCEM)(HCEM)(HCEM): The primaryresponsibilities of HCEM is to perform gateway for multiple clouds to beparticipate in IIoT ecosystem.

� AgentAgentAgentAgent OrchestratorOrchestratorOrchestratorOrchestrator andandandand CorrelationCorrelationCorrelationCorrelation EngineEngineEngineEngine (AOCE)(AOCE)(AOCE)(AOCE): AOCE constitutes themodeling capabilities for the agent specification, the formalization of agentroles and responsibilities, and defines relationship with the specific set ofcloud services.

� CognitiveCognitiveCognitiveCognitive MicroservicesMicroservicesMicroservicesMicroservices LayerLayerLayerLayer ((((CMsLCMsLCMsLCMsL)))): CMsL is primarily responsible forcomposition of the agents identified and modeled in AOCE with either new ordeveloped set of Microservices. It leverages IIoT APIs specific knowledgemanagement and decision tree to provide context to the agents.

� IIoTIIoTIIoTIIoT APIAPIAPIAPI ElementElementElementElement ManagerManagerManagerManager (IAEM)(IAEM)(IAEM)(IAEM): IAEM is responsible for configuring theservices running on the CMsL associating IIoT APIs. Metadata are managedwithin IAEM.

� IIoTIIoTIIoTIIoT APIAPIAPIAPI AdministratorAdministratorAdministratorAdministrator (IAA)(IAA)(IAA)(IAA): IAA is responsible for introducing, managing,monitoring, and deprecating IIoT APIs. IoT APIs are published by numeroususers or participants of heterogeneous Clouds.


Architecture Principles

� IIoT vertically layered two pass cognitive agent architecture and dynamic control process of the agent

� Layered architecture needs to place Microservices with various degrees of abstraction and complexity of events of HCs

� Every layer uses the operations primitives of the lower layer to achieve its goals

� Both, the CAs and Microservices, are multi-layered

� The definition of artifacts are bottom-up, that is specific IIoT API receives control over a functionalities only when this exceeds the capabilities of the layer beyond

� Every CA consists of two functionalities:

- scenario recognition and goal activation

- planning and scheduling module


Advantages

• Technology CompetitivenessTechnology CompetitivenessTechnology CompetitivenessTechnology Competitiveness

• Dynamic ChangesDynamic ChangesDynamic ChangesDynamic Changes

• Streamlining DeploymentStreamlining DeploymentStreamlining DeploymentStreamlining Deployment

• GovernanceGovernanceGovernanceGovernance

• IntelligenceIntelligenceIntelligenceIntelligence

• IntegrityIntegrityIntegrityIntegrity

• AgilityAgilityAgilityAgility

• Operational VisibilityOperational VisibilityOperational VisibilityOperational Visibility

Implication of Ubiquitous Implication of Ubiquitous Implication of Ubiquitous Implication of Ubiquitous IIoTIIoTIIoTIIoT in HCs in HCs in HCs in HCs

�Flexibility and adaptability in the

management and deployment of

workflows of HCs

�Eliminates the need for new IIoT APIs due

to changing requirements of users or

devices

�Introduces an intrinsic competitive

environment for IIoT ecosystem

�Features decentralization allowing

abstraction to HCs and system resources

�CA have self-properties and allow the

direct manipulation of the devices and

their APIs

�Distributed information processing closer

to actual data sources in HCs

�Transmits execution state from devices to

APIs in HC

�Dynamic reuse of hardware components

�Monitoring, alerting, and dynamic changes

based on runtime behavior and conditions


Usecase View after implying Framework


Cognitive Agents of IIoT EcosystemModeling, Implementing, and Classification


Modeling CA for IIoT Framework


General IIoT Multiple Cognitive Agents Architecture

Reasoner

Information aboutInformation aboutInformation aboutInformation about

HCHCHCHC

- what it knows

- what it believes

- what is able to do

- how it is able to do

- what it wants

HC Environment HC Environment HC Environment HC Environment andandandand

other other other other AgentsAgentsAgentsAgents

- knowledge

- beliefsControls

Planner

Scheduler

& Executer

States

IIoTIIoTIIoTIIoT APIAPIAPIAPI

IIoT API

Serviceability

IIoT API Initial

Criteria


Classification of Cognitive Agents (CAs)

The classification emphasizes to precisely representing thecharacteristics of the IIoT services to CAs with respective to the contextsof heterogeneous clouds

� Cloud Service Subscriber CA (CSSCloud Service Subscriber CA (CSSCloud Service Subscriber CA (CSSCloud Service Subscriber CA (CSS))))

� Information Distribution CA (Information Distribution CA (Information Distribution CA (Information Distribution CA (ID)ID)ID)ID)

� MicroservicesMicroservicesMicroservicesMicroservices Deployment and Migration CA (Deployment and Migration CA (Deployment and Migration CA (Deployment and Migration CA (MDM)MDM)MDM)MDM)

� Edge Edge Edge Edge User Access CA User Access CA User Access CA User Access CA (EUA)(EUA)(EUA)(EUA)

� API Control CA (API Control CA (API Control CA (API Control CA (AC)AC)AC)AC)

� Edge Edge Edge Edge Interaction CA (EIInteraction CA (EIInteraction CA (EIInteraction CA (EI))))

� IIoTIIoTIIoTIIoT API API API API Regulatory CA (Regulatory CA (Regulatory CA (Regulatory CA (AR)AR)AR)AR)

� Resource Resource Resource Resource Reconfiguration CA (RRReconfiguration CA (RRReconfiguration CA (RRReconfiguration CA (RR))))

� Resource Utility CA (Resource Utility CA (Resource Utility CA (Resource Utility CA (RU)RU)RU)RU)

� Proactive Proactive Proactive Proactive Enablement of API CA (PEAEnablement of API CA (PEAEnablement of API CA (PEAEnablement of API CA (PEA))))

� Operational Response CA (OPROperational Response CA (OPROperational Response CA (OPROperational Response CA (OPR))))


CCCC

AAAA

Key Enablers of Microservices’ Implementation

Platform Level

Microservice

AAAA

PPPP

IIII

oData

Composite

Microservice

AAAA

PPPP

IIII

AAAA

PPPP

IIII

Data Level

MicroserviceData

Business

Logic

CCCC

AAAA

CCCC

AAAA

IIoTIIoTIIoTIIoT

Ubiquitous Ubiquitous Ubiquitous Ubiquitous

FrameworkFrameworkFrameworkFramework

JSON

XML

REST


Microservices at Scale of CAs

• Derive device layer SLAs Derive device layer SLAs Derive device layer SLAs Derive device layer SLAs within the modeling paradigms of Microservices: Response time, Latency, Availability, and Durability of associated data

• Standardizing integrity Standardizing integrity Standardizing integrity Standardizing integrity of Composite Micoservices and right level of isolation to reduce erroneous execution: Define alternate paths and error conditions

• Consistent behavioral modeling Consistent behavioral modeling Consistent behavioral modeling Consistent behavioral modeling when defining Microservice’s operations in presence of diversified scenarios pertaining to cognitive logic

• Streamline degrade and Streamline degrade and Streamline degrade and Streamline degrade and uuuupgrade of pgrade of pgrade of pgrade of MicroservicesMicroservicesMicroservicesMicroservices’ functionalities ’ functionalities ’ functionalities ’ functionalities based on demand of digital technologies in collaboration with allied business operation

• Utilize dynamic service registries Utilize dynamic service registries Utilize dynamic service registries Utilize dynamic service registries to describe, discover, publish, subscribe, version, and monitor Microservices

• Imply performance optimization Imply performance optimization Imply performance optimization Imply performance optimization techniques in consideration of physical resources utilized during execution of Microservice’s operations


Continuous Integration and Evaluation CA Classification Injection and IIoT Maturity Model


Continuous Integration of CAs to HCs


Scenario based IIoT Metrics in adherence to CA ClassificationHCs’ response latency and number of HCs’ response latency and number of HCs’ response latency and number of HCs’ response latency and number of

IIoTIIoTIIoTIIoT APIsAPIsAPIsAPIs

SLA violations and number of concurrent SLA violations and number of concurrent SLA violations and number of concurrent SLA violations and number of concurrent

IIoTIIoTIIoTIIoT APIs APIs APIs APIs

Average Average Average Average standard deviation in throughput standard deviation in throughput standard deviation in throughput standard deviation in throughput

of of of of the the the the MicroservicesMicroservicesMicroservicesMicroservices and and and and IIoTIIoTIIoTIIoT APIs utilizing APIs utilizing APIs utilizing APIs utilizing

HCs HCs HCs HCs


IIoT Maturity Model based Evaluation


Leverage Rating based Evaluation Graph

0

10

20

30

40

50

60

70

80

90

100

Adaptability + Acceptability (Enterprise)Adaptability + Acceptability (Enterprise)Adaptability + Acceptability (Enterprise)Adaptability + Acceptability (Enterprise)

ConsistencyConsistencyConsistencyConsistency

AdoptionAdoptionAdoptionAdoption

Business ContinuityBusiness ContinuityBusiness ContinuityBusiness Continuity

Standardization and ComplianceStandardization and ComplianceStandardization and ComplianceStandardization and Compliance

Time-to-MarketTime-to-MarketTime-to-MarketTime-to-Market

ManagementManagementManagementManagement

EvolutionEvolutionEvolutionEvolution

Roles and ResponsibilitiesRoles and ResponsibilitiesRoles and ResponsibilitiesRoles and Responsibilities

SecuritySecuritySecuritySecurity

EntitlementEntitlementEntitlementEntitlementSLA FactorizationSLA FactorizationSLA FactorizationSLA Factorization

ScalabilityScalabilityScalabilityScalability

ReliabilityReliabilityReliabilityReliability

PerformancePerformancePerformancePerformance

Complexity v/s CoherencyComplexity v/s CoherencyComplexity v/s CoherencyComplexity v/s Coherency

Change Management and UpdatesChange Management and UpdatesChange Management and UpdatesChange Management and Updates

Provisioning (Capability and Capacity)Provisioning (Capability and Capacity)Provisioning (Capability and Capacity)Provisioning (Capability and Capacity)

Extendibility and MaintainabilityExtendibility and MaintainabilityExtendibility and MaintainabilityExtendibility and Maintainability

Interoperability, Compatibility, andInteroperability, Compatibility, andInteroperability, Compatibility, andInteroperability, Compatibility, and

PortabilityPortabilityPortabilityPortability

PMO/ SDLC MetricsPMO/ SDLC MetricsPMO/ SDLC MetricsPMO/ SDLC Metrics

Weight

[Total: 100]

Rating [1..10]

[10: Best]

Total [%]


Observations and TrendsIIoT Focus Areas


Key Observations

� IndustrialIndustrialIndustrialIndustrial devicedevicedevicedevice----levellevellevellevel connectivityconnectivityconnectivityconnectivity remainsremainsremainsremains oneoneoneone ofofofof thethethethe mostmostmostmostfragmentedfragmentedfragmentedfragmented interfaceinterfaceinterfaceinterface areas for IIoT, with many proprietary and/or defacto standard protocols in use to communicate with HCs

� IIoTIIoTIIoTIIoT willwillwillwill playplayplayplay anananan importantimportantimportantimportant rolerolerolerole acrossacrossacrossacross thethethethe entireentireentireentire lifecyclelifecyclelifecyclelifecycle of anindustrial plant; from initial design and engineering during theupfront, capital expenditure-based project stage, through the muchlonger operational expenditure-based operations and maintenancelifecycle phase, where value is achieved from industrial assets

� Analytics combined with focused, structured data sets enablessystematicsystematicsystematicsystematic andandandand adaptiveadaptiveadaptiveadaptive approachesapproachesapproachesapproaches for IIoT APIs under HCs

� IIoTIIoTIIoTIIoT withwithwithwith intelligenceintelligenceintelligenceintelligence enables proactive maintenance increasesofferings to pursue HC integration specific to configurable dimensionsincluding specific industry verticle

� Ubiquitous framework provides consistentconsistentconsistentconsistent andandandand continuouscontinuouscontinuouscontinuous operationaloperationaloperationaloperationalexcellenceexcellenceexcellenceexcellence across HCs


Trends

Technology TrendsTechnology TrendsTechnology TrendsTechnology Trends Vendors’ LandscapeVendors’ LandscapeVendors’ LandscapeVendors’ Landscape

• Smart sensors and controllers

• Equipment-mounted sensors (temperature,

pressure, vibration, etc.)

• Soft sensors

• Process variable transmitters

• Analytical transmitters

• HMIs

• Mobility devices (smartphones, tablets, etc.)

• Smart positioners and drives

• Industry specific API proliferation

• IIoT Application/data integration

convergence

• Prepackaged devices and plant level

automation

• Internet of Everything - Over 30 billion

objects will be connected by 2020

• Facilitating visibility in operations

through open ended monitoring tools

• Enabling predictive maintenance for

integrated IIoT ecosystem

• Proactive IIoT asset management

platforms, tools, and approaches

• Subscription based intelligence and

knowledge in Clouds from diversified

vendors

• Gathering and analyzing data to

transform Big Data into actionable

intelligence irrespective of the specific

industry segment


Conclusions and Key ReferencesCurrent R & D and Industrialization


Conclusions

�HC computing is quickly becoming a dominant to ubiquitously

managing computational resources across multiple aspects of IIoT.

�IoT devices and their corresponding APIs are predominantly taking

advantages of cloud computing.

�The challenge is in architecting the future of connected IIoT ecosystem

that should capable of simplifying the integration between HCs, IIoT

APIs that are published or subscribed by edge users, and edge users’

devices or services.

�Core concepts and design principles of Microservices to develop

ubiquitous IIoT framework to constitute relationship between IIoT

APIs and HCs provides desired level of extendibility and governance

�An affirmative approach to compose (or model) and classify CAs

within the framework based on characteristics and functionalities of

inherent IIoT devices and representative data sets are essential


Key References

� The Industrial IoT: 56 Startups Transforming Factory Floors, Oil Fields, And Supply

Chains, CB Insight, January 2016

� Roundup Of Internet Of Things Forecasts And Market Estimates, 2016

� Building Microservices – Designing Fine-Grained Systems, Sam Newman, January

2015

� Process Automation and the IoT: Yokogawa’s VigilantPlant Approach to the Connected

Industrial Enterprise, 2015

� Multi-agent Systems, Adina Magda Florea, 2013

� Cognitive Agent Oriented Architecture of Internet-of-Thing Applications in

Heterogeneous Clouds of Mobile Edge Computing, IEEE 2nd International Conference

Cloudification of the Internet of Things (CIoT 2016), November 2016

� PROVISIONING AN INDUSTRIAL INTERNET-OF-THINGS CONTROL FRAMEWORK OF

DYNAMIC MULTI-CLOUD EVENTS AND DEVICES THEREOF, US Patent Application ID:

15/420397, January 2017

� CLASSIFICATION OF HETEROGENEOUS CLOUDS, US Patent Application IS: 15/432,570,

February 2017


Vikas S. Shah

[email protected]

Connected Enterprise Services

Documents

Challenges of Integrating Ubiquitous Industrial ...princetonacm.acm.org/tcfpro/presentations/Challenges of IIoT in Clo… · Collection, analysis and usage of data generated in industrial