Water is going to be scarce with oil in and demand is going to exceed supply by now if you look at water
as a system right there are many many
moving parts in play here right you need
to acquire water you need to treat it
store it transport it to distribute it and
then continue must look like us I'll end
up using it now this whole system
is riddled with inefficiencies it's and
and as demand for water increases we
tend to put more stress on each of these
individual walks now let's dig a little
deeper shall we around % of the earth is
water right that's a lot of water
but and a half percent of that as
oceans so that's not usable that's not
for human use so all of the remaining
three and a half percent seventy percents glaciers and ice so if you if you
keep going down this funnel right you'll
realize that only less than % it's
available for human use and even within
that % is for agriculture % is used
by industrial supply systems so in the
end a teeny tiny . percent of all
available water is for us the consumers
like you and I can consume right
that's a very tiny fraction and and
that's why the cost of water is rapidly
increasing now if you look at silica
the report published over the last three decades if you look at the consumer
price trends and how-how the costs are
varied for like water natural gas
electricity you'll see that the cost of
water has risen by four hundred percent
right and that's more than
electricity and natural gas combined and
that's what and it's it will continue to
get worse right
and so the scarcity factor is gonna just
continue to let worse here over here now
let's see like how did we get here right
it's one of the main reasons is that our
the system is filled with inefficiencies
like leakage and we waste water and the
pipes that are aging pipes and there are
many more reasons right
all these things have contributed
and now we are in this situation where
the cost of war is rapidly increasing
and there's no drinking water available
for everyone on earth and if you talk
about leakage around two points one
trillion gallons of clean treated water
it's just lost every year in u.s.
right it's due to leaks faulty meters
broken water mains and this can actually
feed tens of millions of people for an
entire year right that's that's a lot of
water to waste and this fact is this
activated by the fact that the average
age of pipes when the u.s. in the US and
Canada is around years which means
there are some pipes that are like
hundred years old at just a couple of
weeks ago up in Berkeley, there was a
broken water main like just thousands of
gallons just too wasted and water supply
was shut out for for many people and the
the reason this happened was there was a
rushing leak in a cast-iron pipe was
installed in that's years old
it's crazy and the problem is this is
pervasive across the way it's not
an easy thing to fix because it's
extremely expensive right just to fix
the aging pipes in the US alone it takes
billion dollars now this is one of
the most developed countries in the
world imagine what's happening and some
of the less developed countries, it's
it's enormous the problem is huge and so
one of the ways to solve
this problem is we can do smart things
with data right
if you cannot replace all the hardware
if you cannot replace all the pipes
why don't we monitor it and be smart
about it and that's where that's where
data comes into the picture we currently
live in a data-driven economy where huge
industries are completely changing the
ways they work using data right for
example let us take manufacturing just a
a few decades ago people like humans used
to stand in assembly lines and do them
jobs and now it's completely like robots
and automation has completely taken over
this industry same thing is true with a
the supply chain for example all right
instead of humans trying to figure out
with their spreadsheets how do we know
route and plan this whole thing they are
using data analytics
to optimally are out in the schedule
everything so that they're way more
efficient it's much faster and they're
just way more comfortable as well right
and so water is the one is one
such industry all right if you look at
the trend right now to take the balls
over time all right back in the s
people were using spreadsheets and
manual tools right and then people moved
on from software is basically
something that's installed on your
the computer is no cloud right and then
people move to cloud-based solutions so
that person any people anywhere can
access the same solution from from
multiple geographical locations and then
they are now living in an AI turbid
a world where everything is powered by
data and artificial intelligence all
right so the entities that refuse to
adapt are just going to perish because
the world now is a lot less tolerant to
encompass right then it was said in
so data is the answer right so how do we
how do we use data what is data anyway
right number one is we just need to do like
two things efficiently like collect data
efficiently and do smart things with it
all right looks pretty straightforward
what's the big problem here well let's
see all right let's say let's say
there's a pipe that's installed like
lets like a hundred miles away from
where you live right and you want to
measure how much water is flowing
through it and so what you do is you
install a nice meter and then it's going
to measure the water now the problem is
for you to know what's happening that
five you need to send a guy who will go
there and look at the readings all right
now imagine they're like every guy's
responsible for like hundreds and
hundreds of these like thousands of
these meters right so obviously you
cannot hope to send a guy every single
the hour which is not gonna happen so and
that's why they take these readings once
every two months so what's gonna happen
between the two successive readings
right anything all the data is just
being lost because it was just not
looking at it right and that's where the
efficiency part comes into the picture how
can we continuously monitor water
everywhere without the kind of without
allocating a lot of manpower to it and
number two is how do we do smart things
with it which means if you collect all
the data what
of doing it if-if you're not extracting
meaningful insights and altering your
behavior so that's where the
smartness things now consider picture so
let's talk a little bit about data
the collection now to collect data
efficiently we do avoid sending like
like a human every time to get those
readings you can use Internet-connected
wireless sensors so what these sensors
can do many things like measure
temperature pressure vibration water
flow soil moisture you can you name it
and these sensors can do it and these
sensors are wirelessly connected to the
Internet so we can just stream data to
the cloud and you can just access it
without signing a human alright and this
this ecosystem of intimate connected
sensors are called the Internet of Things I
don't know if you for the bus I'm sure
you've heard of it like the thing is a
sensor so basically an ecosystem of
sensors I can measure these vital things
that is ecosystem we're heading towards
and that's how you collect data very
efficiently and number two is basically
a lot of water data is already available
there are billions of sensors billions
of sensors and mirrors that are already
collecting data it's all we need to do
some of them are connected to the
Internet most of them are not and that's
where it's why we gotta figure out how
can we get that data so that we can we
can do something with it and thing
is we need that data because of you
should aspire to live in a world where
decisions are driven by data right human
intuition is grossly insufficient when
it comes to decision making right
because we are dealing with like
exabytes of data over here which means
humans cannot just use this guesswork to
solve the problems right and that's
where efficiently understanding data
becomes very very handy and managing
water effectively has a lot of benefits
number one you can cut the usage right
under two we can reduce the risk of
the shortage you can reduce avoid expensively
operations and maintenance so basically
being smart with data has a lot of
benefits especially to water right and
we need to aspire to live this world
driven by data because without data
it was just another person with an
opinion right and that's been happening
a lot right that's why we need to move
away from it in the past people used to
use that experience or guesswork say
that hey you know I've seen this might
look familiar let me let me use my
intuition to do something with it that's
that's usually very very suboptimal now
how does it work in practice
number one is we just we need to collect
data continuously right we need to
collect continuously we need to analyze
the data with algorithms and then need
to take real-time decisions now
collecting data continuously right-that's is that's being done using
Internet of Things which means that all
these sensors that are installed they
are they're gathering data continuously
in streaming to the cloud all right and
let us talk a little bit about
artificial intelligence artificial
intelligence is the field of study that
deals with creating machines that can
can function like humans right that can
perform tasks that require human
intelligence right and a lot of the
research in this field is it's based on
emulating the learning process for human
brain human brain is amazing it's like
the best form of AI artificial
intelligence because it can learn
with effortlessly and it almost
happens subconsciously you don't you
need to try and you know chairs a
chair right it's not so easy for
machines to do it because of machines just
need a lot of data to understand even
simplest things and so artificial
intelligence and there's a paradigm
within artificial intelligence called
deep learning that especially focuses on
on this that they create models based on
human brain so that we can emulate them
learning the process and be really efficient
with it right and so basically you
measure learn and you infer right that's
if you keep this loop in any system you
can be very very efficient all right
let's take an example now let's say you
want to monitor water usage by utilizing
flow sensors and residential
buildings now these flow sensors are
retrofit which means that you can just
they just put it on top of a pipe and i
start measuring the water flow and it'll
start streaming the data to the cloud
now in this case by efficiently
measuring the flow of water right you
can be you can do gonna be way more
efficient but just the act of measuring
it and looking at it will alter your
behavior because if you don't
know how what is being wasted you won't
do anything about it but if you know
hey maybe I'm you know using way more
water for my lawn or this there's a
a lot of leaks and in the building I need
to do something about it just by looking
at the data, I will alter your behavior
and that's why we need to monitor water
usage and then understand how can
we can be more efficient with it right
and so there are many many other
problems you can solve as well right
there can detect leakage can track
trends you can identify anomalies can
forecast demand you can predict the
events and you can do a lot more
thanks, I depending on where you are if
you are dueling dealing with buildings
you can do these things you can do the
same for irrigation systems for
utilities industrial supply systems and
so just the act of measuring it will
make you improve it there's a very
a famous saying by Peter Drucker that goes
like if we cannot measure
it cannot improve it right so mature
measuring is an important part measuring
efficiently is even more important and
then doing smart things with it is how
you make it more efficient now how can
we right how can we contribute what can
we do right here and when water is not a
the new problem right what it's been around
for a while I mean when you hear
people talk about hey waters leaking
water is getting wasted people to don't
have access to clean drinking water if
you look around the world like
million people don't have access to
clean drinking water that's a lot of
people it's and so what's uh what's
what's changed now it's been around for
a while why now the problem is or the
the solution is now we now have the tools to
solve the problem like years ago you
you were not in a position to kind of
install these high-end sensors at low
cost because the cost was very
expensive do you know how to install the sensors
even if you wanted to you don't have them
money but now because of advances and
the technology we can deploy low-cost
sensors everywhere to measure everything
right and so now it's so this you know
we are now in a position to kind of
address this issue that's been around
for a while so when I will be there
whenever you hear people talk about
water right we listen and maybe maybe
we'll feel sorry for a minute and then
we'll continue you know doing exactly
what we've been doing for years right
people and in developing reagents they
are they're fighting for space
necessary every single day right it's
it's they came up we cannot expect them
to go like sophisticated artificial
intelligence algorithms or may be set up
internet of things infrastructure to
solve their own problem because well
they're fighting for water every day so
the responsibility is on us and even
more so because we're privileged enough
to understand the nature of the scale of
this problem right so we should we
should aspire to build these intelligent
machines that can solve this problem
right these artificial intelligence has
has all the answers so we just need to
use it in the right way to solve this
water crisis.
 |
| AI solving water crisis. |
as a system right there are many many
moving parts in play here right you need
to acquire water you need to treat it
store it transport it to distribute it and
then continue must look like us I'll end
up using it now this whole system
is riddled with inefficiencies it's and
and as demand for water increases we
tend to put more stress on each of these
individual walks now let's dig a little
deeper shall we around % of the earth is
water right that's a lot of water
but and a half percent of that as
oceans so that's not usable that's not
for human use so all of the remaining
three and a half percent seventy percents glaciers and ice so if you if you
keep going down this funnel right you'll
realize that only less than % it's
available for human use and even within
that % is for agriculture % is used
by industrial supply systems so in the
end a teeny tiny . percent of all
available water is for us the consumers
like you and I can consume right
that's a very tiny fraction and and
that's why the cost of water is rapidly
increasing now if you look at silica
the report published over the last three decades if you look at the consumer
price trends and how-how the costs are
varied for like water natural gas
electricity you'll see that the cost of
water has risen by four hundred percent
right and that's more than
electricity and natural gas combined and
that's what and it's it will continue to
get worse right
and so the scarcity factor is gonna just
continue to let worse here over here now
let's see like how did we get here right
it's one of the main reasons is that our
the system is filled with inefficiencies
like leakage and we waste water and the
pipes that are aging pipes and there are
many more reasons right
all these things have contributed
and now we are in this situation where
the cost of war is rapidly increasing
and there's no drinking water available
for everyone on earth and if you talk
about leakage around two points one
trillion gallons of clean treated water
it's just lost every year in u.s.
/cdn.vox-cdn.com/uploads/chorus_image/image/61951161/Intel_ArticleIllustration_2_1.1540839268.jpg) |
| sustainable resources using AI |
right it's due to leaks faulty meters
broken water mains and this can actually
feed tens of millions of people for an
entire year right that's that's a lot of
water to waste and this fact is this
activated by the fact that the average
age of pipes when the u.s. in the US and
Canada is around years which means
there are some pipes that are like
hundred years old at just a couple of
weeks ago up in Berkeley, there was a
broken water main like just thousands of
gallons just too wasted and water supply
was shut out for for many people and the
the reason this happened was there was a
rushing leak in a cast-iron pipe was
installed in that's years old
it's crazy and the problem is this is
pervasive across the way it's not
an easy thing to fix because it's
extremely expensive right just to fix
the aging pipes in the US alone it takes
billion dollars now this is one of
the most developed countries in the
world imagine what's happening and some
of the less developed countries, it's
it's enormous the problem is huge and so
one of the ways to solve
this problem is we can do smart things
with data right
if you cannot replace all the hardware
if you cannot replace all the pipes
why don't we monitor it and be smart
about it and that's where that's where
data comes into the picture we currently
live in a data-driven economy where huge
industries are completely changing the
ways they work using data right for
example let us take manufacturing just a
a few decades ago people like humans used
to stand in assembly lines and do them
jobs and now it's completely like robots
and automation has completely taken over
this industry same thing is true with a
the supply chain for example all right
instead of humans trying to figure out
with their spreadsheets how do we know
route and plan this whole thing they are
using data analytics
to optimally are out in the schedule
everything so that they're way more
efficient it's much faster and they're
just way more comfortable as well right
and so water is the one is one
such industry all right if you look at
the trend right now to take the balls
over time all right back in the s
people were using spreadsheets and
manual tools right and then people moved
on from software is basically
something that's installed on your
the computer is no cloud right and then
people move to cloud-based solutions so
that person any people anywhere can
access the same solution from from
multiple geographical locations and then
they are now living in an AI turbid
a world where everything is powered by
data and artificial intelligence all
right so the entities that refuse to
adapt are just going to perish because
the world now is a lot less tolerant to
encompass right then it was said in
so data is the answer right so how do we
how do we use data what is data anyway
right number one is we just need to do like
two things efficiently like collect data
efficiently and do smart things with it
all right looks pretty straightforward
what's the big problem here well let's
see all right let's say let's say
there's a pipe that's installed like
lets like a hundred miles away from
where you live right and you want to
measure how much water is flowing
through it and so what you do is you
install a nice meter and then it's going
to measure the water now the problem is
for you to know what's happening that
five you need to send a guy who will go
there and look at the readings all right
now imagine they're like every guy's
responsible for like hundreds and
hundreds of these like thousands of
these meters right so obviously you
cannot hope to send a guy every single
the hour which is not gonna happen so and
that's why they take these readings once
every two months so what's gonna happen
between the two successive readings
right anything all the data is just
being lost because it was just not
looking at it right and that's where the
efficiency part comes into the picture how
can we continuously monitor water
everywhere without the kind of without
allocating a lot of manpower to it and
number two is how do we do smart things
with it which means if you collect all
the data what
of doing it if-if you're not extracting
meaningful insights and altering your
behavior so that's where the
smartness things now consider picture so
let's talk a little bit about data
the collection now to collect data
efficiently we do avoid sending like
like a human every time to get those
readings you can use Internet-connected
wireless sensors so what these sensors
can do many things like measure
temperature pressure vibration water
flow soil moisture you can you name it
and these sensors can do it and these
sensors are wirelessly connected to the
Internet so we can just stream data to
the cloud and you can just access it
without signing a human alright and this
this ecosystem of intimate connected
sensors are called the Internet of Things I
don't know if you for the bus I'm sure
you've heard of it like the thing is a
sensor so basically an ecosystem of
sensors I can measure these vital things
that is ecosystem we're heading towards
and that's how you collect data very
efficiently and number two is basically
a lot of water data is already available
there are billions of sensors billions
of sensors and mirrors that are already
collecting data it's all we need to do
some of them are connected to the
Internet most of them are not and that's
where it's why we gotta figure out how
can we get that data so that we can we
can do something with it and thing
is we need that data because of you
should aspire to live in a world where
decisions are driven by data right human
intuition is grossly insufficient when
it comes to decision making right
because we are dealing with like
exabytes of data over here which means
humans cannot just use this guesswork to
solve the problems right and that's
where efficiently understanding data
becomes very very handy and managing
water effectively has a lot of benefits
number one you can cut the usage right
under two we can reduce the risk of
the shortage you can reduce avoid expensively
operations and maintenance so basically
being smart with data has a lot of
benefits especially to water right and
we need to aspire to live this world
driven by data because without data
it was just another person with an
opinion right and that's been happening
a lot right that's why we need to move
away from it in the past people used to
use that experience or guesswork say
that hey you know I've seen this might
look familiar let me let me use my
intuition to do something with it that's
that's usually very very suboptimal now
how does it work in practice
number one is we just we need to collect
data continuously right we need to
collect continuously we need to analyze
the data with algorithms and then need
to take real-time decisions now
collecting data continuously right-that's is that's being done using
Internet of Things which means that all
these sensors that are installed they
are they're gathering data continuously
in streaming to the cloud all right and
let us talk a little bit about
artificial intelligence artificial
intelligence is the field of study that
deals with creating machines that can
can function like humans right that can
perform tasks that require human
intelligence right and a lot of the
research in this field is it's based on
emulating the learning process for human
brain human brain is amazing it's like
the best form of AI artificial
intelligence because it can learn
with effortlessly and it almost
happens subconsciously you don't you
need to try and you know chairs a
chair right it's not so easy for
machines to do it because of machines just
need a lot of data to understand even
simplest things and so artificial
intelligence and there's a paradigm
within artificial intelligence called
deep learning that especially focuses on
on this that they create models based on
human brain so that we can emulate them
learning the process and be really efficient
with it right and so basically you
measure learn and you infer right that's
if you keep this loop in any system you
can be very very efficient all right
let's take an example now let's say you
want to monitor water usage by utilizing
flow sensors and residential
buildings now these flow sensors are
retrofit which means that you can just
they just put it on top of a pipe and i
start measuring the water flow and it'll
start streaming the data to the cloud
now in this case by efficiently
measuring the flow of water right you
can be you can do gonna be way more
efficient but just the act of measuring
it and looking at it will alter your
behavior because if you don't
know how what is being wasted you won't
do anything about it but if you know
hey maybe I'm you know using way more
water for my lawn or this there's a
a lot of leaks and in the building I need
to do something about it just by looking
at the data, I will alter your behavior
and that's why we need to monitor water
usage and then understand how can
we can be more efficient with it right
and so there are many many other
problems you can solve as well right
there can detect leakage can track
trends you can identify anomalies can
forecast demand you can predict the
events and you can do a lot more
thanks, I depending on where you are if
you are dueling dealing with buildings
you can do these things you can do the
same for irrigation systems for
utilities industrial supply systems and
so just the act of measuring it will
make you improve it there's a very
a famous saying by Peter Drucker that goes
like if we cannot measure
it cannot improve it right so mature
measuring is an important part measuring
efficiently is even more important and
then doing smart things with it is how
you make it more efficient now how can
we right how can we contribute what can
we do right here and when water is not a
the new problem right what it's been around
for a while I mean when you hear
people talk about hey waters leaking
water is getting wasted people to don't
have access to clean drinking water if
you look around the world like
million people don't have access to
clean drinking water that's a lot of
people it's and so what's uh what's
what's changed now it's been around for
a while why now the problem is or the
the solution is now we now have the tools to
solve the problem like years ago you
you were not in a position to kind of
install these high-end sensors at low
cost because the cost was very
expensive do you know how to install the sensors
even if you wanted to you don't have them
money but now because of advances and
the technology we can deploy low-cost
sensors everywhere to measure everything
right and so now it's so this you know
we are now in a position to kind of
address this issue that's been around
for a while so when I will be there
whenever you hear people talk about
water right we listen and maybe maybe
we'll feel sorry for a minute and then
we'll continue you know doing exactly
what we've been doing for years right
people and in developing reagents they
are they're fighting for space
necessary every single day right it's
it's they came up we cannot expect them
to go like sophisticated artificial
intelligence algorithms or may be set up
internet of things infrastructure to
solve their own problem because well
they're fighting for water every day so
the responsibility is on us and even
more so because we're privileged enough
to understand the nature of the scale of
this problem right so we should we
should aspire to build these intelligent
machines that can solve this problem
right these artificial intelligence has
has all the answers so we just need to
use it in the right way to solve this
water crisis.
