X

Disseminating the experiments from an explorer, Exploring with Oracle Technologies, Cloud and Blockchain

  • August 16, 2015

Internet of Things and Business Process Management – IoT & BPM

Vivek Acharya
Consulting Technical Manager

Data
from devices will need to be analyzed and actions will need to be taken based
on that data. These actions could trigger alerts or invoke corrective processes
before routine issues snowball into disasters (examples include flight delays,
parts replacements, fire emergencies, etc). These actions will impact critical
business processes requiring integration with operational systems, from
enterprise resource planning (ERP) and customer relationship management (CRM)
to specialist vertical applications.

IoT Architectural Framework

The IoT services
layer is completely independent from the underlying devices, communication
protocols and connectivity semantics. This layer includes a core set of services
to build IoT applications (i.e. composite applications) across a range of
industry sectors. The IoT services layer helps the enterprise to:

  • Analyze data in real-time (event processing)
  • Act on M2M data and events (integration services)
  • Provide historical, real-time and predictive analytics
    (analytics services)
  • Visualize operational and analytical data through mobile/desktop
    (UI services)
  • Manage data
    security and identity of devices/apps (security and identity management
    Service)
  • The IoT
    developer services layer enables developers to build applications using
    IoT services, development kits, software tools and services. This layer
    helps expose the platform to a range of applications and use-cases.

The role of middleware is to provide the
infrastructure and IoT services which in turn help drive innovation, enable new
revenue streams, and improve operational efficiencies.

Benefits
of IoT and BPM Integration

Embedding intelligence by way of real-time data
gathering from gateways and devices and consuming them through business
processes helps businesses achieve not just cost savings and efficiency but
also helps them generate more revenue patterns. Businesses need to overcome
several business and service challenges to be able to realize smooth
orchestration and manageability of disparate systems.

Use
Case - Outage Initiation over IoT

When we talk about ‘Internet of things’, we are
talking about -

Sense
à Acquire à Communicate à Event Processing à Integrate à Visualize and Analyze

Let’s walk though this with a sample use case of
outage management.

Use
Case explained
–We are here talking about outage caused due
to brownout (drop in voltage in the supply), faults at power stations, fault in
supply devices etc. If power outage not resolved within limits, can lead to, customer
compensation, penalties and loss of revenue. In this use case, we will talk
about 2 things – first is how IoT can lead to the initiation of a BPM process
which is going to handle the outage incident. Second is the BPM process itself,
where we will Model the process using Process Cloud Service and leave it for
implementation by developers.

  1. Equipment sends
    status information at a regular rate to the smart gateway. [Sense]
  2. Smart gateways
    have Oracle event processing for java embedded, which is a small foot
    print of Oracle event processing which is deployment on the gateways. It
    performs various upstream operations to perform basic filtering and
    aggregation, local decisions eliminate noise/false positive, optimize
    bandwidth, etc.
    Oracle Event Processing for Java Embedded has the ability to
    handle millions of events per second with microseconds of processing
    latencies. Oracle Event Processing for Java Embedded installed in the
    gateway appliance, helps filter and analyze real time usage, and detect fault
    and problem event patterns across thousands of distributed data centers,
    enabling dynamic diagnostic and automatic correction interception. [Acquire]
  3. Events of
    significance are sent from the gateway to the backend systems for a more
    detailed analysis. [Communicate]
  4. Oracle Event
    Processing Server edition performs complex downstream operations combining
    and correlating multiple streams of data putting it in a larger context
    (Customer information from CRM, etc.) Oracle Event Processing delivers on
    real-time analysis of high-velocity data. It is a complete solution for
    building IoT applications to filter, correlate and process events in real
    time so that downstream applications are driven by true, real-time
    intelligence. Oracle Event Processing filters out noise (such as data
    ticks without any change in values) and helps identify critical conditions
    before this data is actually relayed to the back-end. Oracle Event
    Processing can trigger a business event to perform integration with the
    BPM process. [Event processing]
  5. Processes
    essentially “listen” for event patterns & issues as they arise.
    Powered with this insight, systems can trigger alerts or invoke corrective
    processes immediately before routine issues snowball into disasters.[Integrate]

· Using BPEL and BPMN industry standards, users
can model process that capture optimal paths, alternative paths, exception
flows, process conversations, and handling of business events.

· Processes can be invoked directly on receipt of
events from devices or after the events have been pre-processed by event
processing engines such as Oracle Event Processing

· Oracle BPM thus provides the ability to
integrate processes that involve devices, applications and human intervention.

  1. Processes can
    be integrated with business intelligence infrastructure to further
    optimize core processes and operations. SOA processes can invoke Oracle
    Business Intelligence to gather contextual information In IoT use-cases,
    this is essential when sensor data or events lack enough context to
    determine how the data should be processed. Combining real-time sensor
    information with historical and prescriptive analytics can make the
    processes intelligent and responses much more human centric. SOA and BPM
    processes acting on sensor-related data and events are tightly integrated
    with Oracle Business Activity Monitoring, which provides dashboards that
    allow administrators to make decisions based on real-time streaming
    information coming either directly from sensing devices/gateways or events
    arriving from Oracle Event Processing engine. [Visualize and Analyze]

Key
Take away -
Following are the key -

  • Oracle SOA Suite on Oracle Exalogic
    provides IoT applications the ability to scale, delivering faster response
    time, and delivers 15X more throughput gains, 2X faster response time, and
    2X improvement in SOA file processing for large payloads.
  • Integration Pattern - The devices should be decoupled with the
    applications. Loose coupling is achieved through abstracting and resolving
    the differences between two or more systems in order to facilitate a
    seamless integration.
  • Loose Coupling & Virtualization
    - Oracle Service Bus integrates new devices and device services and
    enables true plug and play supporting a wide variety of applications and
    services. Oracle Service Bus is designed to connect, mediate, and manage
    interactions between device communication/event processing modules and
    business services instances across an expanding service network. Oracle
    Service Bus enables loose coupling between the device layer and the
    enterprise services (through virtualization) while providing guaranteed
    reliability and scalability of enterprise services.
  • How the right
    application instance/process is initiated? -
    Oracle
    Service Bus virtualizes the REST services exposed by Oracle Event
    Processing. At the same time, Oracle Service Bus can gather all the
    operational statistics of exposed IoT services, monitor service level
    agreements (SLAs) and then initiate actions such as invoke the right
    application instance in order to initiate action based on the gathered
    statistics.

Use
case – sample process

In the sample process “Outage Incident Management”,
there are various options to initiate the process. Customer can report outage
by one-click option on the ‘Reporting Apps’ or customer can call the back office
representative and report the issue. However it's not just a human intervention
that can lead to process invocation.

Devices too can report the problem. ‘SensorDetection' is a REST service with a POST
method. It accepts device id, area/zip in input. This REST service is invoked
by Oracle Event Processing - OEP. This service results in the invocation of the
‘Outage Incident Management’ process which will take care of the incident
resolution.


How
it works


Oracle Service Bus virtualizes the REST services exposed by Oracle Event
Processing. Oracle Service Bus can
gather all the operational statistics of exposed IoT services, monitor service
level agreements (SLAs) and then initiate actions such as invoke the right
application instance in order to initiate action based on the gathered
statistics.


Processes can be invoked directly on receipt of events from devices or after
the events have been pre-processed by event processing engines such as Oracle
Event Processing.



In this case, Oracle Event Processing REST service is exposed and virtualized
by OSB and is connected with the devices. Oracle event processing in turn will
invoke the right set of process/service and in this case it invokes the outage
management process by invoking the 'SensorDetection' service.

Following is the process flow screen capture , part
encircled shows the message event part of the process, which expose the process
as 'SensorDetection' service.

Conclusion

Self-driving cars will soon be a reality.
Connected devices powered with a unique identifier enabled by IPV6,watches,
home security monitoring, smart phones, home devices such as lights, home alarm
systems, thermostats and other appliances etc are resulting in vast data and
information which needs treatment beyond just transmission and processing, that
need to be managed, tracked and utilized in a seamless manner.

When a sensor receives data from a device, it
transmits/generates events. These events need to be captured to get the essence
out of it and to manage it better and to connect people with the automated systems
and processes. It’s not just about string the information extracted from the
data or it’s just about analyzing the data, it’s also about managing this information
and bringing in automated systems and processes. It’s about connecting smart
objects including human intuition. Smart devices will keep sending data and
intelligent devices will keep generating information out of it, to initiative
response to those events.

Response can be in the form of process initiation
which in turn invokes various systems and processes.

BPM into IoT will result in a properly managed IoT
& an effective communication between devices. Just by capturing events from
devices will not lead to an end-to-end solution, it’s the inclusion of BPM
which will result in an End-to-End solution. Merger of BPM and IoT will bring
in analytics, social and mobile capabilities into current processes. Process
and Business analytics will add more size and depth to data.

Join the discussion

Comments ( 1 )
  • priyaappleinfoway Thursday, September 12, 2019
    This is really a very good blog, and thanks for sharing it with the community.
    bpm company in chennai
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.