WEBVTT
Language: en-US

1
00:00:01.840 --> 00:00:07.070
<v Speaker 1>Welcome to the trend detection podcast
powered by Siemens, and now your host Nel

2
00:00:07.070 --> 00:00:07.670
<v Speaker 1>Sullivan.

3
00:00:10.560 --> 00:00:14.510
<v Speaker 2>Hello and welcome to the latest episode of
the trend detection podcast.

4
00:00:15.170 --> 00:00:17.530
<v Speaker 2>I'm really pleased to be joined by my
colleague Dan

5
00:00:17.530 --> 00:00:17.770
<v Dan Reid>Reed.

6
00:00:17.889 --> 00:00:19.229
<v Dan Reid>Hi, thank you for having me.

7
00:00:19.489 --> 00:00:23.069
<v Speaker 2>Um, once again, actually, so you're one of
our regular guests now, so it's really

8
00:00:23.069 --> 00:00:23.729
<v Speaker 2>3rd

9
00:00:23.729 --> 00:00:24.739
<v Dan Reid>time, 3rd time too many.

10
00:00:25.959 --> 00:00:27.209
<v Speaker 2>3rd time lucky, I hope.

11
00:00:27.479 --> 00:00:32.250
<v Speaker 2>But today we're talking about, um, another
enhancement to our maintenance co-pilot,

12
00:00:32.349 --> 00:00:35.069
<v Speaker 2>um, within the Sensi Predict maintenance
platform, of course.

13
00:00:35.680 --> 00:00:36.840
<v Speaker 2>Um, so we're gonna dive into that.

14
00:00:36.889 --> 00:00:38.409
<v Speaker 2>That's Knowledge Builder, um.

15
00:00:38.900 --> 00:00:42.770
<v Speaker 2>And yeah, we're gonna go into that today,
but um I guess first of all, Dan, um,

16
00:00:42.779 --> 00:00:46.810
<v Speaker 2>let's sort of zoom out I guess because we
wanted to look at sort of the, you know,

17
00:00:46.979 --> 00:00:50.540
<v Speaker 2>the context to to all this, what, you
know, why we've developed it and we'll talk

18
00:00:50.540 --> 00:00:55.700
<v Speaker 2>about the actual functionality, but if we
just start with the, the big issue in

19
00:00:55.700 --> 00:01:00.279
<v Speaker 2>industry today, I talked to a lot of
people um around knowledge loss, um, in

20
00:01:00.290 --> 00:01:05.440
<v Speaker 2>industrial maintenance, um, and what
impact that's had on operational efficiency.

21
00:01:06.029 --> 00:01:10.059
<v Dan Reid>Yeah, so it's certainly, you know, we, we
talk to customers a lot and obviously a

22
00:01:10.059 --> 00:01:14.470
<v Dan Reid>lot of what we talk around is predictive
maintenance and downtime and, and related

23
00:01:14.470 --> 00:01:14.980
<v Dan Reid>topics.

24
00:01:15.349 --> 00:01:19.709
<v Dan Reid>But the other major topic which gets
brought up is, is knowledge and capturing it

25
00:01:19.709 --> 00:01:20.849
<v Dan Reid>and surfacing it.

26
00:01:21.220 --> 00:01:24.809
<v Dan Reid>Um, and uh there's some quite interesting
reports on this topic.

27
00:01:24.949 --> 00:01:30.279
<v Dan Reid>So, um, Deloitte published a report and it
said that 62% of manufacturers.

28
00:01:30.680 --> 00:01:35.559
<v Dan Reid>Uh, their biggest issue is, um, talent
recruitment and retention, and that's their

29
00:01:35.559 --> 00:01:36.110
<v Dan Reid>biggest issue.

30
00:01:36.160 --> 00:01:38.279
<v Dan Reid>It's not one of their biggest, it's, it's
the biggest.

31
00:01:38.639 --> 00:01:43.230
<v Dan Reid>So this is a problem that's very much
front of mind for manufacturing and our

32
00:01:43.230 --> 00:01:44.720
<v Dan Reid>customers in, in particular.

33
00:01:45.879 --> 00:01:49.849
<v Dan Reid>Um, and I think it stems from sort of 22
problems.

34
00:01:49.930 --> 00:01:53.040
<v Dan Reid>The first is there's, it's an aging
workforce.

35
00:01:53.370 --> 00:01:58.910
<v Dan Reid>So in that same Deloitte report, they
identified that 25% of manufacturing employees

36
00:01:59.129 --> 00:02:00.529
<v Dan Reid>are 55 or older.

37
00:02:00.690 --> 00:02:06.410
<v Dan Reid>So that's a big portion of, um, you know,
manufacturers, maintainers who are going

38
00:02:06.410 --> 00:02:08.229
<v Dan Reid>to be heading towards retirement.

39
00:02:08.779 --> 00:02:10.250
<v Dan Reid>And obviously through retirement.

40
00:02:10.710 --> 00:02:15.429
<v Dan Reid>These organizations are losing the skills,
the experience, all that knowledge, um,

41
00:02:15.509 --> 00:02:17.660
<v Dan Reid>from these experienced, uh, individuals.

42
00:02:18.229 --> 00:02:22.630
<v Dan Reid>And then on the other side of this
problem, over the last sort of two or three

43
00:02:22.630 --> 00:02:27.710
<v Dan Reid>decades, there's been a declining interest
in manufacturing, mechanical engineering,

44
00:02:27.750 --> 00:02:30.220
<v Dan Reid>and, you know, all of the related, uh,
topics.

45
00:02:30.550 --> 00:02:35.149
<v Dan Reid>And it means that in the job market, you
just don't have the people, um, who have

46
00:02:35.149 --> 00:02:39.029
<v Dan Reid>the skills, uh, to, to backfill these, um,
these positions.

47
00:02:39.630 --> 00:02:45.940
<v Dan Reid>So it does lead to quite a a troublesome
situation and um in terms of efficiency,

48
00:02:46.100 --> 00:02:49.940
<v Dan Reid>what it means is that you're having people
who are less skilled at maintenance or

49
00:02:49.940 --> 00:02:51.179
<v Dan Reid>less experienced.

50
00:02:51.539 --> 00:02:56.059
<v Dan Reid>Um, the maintenance obviously takes
longer, um, it's more prone to mistakes and that

51
00:02:56.059 --> 00:03:00.949
<v Dan Reid>will have a knock-on effect to, um, up
time, mean time to repair and just general

52
00:03:00.949 --> 00:03:03.600
<v Dan Reid>efficiency of the assets and, and the
sites.

53
00:03:04.539 --> 00:03:09.139
<v Speaker 2>And I guess if we're, because I mean, even
when I sort of started within the sense

54
00:03:09.139 --> 00:03:11.979
<v Speaker 2>our team that the talk was a lot about
unplanned downtime.

55
00:03:12.179 --> 00:03:14.729
<v Speaker 2>I mean, that used to be the focus and
maybe and it's still a very important metric,

56
00:03:14.770 --> 00:03:19.539
<v Speaker 2>but what we're saying is this is sort of
shifted, I guess in the past maybe 123

57
00:03:19.539 --> 00:03:21.259
<v Speaker 2>years towards, you know, listen.

58
00:03:21.350 --> 00:03:22.449
<v Speaker 2>Becoming a real issue now.

59
00:03:22.710 --> 00:03:23.910
<v Dan Reid>Yes, yeah, very much so.

60
00:03:24.070 --> 00:03:28.619
<v Dan Reid>So, you know, it's downtime and it's still
an issue and making sure that you're

61
00:03:28.619 --> 00:03:30.729
<v Dan Reid>doing precise maintenance is, is still an
issue.

62
00:03:31.190 --> 00:03:34.669
<v Dan Reid>But yeah, it's certainly how, how do you
go about capturing all of the knowledge,

63
00:03:34.830 --> 00:03:38.910
<v Dan Reid>um, especially as people are retiring, you
want to capture it before they, they

64
00:03:38.910 --> 00:03:43.869
<v Dan Reid>retire, and then making sure that that
knowledge is accessible and you get value

65
00:03:43.869 --> 00:03:44.229
<v Dan Reid>from it.

66
00:03:44.529 --> 00:03:48.940
<v Dan Reid>Um, and it's, it's a difficult problem,
um, and something that, you know, over the

67
00:03:48.940 --> 00:03:54.660
<v Dan Reid>last maybe 2 and a bit years Sens Ii has
been focused on, um, uh, and you know,

68
00:03:54.740 --> 00:03:58.020
<v Dan Reid>we're seeing great, uh, benefit from that
for our, for our customers.

69
00:03:58.970 --> 00:04:03.740
<v Speaker 2>And I, and I guess the, um, the, the
maintenance co-pilot to start with has been a

70
00:04:03.740 --> 00:04:05.059
<v Speaker 2>big step towards that.

71
00:04:05.139 --> 00:04:08.770
<v Speaker 2>So maybe for, for people maybe not 100%
aware, we could talk about sort of the

72
00:04:08.770 --> 00:04:13.059
<v Speaker 2>development of that over the past year,
year or so in general and where we are

73
00:04:13.059 --> 00:04:13.300
<v Speaker 2>today.

74
00:04:13.460 --> 00:04:13.759
<v Speaker 2>Yes,

75
00:04:13.979 --> 00:04:19.779
<v Dan Reid>so, um, this was, uh, I think we released
it probably a year ago at this point, um,

76
00:04:19.899 --> 00:04:22.019
<v Dan Reid>and there's sort of multiple elements of
it.

77
00:04:22.140 --> 00:04:27.059
<v Dan Reid>First of all, within Sensi now when we
raise an alert to, uh, a user.

78
00:04:27.480 --> 00:04:32.850
<v Dan Reid>Um, so that we believe, uh, an asset needs
attention or it needs maintenance.

79
00:04:33.209 --> 00:04:37.450
<v Dan Reid>Um, we obviously provide, you know,
details of what we've, we've seen in the data

80
00:04:37.450 --> 00:04:39.239
<v Dan Reid>and graphs alongside it.

81
00:04:39.649 --> 00:04:45.170
<v Dan Reid>Um, but now you also have an interactive
co-pilot you can talk to and it's like any

82
00:04:45.170 --> 00:04:50.179
<v Dan Reid>chatbot that you're used to, um, but
within, within Sensei, and it has the, the

83
00:04:50.179 --> 00:04:51.160
<v Dan Reid>context that it needs.

84
00:04:51.220 --> 00:04:54.970
<v Dan Reid>So it understands the asset you're looking
at, the case, um, you know, what we've

85
00:04:54.970 --> 00:04:55.940
<v Dan Reid>seen in the data.

86
00:04:56.299 --> 00:05:00.970
<v Dan Reid>Uh, historic cases which are of, of
relevance, recent work events, all of this

87
00:05:00.970 --> 00:05:02.859
<v Dan Reid>information it has, has available to it.

88
00:05:03.260 --> 00:05:08.279
<v Dan Reid>So you're able to ask questions about, um,
you know, what, what's the situation,

89
00:05:08.619 --> 00:05:12.799
<v Dan Reid>what should I do next, um, and it can help
guide the user, and this is coming back

90
00:05:12.799 --> 00:05:17.739
<v Dan Reid>to this, um, this issue of, you know,
having maintainers who may be less

91
00:05:17.739 --> 00:05:21.600
<v Dan Reid>experienced, um, so we're trying to
provide guidance to them.

92
00:05:22.010 --> 00:05:26.899
<v Dan Reid>And uh just one more thing that I think
the more interesting topic is obviously it's

93
00:05:26.899 --> 00:05:31.010
<v Dan Reid>nice having that interactive co-pilot to
talk to, but it's even better if you don't

94
00:05:31.010 --> 00:05:32.350
<v Dan Reid>have to talk to the co-pilot, right?

95
00:05:32.470 --> 00:05:37.929
<v Dan Reid>So when we raise cases, not only do we
present the information we've seen, we also

96
00:05:37.929 --> 00:05:40.790
<v Dan Reid>then provide prescriptive guidance
alongside it.

97
00:05:41.230 --> 00:05:46.890
<v Dan Reid>Um, and you know, that we suggest what are
the potential root causes to this

98
00:05:46.890 --> 00:05:50.850
<v Dan Reid>problem, so that, you know, the engineers
can, the maintainers can look at the, the

99
00:05:50.850 --> 00:05:55.429
<v Dan Reid>data, look at the, um, potential root
causes and know what to, what to do next.

100
00:05:56.750 --> 00:05:57.600
<v Speaker 2>No, no, absolutely.

101
00:05:57.670 --> 00:06:01.209
<v Speaker 2>And what, what you talked about there
within the sense applications, there's already

102
00:06:01.209 --> 00:06:04.630
<v Speaker 2>a vast amount of data which the co-pilot
could use to provide that prescriptive

103
00:06:04.630 --> 00:06:08.829
<v Speaker 2>guidance, whether in, in an interactive
way or in the, um, in the cases, which

104
00:06:08.829 --> 00:06:12.899
<v Speaker 2>again, I think you raised on one of our
previous, um, when we launched it, it was

105
00:06:12.899 --> 00:06:13.390
<v Speaker 2>very important.

106
00:06:13.429 --> 00:06:16.829
<v Speaker 2>You sort of told that story of the, you
know, that there's a great summary here.

107
00:06:16.910 --> 00:06:18.950
<v Speaker 2>So people like me, I could go in there and
understand.

108
00:06:19.026 --> 00:06:20.455
<v Speaker 2>Stuff straight away, which I thought was
really cool.

109
00:06:22.515 --> 00:06:24.256
<v Speaker 2>Exactly, exactly, really cool
functionality.

110
00:06:24.466 --> 00:06:26.346
<v Speaker 2>But all that knowledge within the app is
one thing.

111
00:06:26.425 --> 00:06:30.425
<v Speaker 2>But, um, and back, I guess to our first
question, and we talked about the knowledge

112
00:06:30.425 --> 00:06:32.716
<v Speaker 2>that's actually being lost, um, due to
retirement.

113
00:06:32.855 --> 00:06:35.506
<v Speaker 2>So there's knowledge within the app, but
there's also all this other knowledge

114
00:06:35.506 --> 00:06:36.865
<v Speaker 2>that's leaving the organization.

115
00:06:37.026 --> 00:06:41.425
<v Speaker 2>So what, what type of knowledge is sort
of, you know, leaving with retirement.

116
00:06:41.941 --> 00:06:43.941
<v Speaker 2>People changing industries, etc.

117
00:06:44.342 --> 00:06:44.541
<v Speaker 2>Yeah,

118
00:06:44.592 --> 00:06:45.701
<v Dan Reid>a lot unfortunately.

119
00:06:45.821 --> 00:06:51.302
<v Dan Reid>So I, I think this is, um, you know, the
individuals who are retiring have been in

120
00:06:51.302 --> 00:06:52.661
<v Dan Reid>the industry for decades.

121
00:06:52.821 --> 00:06:56.731
<v Dan Reid>They've built up loads of knowledge
through that time and especially that that

122
00:06:56.731 --> 00:07:00.761
<v Dan Reid>generation, actually people's tenure at
their current employers are normally longer

123
00:07:00.761 --> 00:07:03.511
<v Dan Reid>than, you know, what, what we might have
in our generation.

124
00:07:03.622 --> 00:07:07.582
<v Dan Reid>So you might be talking that they've
actually had decades of experience in that

125
00:07:07.582 --> 00:07:11.141
<v Dan Reid>factory, right, working with them assets,
working with the processes.

126
00:07:11.450 --> 00:07:16.250
<v Dan Reid>And they know it inside out, really in
depth, and you are, you are losing all of

127
00:07:16.250 --> 00:07:21.119
<v Dan Reid>that, um, and it's, it's almost to the
level where you get that intuition, you know,

128
00:07:21.200 --> 00:07:25.890
<v Dan Reid>when something is breaking, you know,
these these individuals will know what the

129
00:07:25.890 --> 00:07:29.130
<v Dan Reid>likely root cause will be, they will know
how to fix it efficiently, they'll know

130
00:07:29.130 --> 00:07:30.359
<v Dan Reid>how to stop it in the future.

131
00:07:30.529 --> 00:07:35.399
<v Dan Reid>So, um, it's, uh, it's all of that
knowledge that will be lost and you know, it's

132
00:07:35.399 --> 00:07:37.970
<v Dan Reid>often called sort of tribal knowledge
within organizations.

133
00:07:38.010 --> 00:07:40.869
<v Dan Reid>It's that internal knowledge which builds
up over years.

134
00:07:41.339 --> 00:07:47.140
<v Dan Reid>Um, which is just, you know, hard earned
hands on experience, which is very hard to,

135
00:07:47.350 --> 00:07:49.130
<v Dan Reid>to capture and to replace.

136
00:07:49.910 --> 00:07:52.720
<v Speaker 2>Yeah, and, and I guess also, I guess that
point as well.

137
00:07:52.829 --> 00:07:58.510
<v Speaker 2>So it's, I guess it also restricts
collaboration as well, um, on, on that side of

138
00:07:58.510 --> 00:07:58.630
<v Speaker 2>things.

139
00:07:58.690 --> 00:08:02.510
<v Speaker 2>So it's it's actually a barrier to
collaboration because again, if people don't have

140
00:08:02.510 --> 00:08:05.230
<v Speaker 2>the knowledge and how are they supposed to
share that, you know, across their

141
00:08:05.230 --> 00:08:07.950
<v Speaker 2>organization, but I guess there's other
barriers too to collaboration that we

142
00:08:07.950 --> 00:08:09.970
<v Speaker 2>should, that we could raise as well.

143
00:08:10.149 --> 00:08:10.869
<v Dan Reid>Yes, certainly.

144
00:08:11.029 --> 00:08:14.589
<v Dan Reid>So it, it, it might be worth sort of
talking through the, the types of knowledge

145
00:08:14.589 --> 00:08:15.190
<v Dan Reid>we're dealing with.

146
00:08:15.269 --> 00:08:20.959
<v Dan Reid>So, um, on one side you've got a certain,
um, You know, type of knowledge which is

147
00:08:20.959 --> 00:08:27.149
<v Dan Reid>documented, it's explicit, um, and, er,
these are things, you know, on generally you

148
00:08:27.149 --> 00:08:29.130
<v Dan Reid>have sort of manuals for your assets.

149
00:08:29.200 --> 00:08:32.929
<v Dan Reid>So say you buy a Siemens drive, you'll get
a manual with that, and that will talk

150
00:08:32.929 --> 00:08:34.859
<v Dan Reid>you through the maintenance you're
supposed to be doing.

151
00:08:34.929 --> 00:08:38.960
<v Dan Reid>It will have troubleshooting and all of
this information, the components, um, so

152
00:08:38.960 --> 00:08:43.409
<v Dan Reid>that's, you know, one form of very
explicit, uh, knowledge within, within an

153
00:08:43.409 --> 00:08:43.429
<v Dan Reid>organization.

154
00:08:44.030 --> 00:08:47.479
<v Dan Reid>But there's much more local knowledge as
well, which is typically documented.

155
00:08:47.719 --> 00:08:51.679
<v Dan Reid>So they, most organizations will have
standard operating procedures, which are

156
00:08:51.679 --> 00:08:56.679
<v Dan Reid>essentially like step by step guides, you
know, when X happens, you do ABC, um, and

157
00:08:56.679 --> 00:09:02.400
<v Dan Reid>this is local to the organization,
normally local to a particular factory, sometimes

158
00:09:02.400 --> 00:09:04.640
<v Dan Reid>even more local to like a line.

159
00:09:04.979 --> 00:09:10.320
<v Dan Reid>Um, in some situations, but most
manufacturers will have this information available,

160
00:09:10.510 --> 00:09:15.489
<v Dan Reid>um, and I think the, the difficulty is
more the accessibility of it.

161
00:09:15.859 --> 00:09:19.500
<v Dan Reid>So say when you're, you're trying to fix
something, you, you're trying to do some

162
00:09:19.500 --> 00:09:22.859
<v Dan Reid>maintenance, um, you're probably under
time pressure, especially if it's like a

163
00:09:22.859 --> 00:09:23.849
<v Dan Reid>reactive situation.

164
00:09:24.809 --> 00:09:29.239
<v Dan Reid>Um, are people going to be able to find
them documents, uh, in the first place, and

165
00:09:29.239 --> 00:09:32.369
<v Dan Reid>even if they do find a document, some of
these documents are hundreds of pages,

166
00:09:32.489 --> 00:09:33.609
<v Dan Reid>right, they're very intense.

167
00:09:33.929 --> 00:09:38.210
<v Dan Reid>So you actually have to find the
information within the document that you, which,

168
00:09:38.330 --> 00:09:39.669
<v Dan Reid>you know, provides value.

169
00:09:40.039 --> 00:09:44.630
<v Dan Reid>Um, so this sort of accessibility of that,
that knowledge is, is problematic.

170
00:09:45.390 --> 00:09:50.140
<v Dan Reid>And then on the other sort of main type of
knowledge is more that sort of tribal

171
00:09:50.140 --> 00:09:50.380
<v Dan Reid>knowledge.

172
00:09:50.460 --> 00:09:55.380
<v Dan Reid>So it's that inbuilt, um, in a company
knowledge that gets built up over time

173
00:09:55.380 --> 00:09:55.989
<v Dan Reid>experience.

174
00:09:56.140 --> 00:10:00.500
<v Dan Reid>It's understanding actually these types of
assets fail in these sort of ways and

175
00:10:00.500 --> 00:10:05.059
<v Dan Reid>actually, you know, you have to tweak this
and turn that to fix it, you know, it's

176
00:10:05.059 --> 00:10:07.500
<v Dan Reid>that the sort of knowledge that gets built
up, um.

177
00:10:08.270 --> 00:10:10.919
<v Dan Reid>And this is just hard knowledge to
document, right?

178
00:10:10.960 --> 00:10:15.070
<v Dan Reid>It's hard to sit someone down and say,
what are your intuitions, write, write them

179
00:10:15.070 --> 00:10:17.340
<v Dan Reid>down for me, er, it's, it's just
difficult.

180
00:10:17.799 --> 00:10:22.159
<v Dan Reid>Um, I think the closest manufacturers have
a sort of maintenance systems.

181
00:10:22.520 --> 00:10:28.229
<v Dan Reid>So these are databases where you log the
work and maintenance which happens on an

182
00:10:28.229 --> 00:10:29.200
<v Dan Reid>asset over time.

183
00:10:29.520 --> 00:10:33.919
<v Dan Reid>So you do build up a history of, um, of
what's happened to that asset.

184
00:10:34.510 --> 00:10:37.390
<v Dan Reid>Um, and that's incredibly useful.

185
00:10:38.109 --> 00:10:42.190
<v Dan Reid>The issues with maintenance systems is,
firstly, they're notoriously, they're

186
00:10:42.190 --> 00:10:44.950
<v Dan Reid>notorious for not being very, uh, high
quality.

187
00:10:45.440 --> 00:10:49.349
<v Dan Reid>So you know, sometimes records aren't
captured or actually the information captured

188
00:10:49.349 --> 00:10:51.739
<v Dan Reid>isn't detailed enough to be of use.

189
00:10:52.280 --> 00:10:53.640
<v Dan Reid>Now that's not always all the time.

190
00:10:53.679 --> 00:10:57.400
<v Dan Reid>I, I should, um, I should comment because
actually a lot of our customers do keep

191
00:10:57.400 --> 00:10:58.140
<v Dan Reid>good records.

192
00:10:58.520 --> 00:11:02.679
<v Dan Reid>Um, but also then you've got another
accessibility problem that you've got a

193
00:11:02.679 --> 00:11:05.219
<v Dan Reid>mountain of records for, for the asset.

194
00:11:05.940 --> 00:11:08.419
<v Dan Reid>You, you're under that time pressure, you
want to get something fixed.

195
00:11:08.830 --> 00:11:14.169
<v Dan Reid>Um, you know, it's hard to dig through
that, that pile of records to find situations

196
00:11:14.169 --> 00:11:16.169
<v Dan Reid>in the past which might give you some
guidance.

197
00:11:16.580 --> 00:11:20.619
<v Dan Reid>Um, and you know, in an ideal world, you
wouldn't just be looking at that single

198
00:11:20.619 --> 00:11:24.859
<v Dan Reid>asset history, you'd be looking at every
asset which is similar in the, in the site

199
00:11:24.859 --> 00:11:29.419
<v Dan Reid>or in the whole organization and picking
out actually that, that incident two years

200
00:11:29.419 --> 00:11:33.179
<v Dan Reid>ago in the site across America is very
similar to what I'm having now.

201
00:11:33.299 --> 00:11:35.020
<v Dan Reid>Maybe I can learn from that.

202
00:11:35.179 --> 00:11:37.630
<v Dan Reid>So, you know, accessibility is.

203
00:11:38.020 --> 00:11:43.940
<v Dan Reid>It is one of the key issues, um, and like
any industry there's, there's silos

204
00:11:43.940 --> 00:11:48.380
<v Dan Reid>between teams and plants and you know, for
a lot of global companies, language

205
00:11:48.380 --> 00:11:50.280
<v Dan Reid>barriers, there's shift work.

206
00:11:50.840 --> 00:11:55.219
<v Dan Reid>Uh, you know, people are busy, very, very,
very responsible for keeping the machine,

207
00:11:55.440 --> 00:11:58.760
<v Dan Reid>the factory running, right, rather than
writing down knowledge.

208
00:11:58.840 --> 00:12:00.580
<v Dan Reid>So I don't know, it's a tricky problem.

209
00:12:01.299 --> 00:12:04.900
<v Speaker 2>And like I say, having the time to
actually read through manuals and find the exact

210
00:12:05.440 --> 00:12:06.039
<v Speaker 2>part they need.

211
00:12:06.119 --> 00:12:09.799
<v Speaker 2>So what, what we're hearing here is
there's, there's these, all these pockets of

212
00:12:09.799 --> 00:12:12.719
<v Speaker 2>knowledge, whether it's people, whether
it's manuals, we could talk about a manual

213
00:12:12.719 --> 00:12:17.700
<v Speaker 2>in a dusty cupboard somewhere or a manual
uploaded into, you know, a file sharing

214
00:12:17.979 --> 00:12:19.700
<v Speaker 2>page, but there's just all this knowledge.

215
00:12:20.409 --> 00:12:23.549
<v Speaker 2>Dotted around, what you want to do is
bring it into one place together, which leads

216
00:12:23.549 --> 00:12:24.280
<v Speaker 2>us to

217
00:12:24.590 --> 00:12:25.510
<v Dan Reid>perfect introduction

218
00:12:26.710 --> 00:12:30.190
<v Speaker 2>to to knowledge builder, which we're
introducing in Sensi predictive maintenance.

219
00:12:30.229 --> 00:12:34.250
<v Speaker 2>And I guess let's just say, how does it
work, Dan, I guess is a simple question.

220
00:12:35.039 --> 00:12:39.369
<v Dan Reid>Yeah, so, um, yeah, Knowledge Builder is
one of our sort of newest features related

221
00:12:39.369 --> 00:12:42.909
<v Dan Reid>to, to knowledge, but obviously, you know,
like I said, you know, we've been doing

222
00:12:42.909 --> 00:12:45.909
<v Dan Reid>this for a couple of years now, so there's
others which we can talk through in a

223
00:12:45.909 --> 00:12:47.229
<v Dan Reid>bit, um.

224
00:12:48.260 --> 00:12:52.469
<v Dan Reid>But KnowledgeBuilder is all around um
capturing that explicit documented knowledge

225
00:12:52.469 --> 00:12:53.359
<v Dan Reid>into the system.

226
00:12:53.739 --> 00:12:58.619
<v Dan Reid>So users are able to take their manuals,
their procedures, upload it into Sensei.

227
00:12:59.260 --> 00:13:04.039
<v Dan Reid>um, Sensei then processes that information
and makes it available er for us to use.

228
00:13:04.469 --> 00:13:09.960
<v Dan Reid>And the way we use it in is um in, in two
ways, uh, and it relates to the co-pilot

229
00:13:09.960 --> 00:13:10.820
<v Dan Reid>we were discussing earlier.

230
00:13:10.960 --> 00:13:15.580
<v Dan Reid>So when you're having an interactive
discussion with the co-pilot, you can ask

231
00:13:15.580 --> 00:13:19.000
<v Dan Reid>questions about what you're doing or the
maintenance you want to perform, and then

232
00:13:19.000 --> 00:13:23.479
<v Dan Reid>the co-pilot can refer to the
documentation you've provided, uh, to provide, you

233
00:13:23.479 --> 00:13:24.659
<v Dan Reid>know, the user guidance.

234
00:13:25.090 --> 00:13:28.929
<v Dan Reid>Um, but as I said before, it's nice to not
have to ask for this information.

235
00:13:29.049 --> 00:13:35.479
<v Dan Reid>So alongside that, we're also, um,
proactively looking at the documentation and, um,

236
00:13:35.530 --> 00:13:39.440
<v Dan Reid>identifying, um, information which would
be useful to bring into cases.

237
00:13:39.890 --> 00:13:44.609
<v Dan Reid>So when we are providing that prescriptive
guidance, we're now supplementing it with

238
00:13:44.609 --> 00:13:48.650
<v Dan Reid>the, um, that explicit knowledge which is
captured in that documentation.

239
00:13:49.179 --> 00:13:53.580
<v Dan Reid>So you know, the step, the troubleshooting
section of your manual, we're, you know,

240
00:13:53.630 --> 00:13:57.179
<v Dan Reid>we're bringing that into, into cases or
the procedures that you're supposed to do

241
00:13:57.179 --> 00:14:01.330
<v Dan Reid>in, in response to certain situations, we
can bring that into the case as well.

242
00:14:01.700 --> 00:14:07.940
<v Dan Reid>So we are solving that sort of
accessibility problem um by, by essentially serving

243
00:14:07.940 --> 00:14:09.320
<v Dan Reid>up the information people need.

244
00:14:10.039 --> 00:14:14.000
<v Dan Reid>Um, and we're also trying to address that,
you know, the, the guidance side of

245
00:14:14.000 --> 00:14:18.659
<v Dan Reid>things, so when you, when people do need
more guidance, we can actually give them,

246
00:14:18.789 --> 00:14:23.380
<v Dan Reid>um, that guidance upfront, sourced from,
from their own documentation.

247
00:14:24.900 --> 00:14:28.140
<v Speaker 2>I guess, uh, I guess the first question
people would be interested to know is what,

248
00:14:28.219 --> 00:14:31.729
<v Speaker 2>what types of results, I know we talked
about manuals and, and different things, but

249
00:14:31.729 --> 00:14:34.700
<v Speaker 2>are there sort of specific examples you
could give which would provide the most

250
00:14:34.700 --> 00:14:38.479
<v Speaker 2>value, you know, to enhance the, you know,
the out the outputs.

251
00:14:39.140 --> 00:14:46.140
<v Dan Reid>Yeah, so you, you can upload pretty much
any, um, file, um, so you

252
00:14:46.140 --> 00:14:51.200
<v Dan Reid>know, text files, um, PDFs, uh,
office-based documents, audio video.

253
00:14:51.609 --> 00:14:55.659
<v Dan Reid>Um, and that's housed within, um, Sensi
for people to make use of.

254
00:14:56.179 --> 00:15:01.619
<v Dan Reid>Now right now the co-pilot can only make
use of, uh, text and PDF, um, files, but

255
00:15:01.619 --> 00:15:05.020
<v Dan Reid>we're obviously going to broaden out, um,
support for other, other files in the

256
00:15:05.020 --> 00:15:05.359
<v Dan Reid>future.

257
00:15:06.140 --> 00:15:09.580
<v Dan Reid>Um, and what's important to understand is
actually when we're processing these

258
00:15:09.580 --> 00:15:14.780
<v Dan Reid>files, we use, um, vision models, and that
what that means is if you're uploading

259
00:15:14.780 --> 00:15:19.460
<v Dan Reid>documents which have a lot of flow charts
or diagrams or tables of information, we

260
00:15:19.460 --> 00:15:22.400
<v Dan Reid>can, um, consume that information and make
it available.

261
00:15:22.830 --> 00:15:28.659
<v Dan Reid>Uh, and you know, er, many people might
have er scanned er paper documents and we

262
00:15:28.659 --> 00:15:31.840
<v Dan Reid>can make use of that as well through these
vision um technologies.

263
00:15:32.320 --> 00:15:36.299
<v Dan Reid>So you know, we can accept quite a broad
range of documentation.

264
00:15:36.960 --> 00:15:41.780
<v Dan Reid>And that obviously complements the
knowledge we already have inside of Sensai.

265
00:15:42.280 --> 00:15:47.799
<v Dan Reid>So, you know, the typical way we work is
we collect um measurements from people's

266
00:15:47.799 --> 00:15:50.580
<v Dan Reid>machines and we obviously track that for a
very long time.

267
00:15:50.640 --> 00:15:54.440
<v Dan Reid>Um, so we're sort of building up a history
of that asset within Senseii.

268
00:15:55.130 --> 00:15:58.739
<v Dan Reid>Um, and then we complement that with, you
know, integrations into maintenance

269
00:15:58.739 --> 00:16:02.700
<v Dan Reid>systems, so we're bringing in information
about what work has been performed on

270
00:16:02.700 --> 00:16:03.299
<v Dan Reid>assets.

271
00:16:03.859 --> 00:16:08.780
<v Dan Reid>Um, and then we also collect our own
information so people can leave comments in

272
00:16:08.780 --> 00:16:13.549
<v Dan Reid>Sensai if they can provide feedback on, on
cases, uh, we collect manual

273
00:16:13.549 --> 00:16:13.880
<v Dan Reid>measurements.

274
00:16:13.989 --> 00:16:18.780
<v Dan Reid>There's a whole, you know, load of
information that we have, and that information is

275
00:16:18.780 --> 00:16:20.669
<v Dan Reid>much closer to that more tribal knowledge.

276
00:16:20.750 --> 00:16:24.630
<v Dan Reid>It's more of that sort of history of an
asset over time, what, what fails and what

277
00:16:24.630 --> 00:16:27.020
<v Dan Reid>do we typically do with that, with that
asset.

278
00:16:27.549 --> 00:16:31.109
<v Dan Reid>Um, so I, I think knowledge builder is
really interesting because we're, we're

279
00:16:31.109 --> 00:16:34.950
<v Dan Reid>taking what I would consider the more
tribal and then we're adding in the, the

280
00:16:34.950 --> 00:16:40.059
<v Dan Reid>explicit documented and that sort of
combination of all of that in one place, um,

281
00:16:40.349 --> 00:16:44.270
<v Dan Reid>being able to pull out, pull on that
information as people need it and surface it

282
00:16:44.270 --> 00:16:45.140
<v Dan Reid>when they need it.

283
00:16:45.469 --> 00:16:49.429
<v Dan Reid>We're really, you know, tackling the core
problems that lie at this sort of

284
00:16:49.429 --> 00:16:51.330
<v Dan Reid>knowledge problem, uh, for customers.

285
00:16:52.109 --> 00:16:55.369
<v Speaker 2>Yeah, and I guess actually specifically,
so combining all this knowledge makes a lot

286
00:16:55.369 --> 00:16:58.710
<v Speaker 2>of sense and it's continuously learning
and improving from all these different

287
00:16:58.710 --> 00:17:00.690
<v Speaker 2>sources of knowledge, which is fantastic.

288
00:17:00.969 --> 00:17:05.170
<v Speaker 2>But the, the output of this is obviously
prescriptive guide, you know, like

289
00:17:05.170 --> 00:17:07.719
<v Speaker 2>suggestions on, you know, what, what they
should do, what should they investigate,

290
00:17:07.770 --> 00:17:09.819
<v Speaker 2>how does it Sort of fit, fit in with that.

291
00:17:09.849 --> 00:17:11.890
<v Speaker 2>How does it produce that output, I guess.

292
00:17:12.170 --> 00:17:14.770
<v Dan Reid>Yeah, so it's, um, it's very interesting.

293
00:17:14.849 --> 00:17:19.680
<v Dan Reid>So, uh, when we, when we spot something in
the data, um, so this might be an

294
00:17:19.680 --> 00:17:26.530
<v Dan Reid>abnormality that we've seen a step or a
spike or a sort of a trend, um, we'll raise

295
00:17:26.530 --> 00:17:27.770
<v Dan Reid>that to a customer in a case.

296
00:17:28.010 --> 00:17:31.719
<v Dan Reid>So the cases start off with describing
what we've seen in the data.

297
00:17:31.930 --> 00:17:35.349
<v Dan Reid>So we describe, um, the actual
measurements which have changed.

298
00:17:35.949 --> 00:17:41.430
<v Dan Reid>And then we, we take them patterns we've
seen, we'll look back, um, in the asset's

299
00:17:41.430 --> 00:17:45.989
<v Dan Reid>history, er, but not only the asset in
question, it's any asset which is similar

300
00:17:45.989 --> 00:17:47.030
<v Dan Reid>across the organization.

301
00:17:47.109 --> 00:17:50.829
<v Dan Reid>So that might be across sites if you've
got very good sort of uniformity of, of

302
00:17:50.829 --> 00:17:55.709
<v Dan Reid>assets, and we'll try to identify cases or
situations in the past which look similar

303
00:17:55.709 --> 00:17:56.589
<v Dan Reid>to what we've seen now.

304
00:17:57.849 --> 00:18:01.719
<v Dan Reid>Um, and when we find them situations in
the past, we'll gather up all of the

305
00:18:01.719 --> 00:18:02.530
<v Dan Reid>knowledge around them.

306
00:18:02.609 --> 00:18:05.969
<v Dan Reid>So there might be comments from people,
there might be work events, there might be

307
00:18:05.969 --> 00:18:10.349
<v Dan Reid>feedback on cases, and we'll gather all of
that knowledge together and then we'll

308
00:18:10.349 --> 00:18:14.209
<v Dan Reid>from that we'll try to determine what the
root causes of that problem might be.

309
00:18:14.489 --> 00:18:19.030
<v Dan Reid>And obviously the history is a good guide
to, to what that the issue may be.

310
00:18:19.680 --> 00:18:22.930
<v Dan Reid>And now with Knowledge Builder, we're
supplementing that with the explicit

311
00:18:22.930 --> 00:18:23.729
<v Dan Reid>documentation.

312
00:18:24.150 --> 00:18:27.859
<v Dan Reid>So as well as looking at that history,
we'll bring in information from the manuals,

313
00:18:27.949 --> 00:18:32.310
<v Dan Reid>from the procedures, er, to try to
identify, you know, in, you know, the manual

314
00:18:32.310 --> 00:18:36.510
<v Dan Reid>might say if you're, if you're suffering
this situation, then this is what it might

315
00:18:36.510 --> 00:18:39.229
<v Dan Reid>be and how to solve it, and we can
obviously bring that into the case.

316
00:18:39.699 --> 00:18:44.689
<v Dan Reid>er, likewise with the procedures, if we
see, um, an incident that has a procedure

317
00:18:45.030 --> 00:18:48.510
<v Dan Reid>associated with it, we can obviously bring
you information about that procedure into

318
00:18:48.510 --> 00:18:49.689
<v Dan Reid>the case as well.

319
00:18:50.069 --> 00:18:55.709
<v Dan Reid>So I think it's, you know, incredibly
powerful and um, as I'm sure most of the

320
00:18:55.709 --> 00:19:01.709
<v Dan Reid>listeners know, AI er can make mistakes,
right, and er hallucinations are a thing.

321
00:19:02.069 --> 00:19:08.189
<v Dan Reid>So what we've tried to do is to be very um
try to build trust in the system, I

322
00:19:08.189 --> 00:19:10.410
<v Dan Reid>suppose, and so when we are referencing.

323
00:19:10.900 --> 00:19:15.380
<v Dan Reid>Historic events, we'll provide links to,
to them events so that the user can just

324
00:19:15.380 --> 00:19:19.739
<v Dan Reid>click from, see that historic event in the
application, make sure that actually, you

325
00:19:19.739 --> 00:19:23.939
<v Dan Reid>know, what the AI is deducing from that
past event makes sense and it matches what

326
00:19:23.939 --> 00:19:25.000
<v Dan Reid>you're seeing now.

327
00:19:25.420 --> 00:19:29.410
<v Dan Reid>Um, and likewise with the documents, we're
providing references not only to the

328
00:19:29.410 --> 00:19:32.939
<v Dan Reid>document of a specific page in that
document, because like I say, these things are

329
00:19:32.939 --> 00:19:34.900
<v Dan Reid>hundreds of pages in some instances.

330
00:19:35.510 --> 00:19:40.469
<v Dan Reid>So users can click on that, see the page
in the application, make sure that some

331
00:19:40.469 --> 00:19:43.790
<v Dan Reid>summary is correct and you know, it's,
it's pulling out the right information.

332
00:19:43.869 --> 00:19:48.949
<v Dan Reid>So we're trying to make sure that
customers can trust the AI and take meaningful

333
00:19:48.949 --> 00:19:52.150
<v Dan Reid>action off of what we're, what we're
raising in the application.

334
00:19:52.800 --> 00:19:56.660
<v Speaker 2>Yeah, so I guess it's one thing seeing,
you know, guidance on, on what to do next,

335
00:19:56.729 --> 00:20:00.760
<v Speaker 2>but some, you know, sometime, you know, if
there's a time or not, but it's good to

336
00:20:00.760 --> 00:20:04.920
<v Speaker 2>have that option to be able to dive into
the original source of it and, and dig in

337
00:20:04.920 --> 00:20:05.560
<v Speaker 2>further to the issue.

338
00:20:05.640 --> 00:20:08.530
<v Speaker 2>I mean, because they might be asked
questions, other colleagues might say, you know,

339
00:20:08.760 --> 00:20:12.079
<v Speaker 2>how does that relate to that, you know,
what, what, what's the detail behind that?

340
00:20:12.400 --> 00:20:13.239
<v Speaker 2>And that's that next step.

341
00:20:13.280 --> 00:20:16.640
<v Speaker 2>And obviously, of course, the interactive
element where you can ask questions is a

342
00:20:16.640 --> 00:20:19.400
<v Speaker 2>key part of that as well, where you can,
you know, get those answers as well.

343
00:20:19.439 --> 00:20:22.560
<v Speaker 2>If it, if, if in the summer it doesn't
tell the complete.

344
00:20:22.719 --> 00:20:23.790
<v Speaker 2>Picture in some cases

345
00:20:23.790 --> 00:20:25.030
<v Dan Reid>that yeah, it's a very good point.

346
00:20:25.150 --> 00:20:29.910
<v Dan Reid>So yeah, the co-pilot is already available
there and you can open that up and you

347
00:20:29.910 --> 00:20:31.109
<v Dan Reid>can ask onward questions.

348
00:20:31.229 --> 00:20:36.069
<v Dan Reid>So if you've got, you know, if you want
clarification on certain topics, um, or if

349
00:20:36.069 --> 00:20:40.349
<v Dan Reid>you want to have, you know, more general
advice, it can provide that.

350
00:20:40.469 --> 00:20:41.670
<v Dan Reid>But yeah, exactly that.

351
00:20:41.780 --> 00:20:45.089
<v Dan Reid>So that, you know, that interactive
session is there if you need it.

352
00:20:45.390 --> 00:20:48.869
<v Dan Reid>And then you can also dig into the details
of, you know, past events or the

353
00:20:48.869 --> 00:20:52.449
<v Dan Reid>documentation to, to help guide for the
onward actions.

354
00:20:53.560 --> 00:20:57.719
<v Speaker 2>And I guess another question as well
because we, you know, the day we're recording

355
00:20:57.719 --> 00:21:01.119
<v Speaker 2>this, we've, we've done sort of quite a
big announcement actually on sort of new

356
00:21:01.119 --> 00:21:05.430
<v Speaker 2>capabilities in the app where Um, usually
we focus on sort of time series data, you

357
00:21:05.430 --> 00:21:09.160
<v Speaker 2>know, in order to use this knowledge
functionality, but we're sort of taking a

358
00:21:09.160 --> 00:21:11.160
<v Speaker 2>slightly different approach or we've got a
new approach to that.

359
00:21:11.239 --> 00:21:15.400
<v Speaker 2>So maybe you could sort of, um, you know,
elaborate on, on where we're heading in

360
00:21:15.400 --> 00:21:15.959
<v Speaker 2>that direction.

361
00:21:16.189 --> 00:21:16.839
<v Dan Reid>Yes, that's right.

362
00:21:17.000 --> 00:21:21.280
<v Dan Reid>So, um, in the past we have focused on
produce maintenance, and that's been, you

363
00:21:21.280 --> 00:21:25.260
<v Dan Reid>know, reliant somewhat on, you know, the
measurements and, uh, time series data.

364
00:21:25.670 --> 00:21:29.760
<v Dan Reid>But we, we are, we appreciate that
actually this knowledge problem, it's a hard

365
00:21:29.760 --> 00:21:31.550
<v Dan Reid>problem and people are looking for
solutions.

366
00:21:31.640 --> 00:21:37.599
<v Dan Reid>So we're keen to have that, um, you know,
Sensi support that need without requiring

367
00:21:37.599 --> 00:21:41.920
<v Dan Reid>people to, you know, get data, to get time
series into the cloud and all that

368
00:21:41.920 --> 00:21:42.979
<v Dan Reid>connectivity problem.

369
00:21:43.479 --> 00:21:49.000
<v Dan Reid>So we're, we're working on, um, allowing
users to use Sensi just for the knowledge

370
00:21:49.000 --> 00:21:50.280
<v Dan Reid>part of this equation.

371
00:21:50.770 --> 00:21:55.449
<v Dan Reid>So roughly how that works as a user um
would start using Sensi, they'll define the

372
00:21:55.449 --> 00:21:59.050
<v Dan Reid>sort of the rough asset types they're
working with, so they've got robots or, you

373
00:21:59.050 --> 00:22:02.890
<v Dan Reid>know, conveyor belts or whatever, and then
for each one of them they'd upload the

374
00:22:02.890 --> 00:22:05.650
<v Dan Reid>documentation they have available, um.

375
00:22:06.430 --> 00:22:09.829
<v Dan Reid>In an ideal world, maybe they'd also
upload their maintenance records to give a bit

376
00:22:09.829 --> 00:22:12.079
<v Dan Reid>more, um, sort of er context.

377
00:22:12.349 --> 00:22:14.989
<v Dan Reid>And then you'll have the co-pilot there to
be able to talk to.

378
00:22:15.109 --> 00:22:17.939
<v Dan Reid>So if you, you're in a certain situation,
you're doing a certain type of

379
00:22:17.939 --> 00:22:21.859
<v Dan Reid>maintenance, you're able to ask the
co-pilot to pull out the information which is

380
00:22:21.859 --> 00:22:22.430
<v Dan Reid>relevant.

381
00:22:22.949 --> 00:22:26.609
<v Dan Reid>So it's a, you know, really addressing
that accessibility, um, issue.

382
00:22:26.910 --> 00:22:31.790
<v Dan Reid>Um, so yeah, so going forward, time series
won't be a requirement.

383
00:22:32.270 --> 00:22:36.900
<v Dan Reid>Um, but as you know, as I've described, I,
I think there is huge benefit to

384
00:22:37.239 --> 00:22:41.800
<v Dan Reid>centralizing knowledge into a system like
Sensei because you're combining that

385
00:22:41.800 --> 00:22:46.680
<v Dan Reid>explicit and the sort of the more history,
um, information of these assets and

386
00:22:46.680 --> 00:22:48.119
<v Dan Reid>you'll get much better results.

387
00:22:48.275 --> 00:22:50.645
<v Dan Reid>Uh, from that, but it can be very much a
progression.

388
00:22:50.724 --> 00:22:52.805
<v Dan Reid>You can start with, you know, the
knowledge side.

389
00:22:52.964 --> 00:22:56.604
<v Dan Reid>Then if you, you, you're seeing benefit,
then you can start adding in time series,

390
00:22:56.724 --> 00:23:00.785
<v Dan Reid>unlocking the predictive maintenance side
of things, and the, you know, the greater

391
00:23:00.785 --> 00:23:04.125
<v Dan Reid>context for, for the knowledge, um,
features.

392
00:23:04.619 --> 00:23:06.260
<v Speaker 2>Yeah, and, and, and that's exactly it.

393
00:23:06.339 --> 00:23:09.699
<v Speaker 2>I mean, yeah, being involved in the press
release, you, and if you've anyone out

394
00:23:09.699 --> 00:23:13.089
<v Speaker 2>there who, who might have read, read the
press release, will see these things about

395
00:23:13.089 --> 00:23:16.209
<v Speaker 2>a pathway to predictive maintenance and
things like that, and that's, that's the

396
00:23:16.209 --> 00:23:20.060
<v Speaker 2>vision that really, you, you know, if you
want to start at scale, fantastic, you

397
00:23:20.060 --> 00:23:24.060
<v Speaker 2>know, the full sense capabilities, but
we're actually recognizing that there are

398
00:23:24.060 --> 00:23:26.670
<v Speaker 2>people or organizations that wants.

399
00:23:26.760 --> 00:23:29.920
<v Speaker 2>Start, you know, slightly earlier and
build up, build up to that, build up the

400
00:23:29.920 --> 00:23:33.010
<v Speaker 2>knowledge, and then add it because the way
I see it actually is sort of you're

401
00:23:33.010 --> 00:23:36.650
<v Speaker 2>adding sort of layers of data as you go,
you know, and we've talked about this

402
00:23:36.650 --> 00:23:39.729
<v Speaker 2>different sources of knowledge, and the
more you, you build up, the more you can,

403
00:23:39.969 --> 00:23:43.449
<v Speaker 2>you know, build towards that predict
maintenance at scale when, when you're ready to

404
00:23:43.449 --> 00:23:43.689
<v Speaker 2>do

405
00:23:43.689 --> 00:23:43.869
<v Dan Reid>so.

406
00:23:43.969 --> 00:23:44.449
<v Dan Reid>Yeah, that's right.

407
00:23:44.489 --> 00:23:45.579
<v Dan Reid>It's very much layers, right?

408
00:23:45.689 --> 00:23:48.790
<v Dan Reid>So as well, if you think about sort of
maintenance strategies.

409
00:23:49.250 --> 00:23:53.209
<v Dan Reid>You know, the documentation side is very
much, you know, reactive and, and plan

410
00:23:53.209 --> 00:23:56.689
<v Dan Reid>maintenance, and then you can sort of add
um data on and get more towards the

411
00:23:56.689 --> 00:23:59.849
<v Dan Reid>predictive end of that sort of maintenance
strategy er journey.

412
00:24:00.089 --> 00:24:02.250
<v Dan Reid>So yeah, very much a, a pathway, that's a,

413
00:24:02.390 --> 00:24:02.729
<v Speaker 2>yeah.

414
00:24:03.140 --> 00:24:03.750
<v Dan Reid>Good press release.

415
00:24:04.520 --> 00:24:05.130
<v Speaker 2>Yeah, good, good marketing.

416
00:24:05.219 --> 00:24:11.290
<v Speaker 2>Good marketing, marketing terms, it's my
fault, um, but, but, and you know, if I,

417
00:24:11.500 --> 00:24:15.459
<v Speaker 2>you know, come to the next question, we
talked about a lot about, you know, um, not

418
00:24:15.459 --> 00:24:19.530
<v Speaker 2>having time series data and all the, all
this, you know, data, but data security is

419
00:24:19.530 --> 00:24:22.729
<v Speaker 2>obviously a big concern when you, when we
start talking about Gen AI and

420
00:24:22.729 --> 00:24:26.699
<v Speaker 2>immediately, you know, alarm bells ring,
and we did talk about this and address this

421
00:24:26.699 --> 00:24:27.619
<v Speaker 2>on previous, you know, we.

422
00:24:27.775 --> 00:24:31.145
<v Speaker 2>Talked about the maintenance co-pilot
launch and stuff, but it might be worth Dan,

423
00:24:31.204 --> 00:24:34.645
<v Speaker 2>if we just sort of talk about that again,
you know, how we ensure that sort of

424
00:24:34.645 --> 00:24:34.844
<v Speaker 2>security.

425
00:24:34.885 --> 00:24:36.755
<v Speaker 2>It's obviously a very important, um,
point.

426
00:24:37.045 --> 00:24:40.444
<v Dan Reid>Yeah, it's always been an important point,
obviously for our customers.

427
00:24:40.844 --> 00:24:43.354
<v Dan Reid>We obviously handling a lot of, um,
important data.

428
00:24:43.964 --> 00:24:45.354
<v Dan Reid>The knowledge is no different to that.

429
00:24:45.474 --> 00:24:49.405
<v Dan Reid>So obviously it's encrypted at rest and in
transit, and we obviously make sure that

430
00:24:49.405 --> 00:24:53.785
<v Dan Reid>it doesn't go anywhere it shouldn't leave
it to our customers or, uh, anywhere else.

431
00:24:54.119 --> 00:24:58.800
<v Dan Reid>Um, I, I think the particular problem, you
know, concerns with generative AI is that

432
00:24:58.800 --> 00:25:03.390
<v Dan Reid>you, you take your organisation's
knowledge, which is obviously incredibly valuable,

433
00:25:03.760 --> 00:25:09.359
<v Dan Reid>uh, you upload it into, um, some of these
language models, and then, um, your cloud

434
00:25:09.359 --> 00:25:13.760
<v Dan Reid>providers or the AI providers take that
knowledge, build it into future models, and

435
00:25:13.760 --> 00:25:16.459
<v Dan Reid>then, you know, it's exposed to, to the
public.

436
00:25:16.920 --> 00:25:20.520
<v Dan Reid>Or I suppose the other concern is that you
provide the knowledge to Senseii, and we

437
00:25:20.520 --> 00:25:23.839
<v Dan Reid>build it into models and then expose it to
other customers of Sensei.

438
00:25:24.199 --> 00:25:27.880
<v Dan Reid>Um, and I can say both of these things
don't happen, uh, you'll be pleased to know.

439
00:25:28.319 --> 00:25:33.199
<v Dan Reid>So, um, you know, the information is
provided to the language model, um, to answer

440
00:25:33.199 --> 00:25:36.479
<v Dan Reid>the questions which are asked of it, and
then it's forgotten.

441
00:25:36.560 --> 00:25:40.060
<v Dan Reid>So there's no, um, training of the textual
information within Sensi.

442
00:25:40.589 --> 00:25:44.959
<v Dan Reid>We do work with cloud and AI providers
obviously, but they are not allowed to use

443
00:25:44.959 --> 00:25:46.790
<v Dan Reid>that data to train future models.

444
00:25:46.839 --> 00:25:48.800
<v Dan Reid>So it doesn't go elsewhere, um.

445
00:25:49.359 --> 00:25:51.119
<v Dan Reid>So it's all, all safe, uh,

446
00:25:52.449 --> 00:25:54.209
<v Speaker 2>so it's, yeah, so it's all segregated.

447
00:25:54.369 --> 00:25:58.810
<v Speaker 2>It's, and it's learning, so it's your own
sort of knowledge repo what we're saying

448
00:25:58.810 --> 00:26:02.170
<v Speaker 2>is our own, your own knowledge repository,
and that's what it's learning from only

449
00:26:02.170 --> 00:26:03.150
<v Speaker 2>your, your knowledge.

450
00:26:03.290 --> 00:26:04.849
<v Speaker 2>The knowledge you put in there is what it
learns.

451
00:26:04.920 --> 00:26:05.160
<v Speaker 2>That's

452
00:26:05.160 --> 00:26:06.400
<v Dan Reid>it, yeah, yeah, very much so.

453
00:26:06.569 --> 00:26:09.770
<v Dan Reid>So you, yeah, you can think of it as each
customer has their own knowledge

454
00:26:09.770 --> 00:26:12.530
<v Dan Reid>repository that is being used by the
language model.

455
00:26:12.569 --> 00:26:16.770
<v Dan Reid>It doesn't go to other customers or to,
um, providers other than to provide, you

456
00:26:16.770 --> 00:26:18.839
<v Dan Reid>know, the language models they're supposed
to be providing.

457
00:26:19.380 --> 00:26:24.219
<v Dan Reid>Um, we do have a certain element of, you
know, knowledge we build into it, um, which

458
00:26:24.219 --> 00:26:27.780
<v Dan Reid>is the sort of a baseline, but obviously
that complements the the knowledge that

459
00:26:27.780 --> 00:26:28.670
<v Dan Reid>customers provide.

460
00:26:29.319 --> 00:26:29.900
<v Speaker 2>Absolutely.

461
00:26:30.300 --> 00:26:34.819
<v Speaker 2>I'm sure I'll be asking you that that
question again as we continue to iterate on

462
00:26:34.819 --> 00:26:36.260
<v Speaker 2>the uh the copilot functionality.

463
00:26:36.739 --> 00:26:41.300
<v Dan Reid>It's an important point and obviously
there's, um, and rightly so there's lots of

464
00:26:41.300 --> 00:26:43.729
<v Dan Reid>concern around it, so it's a, it's a good
thing to we'll,

465
00:26:44.339 --> 00:26:45.329
<v Speaker 2>we'll continue to do.

466
00:26:45.420 --> 00:26:46.239
<v Speaker 2>We might do an episode on it

467
00:26:47.930 --> 00:26:48.670
<v Dan Reid>that might be a bit dry,

468
00:26:50.260 --> 00:26:51.640
<v Speaker 2>maybe, maybe, but, um.

469
00:26:52.199 --> 00:26:55.510
<v Speaker 2>I guess, I mean, of course, we want, we
want to look at the future.

470
00:26:55.560 --> 00:26:57.199
<v Speaker 2>I mean that this is fantastic development.

471
00:26:57.209 --> 00:27:02.280
<v Speaker 2>If we take a step back and look, um, you
and your team or teams of, um, working on

472
00:27:02.280 --> 00:27:07.640
<v Speaker 2>this and, and where we've come from in a,
in a year who's just flown by and, and

473
00:27:07.640 --> 00:27:08.400
<v Speaker 2>there's so much more to come.

474
00:27:08.439 --> 00:27:12.000
<v Speaker 2>But I guess in terms of that knowledge
sharing element, I mean, what do you see and

475
00:27:12.000 --> 00:27:14.819
<v Speaker 2>where that's sort of going to shift in the
next, in the next few years?

476
00:27:15.349 --> 00:27:21.670
<v Dan Reid>Yeah, it's always dangerous making
predictions nowadays, especially, um, but I, I

477
00:27:21.670 --> 00:27:25.750
<v Dan Reid>think it's very much, you know, um,
there's some threads in this conversation, which

478
00:27:25.750 --> 00:27:26.979
<v Dan Reid>I think is the key bits, right?

479
00:27:27.099 --> 00:27:30.189
<v Dan Reid>Accessibility keeps them up, and I think
that's how you solve this knowledge

480
00:27:30.189 --> 00:27:30.449
<v Dan Reid>problem.

481
00:27:30.469 --> 00:27:34.069
<v Dan Reid>There's one thing capturing it, but you
need to unlock the value of it, which means

482
00:27:34.069 --> 00:27:35.689
<v Dan Reid>accessing it when you, when you need it.

483
00:27:36.069 --> 00:27:40.790
<v Dan Reid>Um, so I can see very much, uh, that
developing, and in the ideal world you'd be on

484
00:27:40.790 --> 00:27:43.459
<v Dan Reid>the, you know, the maintenance would be on
the shop floor if they want, might want

485
00:27:43.459 --> 00:27:44.329
<v Dan Reid>their hands free.

486
00:27:44.900 --> 00:27:48.380
<v Dan Reid>Uh, you know, so you could imagine it, you
know, this, this knowledge coming to the

487
00:27:48.380 --> 00:27:52.500
<v Dan Reid>shop floor, and I suppose it's going back
to this, uh, you know, co-pilot or wingman

488
00:27:52.500 --> 00:27:52.819
<v Dan Reid>concept.

489
00:27:52.859 --> 00:27:56.849
<v Dan Reid>You, you do literally want an AI there
beside you providing the knowledge as you

490
00:27:56.849 --> 00:27:57.420
<v Dan Reid>need it, right?

491
00:27:57.579 --> 00:28:02.099
<v Dan Reid>So you could imagine much more natural
language interfaces and, and the like coming.

492
00:28:02.540 --> 00:28:09.140
<v Dan Reid>Um, the other topic we touched on is
around, um, around sort of the lack of

493
00:28:09.140 --> 00:28:12.160
<v Dan Reid>skills within, uh, the, the work,
workforce.

494
00:28:12.650 --> 00:28:18.660
<v Dan Reid>Um, and there, you know, we're solving
that to a certain extent because when we see

495
00:28:18.660 --> 00:28:23.979
<v Dan Reid>an issue we're obviously providing that
prognostic, uh, guidance, uh, um, you know,

496
00:28:24.140 --> 00:28:29.260
<v Dan Reid>prognostic guidance on it, um, but you
know, ideally you'd probably want to be

497
00:28:29.260 --> 00:28:34.619
<v Dan Reid>providing that, um, that knowledge and,
um, you know, training before them

498
00:28:34.619 --> 00:28:35.040
<v Dan Reid>situations happen.

499
00:28:35.339 --> 00:28:38.540
<v Dan Reid>So you could imagine that actually you
take all of the knowledge which Sensi has.

500
00:28:38.930 --> 00:28:42.810
<v Dan Reid>Um, and producing sort of documentation so
you could actually produce, um,

501
00:28:42.819 --> 00:28:48.060
<v Dan Reid>procedures based on the what we've seen in
the past or, uh, training programs and

502
00:28:48.060 --> 00:28:52.569
<v Dan Reid>you start, you know, Siemens has lots of,
um, you know, VR training and the like, so

503
00:28:52.569 --> 00:28:54.329
<v Dan Reid>you could start seeing these things
combined.

504
00:28:54.579 --> 00:28:58.689
<v Dan Reid>I should note that these aren't on the
short term or medium time roadmap.

505
00:28:59.939 --> 00:29:03.780
<v Dan Reid>But you know, you can see this sort of the
directions of this, uh, as it progresses.

506
00:29:03.939 --> 00:29:06.300
<v Speaker 2>But we see because a lot of this stuff has
been very fast moving.

507
00:29:06.420 --> 00:29:09.780
<v Speaker 2>So even like the steps we've made in the
past year, so yeah, although we're not, you

508
00:29:09.780 --> 00:29:12.739
<v Speaker 2>know, we're not explicitly promising these
things, it's not like I said, it's not

509
00:29:12.739 --> 00:29:14.780
<v Speaker 2>beyond the realms of possibility.

510
00:29:14.890 --> 00:29:17.140
<v Dan Reid>No, no, I think this is where, where
you'll see things.

511
00:29:17.180 --> 00:29:20.660
<v Dan Reid>And obviously, you know, the, the AI is
just getting smarter and smarter.

512
00:29:20.739 --> 00:29:23.069
<v Dan Reid>So you'd bring in more and more data.

513
00:29:23.219 --> 00:29:27.689
<v Dan Reid>Um, some of our customers actually sort of
record videos of the experienced

514
00:29:27.689 --> 00:29:29.839
<v Dan Reid>maintainers working on machines.

515
00:29:30.199 --> 00:29:33.930
<v Dan Reid>Um, and it'd be great if you could
actually sort of take that information and

516
00:29:33.930 --> 00:29:36.640
<v Dan Reid>understand it and then surface it when
these situations occur.

517
00:29:36.719 --> 00:29:40.439
<v Dan Reid>And you can imagine, you know, situations
like that, you just extend and extend the

518
00:29:40.439 --> 00:29:44.760
<v Dan Reid>knowledge which is captured and, um, you
know, the ability to surface it just when

519
00:29:44.760 --> 00:29:46.420
<v Dan Reid>it's needed, um,

520
00:29:46.439 --> 00:29:46.880
<v Speaker 2>going forward.

521
00:29:47.000 --> 00:29:50.719
<v Speaker 2>I think that actually to pick out one of
those points you made about, so you know,

522
00:29:50.959 --> 00:29:52.979
<v Speaker 2>produce the actual manuals based on.

523
00:29:53.239 --> 00:29:57.400
<v Speaker 2>The inputs from, you know, as it's
learning and learning and improving it's not,

524
00:29:57.599 --> 00:30:01.640
<v Speaker 2>it's knowledge, the system, I mean, and
just being able to produce, you know, new,

525
00:30:01.910 --> 00:30:05.000
<v Speaker 2>it's almost like new manuals for people
to, to use.

526
00:30:05.540 --> 00:30:06.520
<v Speaker 2>I think that's just so powerful.

527
00:30:06.640 --> 00:30:07.319
<v Speaker 2>That'd be amazing,

528
00:30:07.520 --> 00:30:12.560
<v Dan Reid>yeah, and it really, it really helps
because, um, you know, when, when you're, like

529
00:30:12.560 --> 00:30:15.839
<v Dan Reid>I say, there's, there's not much, um, you
know, they're struggling with this sort of

530
00:30:15.839 --> 00:30:20.640
<v Dan Reid>skills gap in the, in the market, and one
of the issues here is that obviously when

531
00:30:20.640 --> 00:30:22.420
<v Dan Reid>newcomers come into an organization.

532
00:30:23.189 --> 00:30:27.750
<v Dan Reid>They typically go through a period of
training and mentorship, right, and the people

533
00:30:27.750 --> 00:30:30.589
<v Dan Reid>delivering that mentorship are the
experienced people who are leaving for

534
00:30:30.589 --> 00:30:31.310
<v Dan Reid>retirement.

535
00:30:31.729 --> 00:30:37.150
<v Dan Reid>So actually you don't have that sort of
mentor network and support, um, as you did

536
00:30:37.150 --> 00:30:42.150
<v Dan Reid>in the past, and that learning curve is so
much steeper for these individuals, and

537
00:30:42.150 --> 00:30:45.390
<v Dan Reid>for the organization, obviously it's going
to cost more time and resources getting

538
00:30:45.390 --> 00:30:46.790
<v Dan Reid>them to the point where they're
productive.

539
00:30:46.910 --> 00:30:51.390
<v Dan Reid>So if you could, Uh, proactively provide,
um, you know, that knowledge and the

540
00:30:51.390 --> 00:30:56.750
<v Dan Reid>training, um, based on, um, you know,
information with knowledge, er, within sort of

541
00:30:56.750 --> 00:31:01.270
<v Dan Reid>knowledge products like Sensei, it could
be very beneficial to, to the upskilling

542
00:31:01.270 --> 00:31:02.430
<v Dan Reid>of, of individuals.

543
00:31:03.319 --> 00:31:06.609
<v Speaker 2>Yeah, and I guess actually, because I was
just looking at sort of the final question

544
00:31:06.609 --> 00:31:10.319
<v Speaker 2>I was going to ask you in a, in a way that
what you're saying just then, um, has

545
00:31:10.319 --> 00:31:11.050
<v Speaker 2>answered that question.

546
00:31:11.130 --> 00:31:14.250
<v Speaker 2>But if we're going to narrow it down to
sort of one piece of advice and maintenance

547
00:31:14.250 --> 00:31:17.489
<v Speaker 2>teams, we know you described very
brilliantly at the beginning about, you know, the

548
00:31:17.489 --> 00:31:20.530
<v Speaker 2>struggles about knowledge loss and being
able to collaborate with colleagues if, you

549
00:31:20.530 --> 00:31:22.930
<v Speaker 2>know, from this side of the world to this
side of the world, but even in plants

550
00:31:22.930 --> 00:31:24.329
<v Speaker 2>sometimes there's, there's barriers.

551
00:31:24.594 --> 00:31:27.925
<v Speaker 2>But what, what's one piece of advice you'd
sort of leave the, the audience with, you

552
00:31:27.925 --> 00:31:28.925
<v Speaker 2>know, on, on that topic?

553
00:31:29.244 --> 00:31:29.364
<v Speaker 2>Yeah,

554
00:31:29.484 --> 00:31:32.604
<v Dan Reid>so I, I think not capturing knowledge is,
is difficult, right?

555
00:31:32.645 --> 00:31:38.165
<v Dan Reid>And you know, we've touched on that, but I
think the hardest, hardest thing is just

556
00:31:38.165 --> 00:31:38.844
<v Dan Reid>to get started.

557
00:31:39.005 --> 00:31:43.724
<v Dan Reid>So I would just start capturing it and
obviously as you go along, you iterate and

558
00:31:43.724 --> 00:31:45.625
<v Dan Reid>you improve it and you get value from it.

559
00:31:46.069 --> 00:31:50.630
<v Dan Reid>Um, but I, I think often when you look at
sort of, you know, data related problems,

560
00:31:50.719 --> 00:31:56.569
<v Dan Reid>um, people have lots of data most of the
time, they just don't get value from it.

561
00:31:56.839 --> 00:32:00.310
<v Dan Reid>And it's very much like we, you know, we
talk about time series in the past and that

562
00:32:00.310 --> 00:32:06.150
<v Dan Reid>actually many, um, organizations we work
with, they don't have particular sensors,

563
00:32:06.160 --> 00:32:07.849
<v Dan Reid>um, set up potentially.

564
00:32:08.170 --> 00:32:12.270
<v Dan Reid>But they still have a load of time series
data and we've been able to unlock that,

565
00:32:12.300 --> 00:32:16.650
<v Dan Reid>the value in that data by, you know,
pulling information from control systems and

566
00:32:16.650 --> 00:32:19.319
<v Dan Reid>PLCs and providing predictive maintenance
on top.

567
00:32:19.729 --> 00:32:23.660
<v Dan Reid>And I think it's very similar to this
knowledge topic, although organizations, you

568
00:32:23.660 --> 00:32:26.719
<v Dan Reid>know, listeners might be thinking actually
I don't have much knowledge.

569
00:32:27.189 --> 00:32:30.270
<v Dan Reid>Um, a lot of the time it's just that that
knowledge is fragmented.

570
00:32:30.349 --> 00:32:31.630
<v Dan Reid>It's not centralized.

571
00:32:31.910 --> 00:32:35.030
<v Dan Reid>It's not available and you're not making,
you're not getting value from it.

572
00:32:35.189 --> 00:32:39.670
<v Dan Reid>So, you know, applications like Sensi,
being able to centralize that and surface

573
00:32:39.670 --> 00:32:44.329
<v Dan Reid>that knowledge when you need it may unlock
a lot of value in information that you

574
00:32:44.329 --> 00:32:46.770
<v Dan Reid>already have, you're just not benefiting
from.

575
00:32:47.140 --> 00:32:49.930
<v Dan Reid>Um, so that's what my, uh, my closing
statement.

576
00:32:51.560 --> 00:32:55.040
<v Speaker 2>There's, it's essentially, it's I don't,
we've talked about this again with like,

577
00:32:55.160 --> 00:32:56.290
<v Speaker 2>you know, time series data.

578
00:32:56.369 --> 00:32:59.239
<v Speaker 2>It's like don't underestimate the data you
have because I guess sometimes you have

579
00:32:59.239 --> 00:33:03.290
<v Speaker 2>customers who underestimate and maybe
sometimes overestimate, but don't just say,

580
00:33:03.449 --> 00:33:06.170
<v Speaker 2>oh, you know, I don't have anything to
share because we've talked about all these

581
00:33:06.170 --> 00:33:08.569
<v Speaker 2>different areas you can bring in
knowledge.

582
00:33:08.849 --> 00:33:12.489
<v Speaker 2>It's, you, you'll have useful stuff that
you can get you started on this journey.

583
00:33:12.640 --> 00:33:13.250
<v Speaker 2>Yeah, almost

584
00:33:13.250 --> 00:33:15.930
<v Dan Reid>certainly, and I, I would imagine most,
um, most manufacturers will.

585
00:33:16.069 --> 00:33:18.660
<v Dan Reid>Have, well, everyone will have manuals of
their, their assets.

586
00:33:19.229 --> 00:33:21.949
<v Dan Reid>Most will have standard operating
procedures.

587
00:33:22.229 --> 00:33:27.150
<v Dan Reid>Many will have maintenance systems, and
that in itself is loads of information that

588
00:33:27.150 --> 00:33:28.310
<v Dan Reid>could be put to good use.

589
00:33:28.670 --> 00:33:32.670
<v Dan Reid>And there's also this really lovely
feedback loop that when you start making use of

590
00:33:32.670 --> 00:33:35.109
<v Dan Reid>the knowledge, people see the benefit in
it, right?

591
00:33:35.229 --> 00:33:38.910
<v Dan Reid>So when, when there's an alert raised in
Sensai, and it tells them actually this is

592
00:33:38.910 --> 00:33:40.650
<v Dan Reid>what we think the root cause would be.

593
00:33:41.150 --> 00:33:43.729
<v Dan Reid>They see the value in what they've
provided in the past.

594
00:33:43.949 --> 00:33:47.829
<v Dan Reid>And guess what, they, they're motivated to
provide more knowledge in future and you

595
00:33:47.829 --> 00:33:51.869
<v Dan Reid>get into this nice loop where people see
the value in providing knowledge.

596
00:33:51.910 --> 00:33:55.550
<v Dan Reid>It's not just a document they create that
no one looks at, it gets surfaced again

597
00:33:55.550 --> 00:33:57.050
<v Dan Reid>and it helps people in future.

598
00:33:57.510 --> 00:34:02.180
<v Dan Reid>So if you can get into that, that routine
and that cadence, like, you know, it's,

599
00:34:02.189 --> 00:34:04.290
<v Dan Reid>uh, it's good from there on.

600
00:34:06.040 --> 00:34:07.479
<v Speaker 2>And, and that's your closing.

601
00:34:07.680 --> 00:34:08.409
<v Dan Reid>That's my closing statement.

602
00:34:10.080 --> 00:34:10.610
<v Speaker 2>No, absolutely.

603
00:34:10.760 --> 00:34:15.479
<v Speaker 2>And yeah, and so yeah, so I think that's a
really great way to sort of close out the

604
00:34:15.479 --> 00:34:15.840
<v Speaker 2>episode.

605
00:34:16.000 --> 00:34:19.439
<v Speaker 2>Um, again, if you want to find out more,
you know, visit the website.

606
00:34:19.639 --> 00:34:23.080
<v Speaker 2>Um, we've got lots of information on
there, um, about, about the co-pilot

607
00:34:23.080 --> 00:34:26.020
<v Speaker 2>specifically as well, as well as Sens I
predictctor maintenance of the platform.

608
00:34:26.629 --> 00:34:29.000
<v Speaker 2>Um, thank you, Dan, for joining us today
once again.

609
00:34:29.189 --> 00:34:29.409
<v Speaker 2>Thank you for having me.

610
00:34:29.419 --> 00:34:31.310
<v Speaker 2>Um, come back for a 4th time.

611
00:34:31.320 --> 00:34:31.330
<v Speaker 2>Yeah.

612
00:34:33.159 --> 00:34:36.679
<v Speaker 2>And I'd like to thank the audience out
there for listening and watching wherever you

613
00:34:36.679 --> 00:34:38.699
<v Speaker 2>are and see you on the next episode.

614
00:34:41.310 --> 00:34:46.239
<v Speaker 1>Please like and subscribe on all major
podcast channels and find out more about

615
00:34:46.239 --> 00:34:49.830
<v Speaker 1>Sensi predictive maintenance at
Siemens.com.
