XMPRO Recommendations - Event Intelligence Applications
XMPRO Recommendations are advanced event alerts that combine alerts, actions, and monitoring. You can create recommendations based on business rules and AI logic to recommend the best next actions to take when a certain event happens.
Go to https://xmpro.com if you want to learn more about how you can build real-time applications with our low-code Event Intelligence Platform.
Transcript
ex-emperor recommendations are advanced
event alerts that combines alerts
actions and monitoring it creates new
event alerts based on business rules and
or AR logic and it recommends the best
next action based on expert suggestions
it also monitors the actions and the
outcomes to close the loop on event
response
this means that operations can respond
to critical events based on expert
knowledge in the organization before the
opportunity expires while managers can
close the loop by monitoring that it's
done in a timely and appropriate manner
let me run you through the exemplary
recommendations creating simple
applications thought by bringing in data
from multiple data sources and we create
some application or visualization around
it and then put that in front of users
to respond to key or critical business
events with recommended actions based on
the knowledge of those experts in the
organization and that is the role of
recommendations in there in the
excellent pro system so let me show you
how that works this is an example of an
XM pro event app it was configured with
our app designer and it features an
event board in this instance that
highlights key events that are starting
to happen across certain areas of my
plant and certain equipment and for each
of those they are recommendations which
were triggered based on conditions for
this as well as recommended actions on
what to take and again there's some
other additional information from a
application point of view that might be
interesting for us what I'm interested
in at this stage is the recommendation
around this bump now I could drill into
the actual bump and look at it but what
I'll do from here is just have a quick
look and see what this recommendation is
what the alert was that triggered it and
what the recommended action is around
resolving this let me draw into this
bump so this is the data that triggered
that recommendation for certain
conditions when
it and I'll demonstrate that in a minute
what those rules are behind this but
this was the die turn that they
triggered that these are some of the
instructions around how to potentially
do this but because I've been working on
this blonde for a long time and I also
know some of the other conditions and
other equipment I could provide a
comment and you from my expert point of
view and provide input to the person
that actually needs to go and do
something on this now in this example
there's potentially a problem with the
in-line fault on the cooling tower and
that's why we're eating something around
the block suction pipe and it's causing
a challenge with the discharge pressure
so this is one of the mechanisms of
capturing knowledge that's been there
for a long for a long time people who
have had experience on that plant or
that type of equipment you can very
quickly identify some of the key things
that we might need to do so now that
I've done that I can either
automatically create a work order or
work request into a back-end system but
in this instance I will just put in this
work request number that I've created in
my yeah M system or plant maintenance
system and it will then as soon as I
save this and it finds that a
maintenance plan or someone has created
a work order for that it would then
bring that work order information back
for me in real time so that I can see
this has been actioned that information
is then brought together in terms of
looking at how we follow this process so
I can now monitor and that there was an
alert that I put in a request and that
request at some stage is turned into a
work order and I can also monitor the
work order status to make sure that
these alerts are being addressed now
how we set these alerts up is under the
rules side of it so we manage the
recommendations and again this is a kind
of a global view of all the
recommendations and I can resolve them
are they individually or as a group but
what I'll do in this instance is
actually look at the rules behind them
how do we set up these recommendations
now that specific one for the pump on
the discharge brazier and I'm gonna this
is not a detailed explanation but just a
very high-level explanation of how you
would put that together we had that out
of efficiency range rule so that you can
have multiple rules that you set up and
in this instance this information for
this rule comes from the data stream
that sits behind it so we've got real
time flowing data and this then
interrogate that data based on the
frequency that you said it could be
every second every five minutes if
you're half an hour half a day or a day
or whatever works for this specific
business guys in terms of how frequently
we want to run and check that against
this rule we've said the the parameters
for that it doesn't have to be a numeric
number it could also bring back me if I
use for example feed right and versus
flour right so if the flour is really
less than the feed right then I might
want to trigger this so this is how you
set up recommendations and those
recommendations are the ones that you
saw at the front where it creates those
red bubbles or red dots to tell me that
there's something wrong and I can now
quickly see key events I can quickly see
what the recommendations are and then I
can start monitoring the process in
terms of how long did it take to
generate in terms of someone being
assigned assigned to resolve and on how
long does it take us to resolve these
and also how many of these
recommendations are being acted upon and
that provides the full feedback loop to
make sure that we don't only alert but
we also recommend what to do and then
check that that is being done in the
best and timely way
Last updated