WEBVTT
Language: en-US

1
00:00:02.680 --> 00:00:06.199
<v Speaker 1>Welcome to the trend detection podcast
powered by Siemens.

2
00:00:15.159 --> 00:00:20.200
<v Speaker 2>The Sensi tool, obviously AI there's a
broad array of different tools out there for

3
00:00:20.200 --> 00:00:24.680
<v Speaker 2>general use, but Sensi's quite specific,
so I guess, tell us a little bit about the

4
00:00:24.680 --> 00:00:27.620
<v Speaker 2>journey from where you started and, and
how you got there.

5
00:00:28.280 --> 00:00:29.670
<v Speaker 3>It's been a fascinating journey, really.

6
00:00:29.760 --> 00:00:33.490
<v Speaker 3>Uh, we started about 10 years ago, and,
and for me it's crazy to think about that.

7
00:00:33.799 --> 00:00:35.459
<v Speaker 3>I didn't have any gray hairs.

8
00:00:36.340 --> 00:00:38.000
<v Speaker 3>Uh, I've seen a lot of changes since then.

9
00:00:38.529 --> 00:00:41.000
<v Speaker 3>We started as the founders in aerospace
and defense.

10
00:00:41.459 --> 00:00:46.380
<v Speaker 3>And we actually worked on these systems um
and we developed the analysis software to

11
00:00:46.380 --> 00:00:52.779
<v Speaker 3>help the UK MOD monitor its fleet of
helicopters, and, and the OEM to understand how

12
00:00:52.779 --> 00:00:58.659
<v Speaker 3>can we start to deliver not just a, Piece
of equipment and a maintenance package,

13
00:00:59.040 --> 00:01:03.040
<v Speaker 3>but how, but now that our customer who's
the RAF was saying, well, I don't want to

14
00:01:03.040 --> 00:01:03.680
<v Speaker 3>buy that anymore.

15
00:01:03.720 --> 00:01:07.930
<v Speaker 3>I want to buy power by the hour, which was
coined by Rolls-Royce in the 60s.

16
00:01:08.239 --> 00:01:11.419
<v Speaker 3>So this concept's actually been around in
aerospace and defense for a tremendous

17
00:01:11.419 --> 00:01:12.220
<v Speaker 3>amount of time.

18
00:01:12.739 --> 00:01:16.120
<v Speaker 3>And with that one customer saying, hey, I
don't want to buy this, buy it like this

19
00:01:16.120 --> 00:01:20.959
<v Speaker 3>anymore, I want to buy flight hours, and
it is now up to you as the OEM to give me

20
00:01:20.959 --> 00:01:21.419
<v Speaker 3>that.

21
00:01:21.959 --> 00:01:25.760
<v Speaker 3>And we saw that concept potentially making
its way into the industrial space.

22
00:01:26.980 --> 00:01:31.300
<v Speaker 3>To give you an idea of how much it cost to
analyze these very complex machines, the

23
00:01:31.300 --> 00:01:35.260
<v Speaker 3>helicopters are essentially a, a machine
that, that hates itself, it just wants to

24
00:01:35.260 --> 00:01:37.290
<v Speaker 3>vibrate itself apart all the time.

25
00:01:37.660 --> 00:01:42.050
<v Speaker 3>Uh, having spent some time in that
industry, I've not been on a helicopter, and I'm,

26
00:01:42.099 --> 00:01:46.739
<v Speaker 3>I'm not tremendously interested in going
on one, because they're, you know, they're

27
00:01:46.739 --> 00:01:48.050
<v Speaker 3>difficult, uh.

28
00:01:49.099 --> 00:01:50.440
<v Speaker 3>All of this data.

29
00:01:51.699 --> 00:01:57.739
<v Speaker 3>For only about 300 machines, required
about 50 million pounds of investment every

30
00:01:57.739 --> 00:01:59.680
<v Speaker 3>year to analyze and understand.

31
00:02:00.059 --> 00:02:05.339
<v Speaker 3>And so it's a whole building, literally,
or a whole floor of expert reliability

32
00:02:05.339 --> 00:02:08.660
<v Speaker 3>consultants, looking at these things, OK,
what's the gearbox doing, how many metal

33
00:02:08.660 --> 00:02:12.460
<v Speaker 3>shards are detected, who is flying it at
that time, are they, are they a

34
00:02:12.460 --> 00:02:14.339
<v Speaker 3>particularly bad user of these machines?

35
00:02:15.320 --> 00:02:16.940
<v Speaker 3>And we saw this as an opportunity.

36
00:02:17.789 --> 00:02:22.259
<v Speaker 3>Particularly with IOT and cloud
processing, to take that concept, and to build a

37
00:02:22.259 --> 00:02:26.589
<v Speaker 3>system that would automate the role of the
reliability engineer, not to replace

38
00:02:26.589 --> 00:02:31.550
<v Speaker 3>them, but to augment their expertise, so
that you're not having to spend 50 million

39
00:02:31.550 --> 00:02:34.990
<v Speaker 3>pounds per year for 300 machines, because
when you take that to the industrial

40
00:02:34.990 --> 00:02:37.979
<v Speaker 3>space, we, we all know that it's not
scalable.

41
00:02:38.350 --> 00:02:42.550
<v Speaker 3>So we want to develop this scalable system
that can answer a problem right now of

42
00:02:42.550 --> 00:02:44.889
<v Speaker 3>how do I improve the efficiency of my
machinery.

43
00:02:45.470 --> 00:02:49.350
<v Speaker 3>But also answers this question that we
feel is coming still, and it's taking a

44
00:02:49.350 --> 00:02:54.110
<v Speaker 3>little bit longer than we thought, of how
does industry shift to instead of buying a

45
00:02:54.110 --> 00:02:59.350
<v Speaker 3>machine and a maintenance package, how can
we start to enable other organizations to

46
00:02:59.350 --> 00:03:01.380
<v Speaker 3>deliver capability.

47
00:03:03.699 --> 00:03:06.660
<v Speaker 4>Yes, so they don't need to have that team,
that whole floor full of specialist

48
00:03:06.660 --> 00:03:07.539
<v Speaker 4>engineers, they have a tool.

49
00:03:07.740 --> 00:03:11.479
<v Speaker 3>No, and if you're selling something as a
service, so maybe you're selling welds per

50
00:03:11.479 --> 00:03:15.979
<v Speaker 3>hour, or maybe you're selling rivets per
hour, you know, whatever it is, predictive

51
00:03:15.979 --> 00:03:17.899
<v Speaker 3>maintenance is an essential component of
that.

52
00:03:18.100 --> 00:03:21.940
<v Speaker 3>You cannot do that as an OEM without a
deep understanding of what is your machinery

53
00:03:21.940 --> 00:03:25.740
<v Speaker 3>doing at the customer site, how are they
using it, how are they treating it, how do

54
00:03:25.740 --> 00:03:29.619
<v Speaker 3>we need to change ourselves as an
organization to provide that service.

55
00:03:29.869 --> 00:03:32.220
<v Speaker 3>Everything comes down to the service.

56
00:03:32.789 --> 00:03:35.789
<v Speaker 3>And we wanted to build a technology that
would be ready for that when that time

57
00:03:35.789 --> 00:03:36.110
<v Speaker 3>comes.

58
00:03:36.880 --> 00:03:42.610
<v Speaker 2>So to deploy a tool like Sensei, uh, do I
need, does everything need to be in the

59
00:03:42.610 --> 00:03:43.119
<v Speaker 2>one brand?

60
00:03:43.160 --> 00:03:45.880
<v Speaker 2>Is it a Siemens solution that only works
with Siemens products?

61
00:03:45.960 --> 00:03:46.809
<v Speaker 2>What, how does that work?

62
00:03:46.960 --> 00:03:47.559
<v Speaker 3>It's a good question.

63
00:03:47.699 --> 00:03:51.199
<v Speaker 3>Um, so we were not originally Siemens, and
maybe I skipped over that part a little

64
00:03:51.199 --> 00:03:51.369
<v Speaker 3>bit.

65
00:03:51.639 --> 00:03:55.199
<v Speaker 3>So we, we developed this technology and we
were Sensii, and then in 20.

66
00:03:55.339 --> 00:04:00.570
<v Speaker 3>22, we were acquired by Siemens, and one
of our core tenants has always been, we are

67
00:04:00.570 --> 00:04:06.369
<v Speaker 3>agnostic, uh, because every organization
has their own history of solutions that

68
00:04:06.369 --> 00:04:10.330
<v Speaker 3>they've chosen and even within the same
facility, they might have these different

69
00:04:10.330 --> 00:04:10.830
<v Speaker 3>solutions.

70
00:04:11.210 --> 00:04:14.089
<v Speaker 3>And it would be very ignorant of us to
come in and say, no, you need to get rid of

71
00:04:14.089 --> 00:04:16.809
<v Speaker 3>all of that, uh, and, and you can only buy
from Siemens.

72
00:04:17.579 --> 00:04:20.299
<v Speaker 3>Well, Instead, what we want to do is take
this approach that, OK, you've made some

73
00:04:20.299 --> 00:04:24.619
<v Speaker 3>sensible decisions, and you've invested a
lot of money, we don't want to throw that

74
00:04:24.619 --> 00:04:28.140
<v Speaker 3>away, we want to start with what you have
and build up from there.

75
00:04:28.500 --> 00:04:34.059
<v Speaker 3>And maybe what you want out of this
doesn't match with what we need, but we will

76
00:04:34.059 --> 00:04:38.880
<v Speaker 3>choose and help you select the best of
breed, and that might not always be Siemens,

77
00:04:38.940 --> 00:04:43.100
<v Speaker 3>and this is a sacrilegious thing to say,
maybe, uh, maybe this is the last webcast

78
00:04:43.100 --> 00:04:43.839
<v Speaker 3>I'll do.

79
00:04:44.619 --> 00:04:49.619
<v Speaker 3>But, but the idea for us is respect the
customer's decision, respect the investments

80
00:04:49.619 --> 00:04:53.660
<v Speaker 3>that they have made, and leverage that to
the maximum of our abilities.

81
00:04:54.929 --> 00:04:58.829
<v Speaker 2>So The system pulls data from whatever.

82
00:04:59.600 --> 00:05:02.679
<v Speaker 2>PLC, SCAA, whatever database you've

83
00:05:02.679 --> 00:05:06.670
<v Speaker 3>got, we have vibration sensors, there's so
many different ways, we had one customer,

84
00:05:06.720 --> 00:05:12.220
<v Speaker 3>a very large multinational automotive
organization, they sent us an email every hour

85
00:05:12.220 --> 00:05:13.540
<v Speaker 3>and they had a CSV file.

86
00:05:14.549 --> 00:05:17.619
<v Speaker 3>It's as basic as that, it can work like
that, and then you build up on a more

87
00:05:17.619 --> 00:05:18.760
<v Speaker 3>sophisticated integration.

88
00:05:19.140 --> 00:05:19.440
<v Speaker 3>So it's

89
00:05:19.440 --> 00:05:21.000
<v Speaker 4>time series data, it can be IOT
essentially,

90
00:05:21.839 --> 00:05:25.600
<v Speaker 3>as long as it's time series data that we
can relate back to the condition of this

91
00:05:25.600 --> 00:05:29.359
<v Speaker 3>machine, that's what we're interested in,
how you deliver that to us, that's up to

92
00:05:29.359 --> 00:05:29.820
<v Speaker 3>you.

93
00:05:30.299 --> 00:05:34.429
<v Speaker 2>So you mentioned a large automotive
business, what other industries is this being

94
00:05:34.429 --> 00:05:35.130
<v Speaker 2>deployed in?

95
00:05:35.559 --> 00:05:38.799
<v Speaker 3>Literally everything you can imagine from
hydroelectric power plants to confectiony

96
00:05:38.799 --> 00:05:44.700
<v Speaker 3>manufacturing to, To anything, um, we
cover, we cover uranium mining.

97
00:05:46.679 --> 00:05:51.880
<v Speaker 3>Like I said, chocolate bars, ice cream,
cars, you, you name it, we probably look at

98
00:05:51.880 --> 00:05:55.600
<v Speaker 3>it, and that's fascinating for me as well
as a founder to say, like, my favorite

99
00:05:55.600 --> 00:06:00.839
<v Speaker 3>part of this role is really, I get to go
into these organizations and see how this

100
00:06:00.839 --> 00:06:01.440
<v Speaker 3>stuff is made.

101
00:06:01.720 --> 00:06:08.660
<v Speaker 3>One of my favorite, Programs as a, as a,
as a, I say as a teenager, um, was the

102
00:06:08.660 --> 00:06:12.459
<v Speaker 3>how it's Made series, and, and you go into
the factory and these people talk with

103
00:06:12.459 --> 00:06:16.950
<v Speaker 3>passion about something that they've
invested their whole career and their lives in,

104
00:06:17.260 --> 00:06:18.260
<v Speaker 3>and I find that fascinating.

105
00:06:19.040 --> 00:06:20.309
<v Speaker 2>So still watch them on YouTube by the way,

106
00:06:20.420 --> 00:06:22.459
<v Speaker 3>yeah, yeah, yeah, but but that's literally
my life now.

107
00:06:22.500 --> 00:06:25.779
<v Speaker 3>I go in and I get to meet these people,
and I get to ask all kinds of awkward

108
00:06:25.779 --> 00:06:28.890
<v Speaker 3>questions and, you know, try not to press
the big red buttons that everywhere, um,

109
00:06:28.899 --> 00:06:32.660
<v Speaker 3>it's quite compelling, but it's
fascinating to learn about this stuff and then.

110
00:06:33.369 --> 00:06:37.450
<v Speaker 3>To see how we can improve that and how
working with partners we can integrate these

111
00:06:37.450 --> 00:06:41.859
<v Speaker 3>systems, integrate more and more of this
information and give, give our customers

112
00:06:41.859 --> 00:06:46.089
<v Speaker 3>something really powerful where they can,
they can spend time on the things that

113
00:06:46.089 --> 00:06:47.619
<v Speaker 3>they don't want to spend time on.

114
00:06:47.890 --> 00:06:48.910
<v Speaker 3>That, that's key to us.

115
00:06:49.089 --> 00:06:53.329
<v Speaker 3>How can you focus on the things that are
really interesting, that a human is good

116
00:06:53.329 --> 00:06:56.019
<v Speaker 3>at, which is ultimately weird problems.

117
00:06:56.450 --> 00:06:57.209
<v Speaker 3>Condition monitoring.

118
00:06:57.619 --> 00:06:58.660
<v Speaker 3>is not a new problem.

119
00:06:58.820 --> 00:06:59.899
<v Speaker 3>It's very well understood.

120
00:07:00.179 --> 00:07:02.309
<v Speaker 3>It's a very structured discipline.

121
00:07:02.739 --> 00:07:06.649
<v Speaker 4>Something I've noticed about Sensor that I
really like is um a lot of the solutions

122
00:07:06.649 --> 00:07:10.579
<v Speaker 4>I can go buy off the shelf today, I spoke
to a specific industry or a specific use

123
00:07:10.579 --> 00:07:14.820
<v Speaker 4>case, whereas Sens is not sort of agnostic
and use case as well, you know, you

124
00:07:14.820 --> 00:07:18.450
<v Speaker 4>pointed at the data, it doesn't have to be
a pump or something like that, right?

125
00:07:18.700 --> 00:07:22.179
<v Speaker 3>And people don't like hearing that people
like hearing that, OK, my machine is super

126
00:07:22.179 --> 00:07:23.640
<v Speaker 3>special, my industry is super special.

127
00:07:24.269 --> 00:07:27.000
<v Speaker 3>Ultimately, actually, um, condition
monitoring is condition monitoring.

128
00:07:27.359 --> 00:07:32.920
<v Speaker 3>And whilst we focused predominantly on the
heavy industries, automotive, er,

129
00:07:33.000 --> 00:07:37.720
<v Speaker 3>pharmaceutical and food and beverage
manufacturing, It can really apply to anything

130
00:07:38.019 --> 00:07:42.739
<v Speaker 3>because machinery and how you deal with it
and how you look at reliability is, is a

131
00:07:42.739 --> 00:07:45.950
<v Speaker 3>core discipline, no matter which industry
you look at.

132
00:07:46.920 --> 00:07:48.730
<v Speaker 2>So let's talk about implementation then.

133
00:07:48.839 --> 00:07:52.859
<v Speaker 2>So is this something I buy, plug it in,
and it's working?

134
00:07:53.359 --> 00:07:53.880
<v Speaker 2>How does it work?

135
00:07:54.000 --> 00:07:55.040
<v Speaker 2>What's the rollout of the plug?

136
00:07:56.109 --> 00:08:00.429
<v Speaker 3>So the ultimate dream, and when we founded
it, we had this idea that a maintenance

137
00:08:00.429 --> 00:08:03.970
<v Speaker 3>manager can come along and they can type
in the credit card details, and here's my

138
00:08:03.970 --> 00:08:06.850
<v Speaker 3>er database URL and it will just work
magically.

139
00:08:07.190 --> 00:08:09.820
<v Speaker 3>And that's kind of the vision that we
still have in mind of that's where we'd like

140
00:08:09.820 --> 00:08:13.190
<v Speaker 3>to get to, but the reality is that it's
much more complex than that.

141
00:08:13.429 --> 00:08:19.070
<v Speaker 3>And that's why it's imperative uh for us
to rely on partners to perform that

142
00:08:19.070 --> 00:08:23.929
<v Speaker 3>integration work, that integration, that
connectivity, none of this works without

143
00:08:23.940 --> 00:08:24.250
<v Speaker 3>that.

144
00:08:24.579 --> 00:08:27.959
<v Speaker 3>And that's something that we learned very
much at Sensai, that as a start-up, we

145
00:08:27.959 --> 00:08:29.279
<v Speaker 3>didn't answer that question.

146
00:08:29.540 --> 00:08:33.940
<v Speaker 3>As Siemens and with our partners, we can
do that, and, and I'm tremendously excited

147
00:08:33.940 --> 00:08:37.239
<v Speaker 3>about the potential because it solves our
biggest problem.

148
00:08:37.739 --> 00:08:40.859
<v Speaker 4>Yes, if I circle back, I talked about
productivity, you know, we've got a.

149
00:08:41.278 --> 00:08:43.369
<v Speaker 4>A bunch of manufacturers, they want to
apply technology.

150
00:08:43.448 --> 00:08:45.528
<v Speaker 4>They don't necessarily have a big powerful
workforce.

151
00:08:45.609 --> 00:08:48.338
<v Speaker 4>They don't have a lot of like they might
have one or two skilled individuals.

152
00:08:48.528 --> 00:08:49.768
<v Speaker 4>They've got new staff coming in.

153
00:08:49.809 --> 00:08:50.828
<v Speaker 4>They're adding a shift.

154
00:08:51.278 --> 00:08:52.068
<v Speaker 4>Like they're trying to scale.

155
00:08:52.489 --> 00:08:56.169
<v Speaker 4>How do they scale that knowledge and
skills, you know, a tool like Sensor allows

156
00:08:56.169 --> 00:08:58.968
<v Speaker 4>them to sort of leverage those that
existing knowledge, you know.

157
00:08:59.049 --> 00:09:02.049
<v Speaker 4>So there's a bit of a change management
piece there, and it's like you're saying to

158
00:09:02.049 --> 00:09:04.129
<v Speaker 4>them like it's not here to replace you,
it's here to help you.

159
00:09:04.369 --> 00:09:07.028
<v Speaker 4>It's ready to add value to your business
and like, you know, really.

160
00:09:07.195 --> 00:09:10.304
<v Speaker 4>Make better use of your time, you know,
and, and aerospace

161
00:09:10.304 --> 00:09:12.585
<v Speaker 3>and defense term, uh, it's a force
multiplier.

162
00:09:12.784 --> 00:09:16.784
<v Speaker 3>It's, it's, you have this thing, and now
rather than monitoring the 20 critical

163
00:09:16.784 --> 00:09:20.565
<v Speaker 3>machines, you can monitor 2000, and, and
that's a wonderful story.

164
00:09:21.145 --> 00:09:25.335
<v Speaker 2>So if we think of, you know, we all know
about the tools like chat GPT, Copilot,

165
00:09:25.385 --> 00:09:29.224
<v Speaker 2>those sort of things, um, does it work in
the same way as it conversational?

166
00:09:29.265 --> 00:09:33.005
<v Speaker 2>Can you talk to your data in real, you
know, like human language?

167
00:09:33.760 --> 00:09:40.219
<v Speaker 3>So that's a chat GBT large language models
has been wonderful for us in that it

168
00:09:40.479 --> 00:09:44.280
<v Speaker 3>answers, or, or it, it's a missing part of
the puzzle that we always needed.

169
00:09:44.840 --> 00:09:49.559
<v Speaker 3>But our technology and our AI is, we, you
know, we started this 10 years ago when

170
00:09:49.559 --> 00:09:51.070
<v Speaker 3>that kind of stuff just wasn't available.

171
00:09:52.119 --> 00:09:55.719
<v Speaker 3>And so we spent a lot of time
understanding how do we codify condition monitoring,

172
00:09:55.840 --> 00:09:57.520
<v Speaker 3>how do we understand the metadata.

173
00:09:57.760 --> 00:10:00.940
<v Speaker 3>We developed a lot, a lot of patterns, a
lot of technologies specifically around,

174
00:10:01.239 --> 00:10:06.320
<v Speaker 3>well, how do we teach a machine to
understand what machine data looks like and try

175
00:10:06.320 --> 00:10:09.580
<v Speaker 3>to do that automatically, because that's
the only way that you can achieve scale.

176
00:10:10.359 --> 00:10:15.840
<v Speaker 3>Part of our issue then was how do we
communicate this to humans in a really, Easy

177
00:10:15.840 --> 00:10:21.289
<v Speaker 3>way, and that's a massive challenge, one
we've always struggled with, and we kind of

178
00:10:21.289 --> 00:10:25.690
<v Speaker 3>deferred that solving that problem, till
later down the line, we thought, well,

179
00:10:25.929 --> 00:10:29.099
<v Speaker 3>we're, we're gonna have to tackle this
problem, we have some scripted messaging that

180
00:10:29.099 --> 00:10:33.150
<v Speaker 3>we can put in and we can make it smart,
but it was still very difficult, and then,

181
00:10:34.049 --> 00:10:38.210
<v Speaker 3>Large language models came along, and we
fully integrated that, and we are the first

182
00:10:38.210 --> 00:10:40.210
<v Speaker 3>Siemens product, actually, we released it

183
00:10:40.210 --> 00:10:43.119
<v Speaker 3>in February 2024.

184
00:10:43.609 --> 00:10:44.400
<v Speaker 3>That sounds awkward, isn't it?

185
00:10:44.479 --> 00:10:48.570
<v Speaker 3>We've released it in February 2024, and,
and we've passed all of Siemens legal and

186
00:10:48.570 --> 00:10:52.260
<v Speaker 3>commercial and all of that, to have a
fully integrated co-pilot that's aware of the

187
00:10:52.260 --> 00:10:56.450
<v Speaker 3>time series data, it's aware of the
maintenance manuals that you put in, it's aware

188
00:10:56.450 --> 00:10:59.210
<v Speaker 3>of the contextual information that you're
bringing in from your maintenance

189
00:10:59.210 --> 00:11:00.070
<v Speaker 3>management system.

190
00:11:00.880 --> 00:11:02.039
<v Speaker 3>And that is so powerful.

191
00:11:02.200 --> 00:11:02.760
<v Speaker 3>It's incredible.

192
00:11:03.080 --> 00:11:04.150
<v Speaker 2>I'll just pick up on what you said there.

193
00:11:04.239 --> 00:11:05.500
<v Speaker 2>So it's connecting with the manual.

194
00:11:06.159 --> 00:11:09.940
<v Speaker 2>So your machine or whatever you're
monitoring, it's actually taking data from that

195
00:11:10.239 --> 00:11:12.369
<v Speaker 2>and, and providing it to the user.

196
00:11:12.520 --> 00:11:12.729
<v Speaker 2>Yeah,

197
00:11:13.039 --> 00:11:17.330
<v Speaker 3>our philosophy is if the AI has access to
it, give it to the user.

198
00:11:17.950 --> 00:11:22.400
<v Speaker 3>Uh, my one of my co-founders, Rob, his,
his, his pitch was always, we measure your

199
00:11:22.400 --> 00:11:24.539
<v Speaker 3>success by how little time you spend in
the application.

200
00:11:25.090 --> 00:11:28.450
<v Speaker 3>We don't tend to say that anymore because
it's often misunderstood, you know, we

201
00:11:28.450 --> 00:11:29.169
<v Speaker 3>don't want you using it.

202
00:11:29.369 --> 00:11:33.409
<v Speaker 3>No, we do want you using it, but we don't
want you coming in trawling for

203
00:11:33.409 --> 00:11:34.750
<v Speaker 3>information and trawling for value.

204
00:11:35.010 --> 00:11:36.929
<v Speaker 3>If the AI knows it, tell

205
00:11:36.929 --> 00:11:39.890
<v Speaker 4>you, it's not a chore, it's just easy,
yeah, feeds the day to make

206
00:11:39.890 --> 00:11:40.909
<v Speaker 3>your, make your life easier.

207
00:11:42.450 --> 00:11:48.539
<v Speaker 2>So from a personal experience, I've been
looking at AI tools as a uh a cheat code, a

208
00:11:48.539 --> 00:11:52.750
<v Speaker 2>shortcut, you know, a tool to make things
faster and more efficient, um, but that

209
00:11:52.750 --> 00:11:55.789
<v Speaker 2>also brings challenges because there are,
there's inherent fear with it.

210
00:11:55.869 --> 00:11:59.109
<v Speaker 2>There's people don't understand which
tools do you use.

211
00:11:59.390 --> 00:12:02.820
<v Speaker 2>So I guess my question is, is, uh, what,
what are the barriers to implementing

212
00:12:02.820 --> 00:12:03.289
<v Speaker 2>something like Senso.

213
00:12:07.159 --> 00:12:10.219
<v Speaker 3>on the like on the on the implementation
and the data connectivity side.

214
00:12:10.440 --> 00:12:11.270
<v Speaker 4>It's a journey, right?

215
00:12:11.450 --> 00:12:13.960
<v Speaker 4>It's a journey like you said, you don't
just switch it on and walk away.

216
00:12:14.200 --> 00:12:17.479
<v Speaker 4>It's about, you know, bringing people
along with you, sort of explaining the value

217
00:12:17.479 --> 00:12:19.520
<v Speaker 4>to them, and then making, making them part
of the journey.

218
00:12:19.590 --> 00:12:22.640
<v Speaker 4>So they're really like engaged from the
beginning and all the way through.

219
00:12:22.960 --> 00:12:26.919
<v Speaker 4>So it's like, you know, the right support
and training and, and also use case

220
00:12:26.919 --> 00:12:29.419
<v Speaker 4>examples and teaching them how to use it,
supporting them as they go.

221
00:12:29.719 --> 00:12:33.000
<v Speaker 4>And then making them like part of training
it, you know, because what I don't know

222
00:12:33.000 --> 00:12:36.700
<v Speaker 4>what you picked up in there, but when,
when you have a anomaly detected and you

223
00:12:36.700 --> 00:12:40.799
<v Speaker 4>solve the problem, you tell Si how you
solved it, so that gets fed into the model

224
00:12:40.799 --> 00:12:42.799
<v Speaker 4>for next time to tell the, tell the person
down the track.

225
00:12:43.159 --> 00:12:45.719
<v Speaker 4>So you help them with how to give good
information into the system.

226
00:12:45.960 --> 00:12:49.679
<v Speaker 4>So we, we're working with them person to
person to really get the most out of the

227
00:12:49.679 --> 00:12:50.020
<v Speaker 4>tool.

228
00:12:50.580 --> 00:12:52.679
<v Speaker 3>You said something key actually, it's
about people.

229
00:12:53.419 --> 00:12:56.539
<v Speaker 3>And we talk about condition monitoring, we
talk about machines, we talk about

230
00:12:56.539 --> 00:13:01.179
<v Speaker 3>machine data and condition monitoring, but
ultimately, all of this technology and AI

231
00:13:01.619 --> 00:13:05.190
<v Speaker 3>is about saving time and effort for
people.

232
00:13:05.659 --> 00:13:10.820
<v Speaker 3>Do you, as an expert reliability engineer,
want to focus on vibration monitoring and

233
00:13:10.820 --> 00:13:16.380
<v Speaker 3>the 123 times harmonics and trawling
around in some old school system to look at sky

234
00:13:16.380 --> 00:13:17.090
<v Speaker 3>lines?

235
00:13:17.539 --> 00:13:19.700
<v Speaker 3>Our argument is probably not.

236
00:13:20.820 --> 00:13:26.150
<v Speaker 3>The AI can do that all of the time, 24/7
on tens of thousands of machines within a

237
00:13:26.150 --> 00:13:30.390
<v Speaker 3>single account, and it can tell you which
ones you should be paying attention to.

238
00:13:30.669 --> 00:13:34.830
<v Speaker 3>And out of those thousands, it might just
be 2 or 3, and you can use your expertise

239
00:13:34.830 --> 00:13:36.500
<v Speaker 3>then to really zero in on that.

240
00:13:36.940 --> 00:13:39.049
<v Speaker 4>You see that scale, that productivity
we're talking about.

241
00:13:39.869 --> 00:13:40.750
<v Speaker 4>So how, you know,

242
00:13:40.830 --> 00:13:41.710
<v Speaker 2>what's the measure of success?

243
00:13:41.830 --> 00:13:42.669
<v Speaker 2>How do you know it's working?

244
00:13:42.830 --> 00:13:44.409
<v Speaker 2>You, you talked about feedback.

245
00:13:46.010 --> 00:13:49.469
<v Speaker 4>So, so the tool itself, what are we
talking, return on investment, and that's part

246
00:13:49.469 --> 00:13:50.400
<v Speaker 4>of the story, right?

247
00:13:50.590 --> 00:13:50.669
<v Speaker 4>Yeah,

248
00:13:50.750 --> 00:13:53.849
<v Speaker 3>yeah, and that's part of the journey that
you have to take these customers on of

249
00:13:54.070 --> 00:13:55.469
<v Speaker 3>what actually matters to you.

250
00:13:55.549 --> 00:13:56.909
<v Speaker 3>Are you producing milk?

251
00:13:56.960 --> 00:13:58.390
<v Speaker 3>Are you digging up dirt?

252
00:13:58.460 --> 00:14:00.340
<v Speaker 3>Are you producing cars?

253
00:14:00.700 --> 00:14:04.409
<v Speaker 3>How do you, we, and this is a
conversation, and this is where the human element

254
00:14:04.409 --> 00:14:05.330
<v Speaker 3>comes back in.

255
00:14:05.869 --> 00:14:08.710
<v Speaker 3>How do you measure success and how can we
set up the AI?

256
00:14:09.429 --> 00:14:14.190
<v Speaker 3>So that it's as easy for you as possible
to measure that success and capture it.

257
00:14:14.549 --> 00:14:18.729
<v Speaker 3>So, OK, we found something cool on a, on a
gearbox, and it was slightly cracked.

258
00:14:19.419 --> 00:14:23.419
<v Speaker 3>It's OK to say, look, we uh, we fixed
that, wonderful, and that's where we see most

259
00:14:23.419 --> 00:14:24.419
<v Speaker 3>of the market looking at.

260
00:14:24.849 --> 00:14:27.510
<v Speaker 3>But actually what's more, much more
interesting is, what does that mean to

261
00:14:27.510 --> 00:14:27.909
<v Speaker 3>management?

262
00:14:28.219 --> 00:14:29.380
<v Speaker 3>What does that mean to the board level?

263
00:14:29.580 --> 00:14:33.580
<v Speaker 3>How do you integrate that, that data so
that you can say, yeah, I fixed the crack on

264
00:14:33.580 --> 00:14:36.900
<v Speaker 3>the gearbox, and this is how we try to
talk to reliability engineers as well, of,

265
00:14:37.070 --> 00:14:40.979
<v Speaker 3>we're not coming for your job, we're
making sure that you guys are displayed as the

266
00:14:40.979 --> 00:14:41.359
<v Speaker 3>hero.

267
00:14:42.140 --> 00:14:46.664
<v Speaker 3>You use, to find a crack in the gearbox,
and that meant that you saved 10,000 L of

268
00:14:46.664 --> 00:14:49.825
<v Speaker 3>water, you saved 5000 tons of carbon
emissions.

269
00:14:50.104 --> 00:14:56.354
<v Speaker 3>You also saved a $1 million contractual
penalty penalty on, on emissions, and we see

270
00:14:56.354 --> 00:15:00.354
<v Speaker 3>this all day every day, where we can say,
look, somebody used Sensai.

271
00:15:01.859 --> 00:15:07.179
<v Speaker 3>They fixed the problem, they added the KPI
they added the work event, which, you

272
00:15:07.179 --> 00:15:11.940
<v Speaker 3>know, helped Sensie to learn and build
from that, and now your name as the

273
00:15:11.940 --> 00:15:13.700
<v Speaker 3>maintenance reliability Eng engineer.

274
00:15:13.989 --> 00:15:18.020
<v Speaker 3>Is associated with that big capture where
you've saved a contractual penalty of a

275
00:15:18.020 --> 00:15:22.349
<v Speaker 3>million dollars, and previously, that was
a pretty hard story to convince, but as

276
00:15:22.349 --> 00:15:26.270
<v Speaker 3>sensei, it's about bringing that human
element into it.

277
00:15:26.359 --> 00:15:26.520
<v Speaker 3>And,

278
00:15:26.630 --> 00:15:26.710
<v Speaker 4>and

279
00:15:26.710 --> 00:15:26.869
<v Speaker 3>I'll

280
00:15:26.869 --> 00:15:30.030
<v Speaker 4>add to that, like if you've got, you know,
you've got your, uh, engineer who's been

281
00:15:30.030 --> 00:15:33.419
<v Speaker 4>there in a lifetime, you know, he knows
his plant or her plant, but they might be

282
00:15:33.419 --> 00:15:34.030
<v Speaker 4>misunderstood.

283
00:15:34.309 --> 00:15:36.090
<v Speaker 4>They might not get that point across very
often.

284
00:15:36.429 --> 00:15:40.400
<v Speaker 4>But using the data, using the SCSI results
and the calculation of the ROI, the

285
00:15:40.400 --> 00:15:44.599
<v Speaker 4>return on investment, suddenly they're a
hero, you know, um, and they can share that

286
00:15:44.599 --> 00:15:47.520
<v Speaker 4>knowledge with the rest of the team, and
they can really, um, make a difference to

287
00:15:47.520 --> 00:15:48.010
<v Speaker 4>the business.

288
00:15:48.159 --> 00:15:48.359
<v Speaker 4>And

289
00:15:48.359 --> 00:15:49.169
<v Speaker 3>they are heroes.

290
00:15:49.239 --> 00:15:53.429
<v Speaker 3>Like a lot of the, a lot of the difficulty
we have is people get nervous about AI

291
00:15:53.429 --> 00:15:55.419
<v Speaker 3>that they think, oh, it's gonna come and
take my job.

292
00:15:55.799 --> 00:15:59.200
<v Speaker 3>And, and maintainers are in a really
difficult position.

293
00:15:59.590 --> 00:16:03.239
<v Speaker 3>If they do their job well, then there
won't be any breakdowns.

294
00:16:03.479 --> 00:16:06.320
<v Speaker 3>Nobody notices, and then they say, oh we
don't need maintenance, we can get rid of

295
00:16:06.320 --> 00:16:06.479
<v Speaker 3>that.

296
00:16:06.630 --> 00:16:10.020
<v Speaker 3>That's the first thing to get deleted from
the accountancy budget.

297
00:16:10.960 --> 00:16:13.919
<v Speaker 3>But if they're, you know, struggling,
which they often are, because it's

298
00:16:13.919 --> 00:16:16.559
<v Speaker 3>underfunded, because people say, oh, we're
spending too much on maintenance, we

299
00:16:16.559 --> 00:16:17.200
<v Speaker 3>don't need that.

300
00:16:17.979 --> 00:16:22.659
<v Speaker 3>Then bad things happen, and then they get
screamed at, so they've, they are under

301
00:16:22.659 --> 00:16:26.020
<v Speaker 3>pressure from all sides, and a lot of the
work that we do and a lot of the process

302
00:16:26.020 --> 00:16:31.469
<v Speaker 3>and the methodologies and and the way that
we work is to try to say, OK, you did

303
00:16:31.469 --> 00:16:36.590
<v Speaker 3>this thing, you, you fixed the problem,
and now let's automate how we reflect that

304
00:16:36.590 --> 00:16:41.590
<v Speaker 3>into your organization, so that you as the
hero can really easily be identified, and

305
00:16:41.590 --> 00:16:45.989
<v Speaker 3>maybe you're not so good at taking your,
uh, you know, your, your capture and your

306
00:16:45.989 --> 00:16:47.950
<v Speaker 3>win, your fix, and relating

307
00:16:47.950 --> 00:16:50.609
<v Speaker 3>that to senior management, but we can
automate that.

308
00:16:51.359 --> 00:16:53.479
<v Speaker 2>I just want to pick you up, Nick, on, on a
comment you made.

309
00:16:53.760 --> 00:16:57.450
<v Speaker 2>You talked about an engineer that's been
there a million years, uh, and really that,

310
00:16:57.479 --> 00:17:01.159
<v Speaker 2>that points to an issue that everyone's
having at the moment, and that is a lot of

311
00:17:01.159 --> 00:17:05.229
<v Speaker 2>skilled workers are getting to that age
and stage and they're starting to retire and

312
00:17:05.229 --> 00:17:05.880
<v Speaker 2>you're losing that.

313
00:17:05.959 --> 00:17:06.319
<v Speaker 2>So

314
00:17:06.319 --> 00:17:08.680
<v Speaker 4>starting to retire, but also, like I was
talking about that productivity.

315
00:17:08.729 --> 00:17:11.709
<v Speaker 4>If you add a shift, how do you duplicate
that person, you know, or they go on

316
00:17:11.709 --> 00:17:12.489
<v Speaker 4>holiday, you know.

317
00:17:12.935 --> 00:17:14.175
<v Speaker 4>It's not just about when they retire.

318
00:17:14.244 --> 00:17:16.015
<v Speaker 4>It's, it's all facets of the business, you
know.

319
00:17:16.135 --> 00:17:20.295
<v Speaker 4>So having these tools to help gather the
information, intelligence, and the, and the

320
00:17:20.295 --> 00:17:23.895
<v Speaker 4>processes, and I, I, I think there's
always a challenge there to get some of those

321
00:17:23.895 --> 00:17:25.255
<v Speaker 4>guys to let go of that information.

322
00:17:25.375 --> 00:17:27.234
<v Speaker 4>But, you know, generally, they're pretty
good at that.

323
00:17:27.334 --> 00:17:28.675
<v Speaker 4>They love having the apprentice along.

324
00:17:28.974 --> 00:17:30.694
<v Speaker 4>So it's like, it's like a tool for the
apprentice.

325
00:17:30.775 --> 00:17:32.224
<v Speaker 4>So you've got to explain it that way to
them.

326
00:17:32.614 --> 00:17:34.425
<v Speaker 4>Now they're helping the young guy that's
coming up.

327
00:17:34.974 --> 00:17:37.375
<v Speaker 4>And, um, as soon as you explain that,
like, OK, got it.

328
00:17:37.704 --> 00:17:41.594
<v Speaker 2>So obviously since I was ingesting a lot
of data, vibration, all those different,

329
00:17:41.685 --> 00:17:42.015
<v Speaker 2>um.

330
00:17:42.410 --> 00:17:46.849
<v Speaker 2>Uh, feeds of information, time series
data, but how does it capture the knowledge of

331
00:17:46.849 --> 00:17:49.560
<v Speaker 2>somebody who's just got it up here and
knows the machine and

332
00:17:50.520 --> 00:17:50.810
<v Speaker 3>That's great.

333
00:17:53.920 --> 00:17:58.979
<v Speaker 4>I mean, so yeah, so that's about that's
again, I guess it's up to us engaging with

334
00:17:58.979 --> 00:18:02.640
<v Speaker 4>them to help them put really good data
back into to feed it, what, how to write, how

335
00:18:02.640 --> 00:18:05.119
<v Speaker 4>to write the use case up, you know, what
did they do?

336
00:18:05.239 --> 00:18:05.390
<v Speaker 4>What

337
00:18:05.390 --> 00:18:10.000
<v Speaker 3>did they do an acceptance that actually
like when we sometimes some customers, we

338
00:18:10.000 --> 00:18:12.579
<v Speaker 3>say, well, can we look at your maintenance
management system.

339
00:18:13.040 --> 00:18:17.300
<v Speaker 3>Because the way that we work and the way
that we treat assets, each asset is unique.

340
00:18:17.660 --> 00:18:20.900
<v Speaker 3>Yeah, this might be the same fan or the
same motor or whatever, but maybe it was

341
00:18:20.900 --> 00:18:24.619
<v Speaker 3>bolted in differently, maybe somebody
threw a spanner in it at one point and now

342
00:18:24.619 --> 00:18:25.140
<v Speaker 3>it's damaged.

343
00:18:25.339 --> 00:18:27.880
<v Speaker 3>There's a history to these things, and we
need to take that into account.

344
00:18:28.260 --> 00:18:28.270
<v Speaker 3>Hm.

345
00:18:28.795 --> 00:18:32.505
<v Speaker 3>In the same way that each human is unique,
each asset is unique, and yeah, at its

346
00:18:32.505 --> 00:18:36.964
<v Speaker 3>base, it's a, it's a motor, but actually
it has a unique history and a unique way,

347
00:18:37.035 --> 00:18:40.954
<v Speaker 3>and the, and the data that comes from that
is unique, so we need to learn that and

348
00:18:40.954 --> 00:18:41.744
<v Speaker 3>treat it uniquely.

349
00:18:41.915 --> 00:18:44.694
<v Speaker 3>We need to do that automatically in order
for this to be scalable.

350
00:18:45.275 --> 00:18:48.594
<v Speaker 3>And I've completely missed my point of
whatever I was gonna say, so we can edit this

351
00:18:48.594 --> 00:18:48.795
<v Speaker 3>one out.

352
00:18:49.714 --> 00:18:51.354
<v Speaker 3>So knowledge capture, um.

353
00:18:53.260 --> 00:18:59.939
<v Speaker 3>Large language models allow us to make
sure that we can relate this in a way that's

354
00:18:59.939 --> 00:19:03.189
<v Speaker 3>understandable by non-reliability
engineering professionals.

355
00:19:03.500 --> 00:19:07.400
<v Speaker 3>And so reliability, you know, have, have
traditionally struggled to communicate

356
00:19:07.819 --> 00:19:08.339
<v Speaker 3>their value.

357
00:19:08.500 --> 00:19:11.540
<v Speaker 3>Maintenance teams have struggled to
communicate their value.

358
00:19:12.060 --> 00:19:16.949
<v Speaker 3>Now using AI we can turn this into
something very easily understandable by somebody

359
00:19:16.949 --> 00:19:17.880
<v Speaker 3>in senior management.

360
00:19:18.250 --> 00:19:21.140
<v Speaker 3>And my theory, and, and what I've always
been telling the team is, right, when we

361
00:19:21.140 --> 00:19:22.359
<v Speaker 3>need to talk to senior management.

362
00:19:22.939 --> 00:19:27.859
<v Speaker 3>Uh, we've got to be, uh, thinking that
we're communicating with a seven year old,

363
00:19:27.900 --> 00:19:28.489
<v Speaker 3>8-year-old.

364
00:19:28.900 --> 00:19:32.380
<v Speaker 3>They don't have much time and they want a
picture and they want it very simply

365
00:19:32.380 --> 00:19:32.839
<v Speaker 3>explained.

366
00:19:33.140 --> 00:19:36.609
<v Speaker 3>And if you start going into frequencies
and harmonics and all these things, they're

367
00:19:36.609 --> 00:19:37.089
<v Speaker 3>switched off.

368
00:19:37.380 --> 00:19:41.410
<v Speaker 3>If you start talking to them about value
and business outcomes and money saved,

369
00:19:41.859 --> 00:19:42.280
<v Speaker 3>good.

370
00:19:43.099 --> 00:19:46.739
<v Speaker 2>So you talked about scale, and we, we're
talking about, um.

371
00:19:47.119 --> 00:19:50.790
<v Speaker 2>You know, knowledge capture from people
that are at that age and stage, onboarding

372
00:19:50.790 --> 00:19:54.329
<v Speaker 2>new teams and everything, um, but how
scalable actually is it?

373
00:19:54.400 --> 00:20:00.060
<v Speaker 2>You talk about a global organization in
automotive, I think, um, so what is scale?

374
00:20:01.630 --> 00:20:01.959
<v Speaker 2>Where does it start and where does it

375
00:20:01.959 --> 00:20:02.160
<v Speaker 3>stop?

376
00:20:02.339 --> 00:20:03.800
<v Speaker 3>Everyone has a different definition of it,
right?

377
00:20:04.069 --> 00:20:08.750
<v Speaker 3>Um, a lot of people with this technology
is a lot of people will say, oh, I can do

378
00:20:08.750 --> 00:20:10.459
<v Speaker 3>50 machines and I'm doing it at scale.

379
00:20:10.939 --> 00:20:14.930
<v Speaker 3>Uh, OK, cool, well done, but now try to
take it to 500 or 5000.

380
00:20:15.180 --> 00:20:19.579
<v Speaker 3>So we have real customers, like Bluescope
Steel, it's a public case study, who are

381
00:20:19.579 --> 00:20:23.969
<v Speaker 3>currently monitoring over 15,000 machines,
and a machine might be a collection of

382
00:20:23.969 --> 00:20:27.859
<v Speaker 3>assets, and that is millions of data
points, I don't know how many, it's many

383
00:20:27.859 --> 00:20:32.099
<v Speaker 3>millions, and that's being monitored in
real time and helping their maintainers to

384
00:20:32.099 --> 00:20:36.199
<v Speaker 3>focus on, Just what is actually what,
what's the thing that I need to focus on,

385
00:20:36.540 --> 00:20:39.020
<v Speaker 3>because I've got a huge list of scheduled
maintenance, I've got a huge list of

386
00:20:39.020 --> 00:20:42.380
<v Speaker 3>things I need to worry about, but what are
the things that are actually showing me

387
00:20:42.380 --> 00:20:42.760
<v Speaker 3>concern?

388
00:20:44.469 --> 00:20:49.400
<v Speaker 1>Please like and subscribe on all major
podcast channels and find out more about

389
00:20:49.400 --> 00:20:52.780
<v Speaker 1>Sensi predictive maintenance at
Siemens.com.
