WEBVTT
Language: en-US

1
00:00:05.400 --> 00:00:11.239
<v Speaker 1>Welcome to the trend detection podcast,
powered by Siemens, and now your host, Gugan

2
00:00:11.239 --> 00:00:11.350
<v Speaker 1>Sanga.

3
00:00:17.670 --> 00:00:23.000
<v Speaker 1>Welcome to the trend detection podcast,
powered by Siemens, and now your host, Gugan

4
00:00:23.000 --> 00:00:23.020
<v Speaker 1>Sanga.

5
00:00:24.139 --> 00:00:28.940
<v Speaker 1>In today's episode, we'll dive into how
that automotive manufacturer has achieved

6
00:00:28.940 --> 00:00:33.939
<v Speaker 1>its success, the technologies they've
implemented, and what other manufacturers can

7
00:00:33.939 --> 00:00:35.040
<v Speaker 1>learn from their journey.

8
00:00:35.860 --> 00:00:41.009
<v Speaker 1>To help us dig deeper into this success
story, I'm joined by some fantastic guests

9
00:00:41.220 --> 00:00:46.139
<v Speaker 1>who have valuable insights into the
processes and technologies that have driven this

10
00:00:46.139 --> 00:00:47.360
<v Speaker 1>plant's success.

11
00:00:47.779 --> 00:00:51.279
<v Speaker 1>Jan and Peter, thank you for joining me on
today's episode.

12
00:00:51.939 --> 00:00:56.220
<v Speaker 1>We're looking forward to hearing your
insights on what has made the plant a stand

13
00:00:56.220 --> 00:00:58.119
<v Speaker 1>out in the automotive industry.

14
00:00:58.700 --> 00:01:01.680
<v Speaker 1>Please take a moment to introduce
yourselves to our audience.

15
00:01:03.450 --> 00:01:05.889
<v Peter O'Hara>Yeah, hello folks, Peter O'Hara.

16
00:01:06.010 --> 00:01:12.650
<v Peter O'Hara>I am a customer success manager here at
Siemens working with uh Sensai

17
00:01:12.650 --> 00:01:13.589
<v Peter O'Hara>Predictive Maintenance.

18
00:01:13.690 --> 00:01:20.410
<v Peter O'Hara>I've been with Sensai for just over 7
years, and my aim in life is to help our

19
00:01:20.410 --> 00:01:24.349
<v Peter O'Hara>customers to use the application, add more
users.

20
00:01:24.760 --> 00:01:30.110
<v Peter O'Hara>Celebrate any wins, add more machines and
slowly take over their entire facility,

21
00:01:30.279 --> 00:01:34.639
<v Peter O'Hara>going site to site until everything the
customer does is monitored by Sensor.

22
00:01:35.370 --> 00:01:36.209
<v Peter O'Hara>Predictive maintenance.

23
00:01:36.569 --> 00:01:38.309
<v Jan Trnovec>My name is Jan Tarnovitz.

24
00:01:39.269 --> 00:01:45.389
<v Jan Trnovec>I'm a technician, a specialist in Ciprin,
which is a daughter company of Siemens.

25
00:01:46.360 --> 00:01:52.660
<v Jan Trnovec>We provide uh maintenance and engineering
services to our customers,

26
00:01:53.160 --> 00:01:56.000
<v Jan Trnovec>uh, mainly in automotive sector.

27
00:01:57.089 --> 00:01:58.519
<v Jan Trnovec>Um.

28
00:01:59.889 --> 00:02:06.730
<v Jan Trnovec>My work is mainly to reduce the downtimes,
to reduce the price

29
00:02:06.730 --> 00:02:12.539
<v Jan Trnovec>of the maintenance, and to uh achieve that
our customer.

30
00:02:13.449 --> 00:02:16.979
<v Jan Trnovec>can make as many products as he can.

31
00:02:21.830 --> 00:02:22.300
<v Speaker 1>Perfect.

32
00:02:22.449 --> 00:02:28.130
<v Speaker 1>Thank you both for your introductions and
I'm really interested in this topic, so

33
00:02:28.130 --> 00:02:33.119
<v Speaker 1>looking forward to hearing your insights
into the automotive industry and you know,

34
00:02:33.250 --> 00:02:37.270
<v Speaker 1>key successes around some of the asset
types.

35
00:02:37.759 --> 00:02:44.610
<v Speaker 1>Um, so what I wanted to do was to kick off
this um podcast and ask um a

36
00:02:44.610 --> 00:02:47.630
<v Speaker 1>first question around the asset type.

37
00:02:48.199 --> 00:02:54.279
<v Speaker 1>So I understand the asset type is a high
pressure pump injecting foam into car

38
00:02:54.279 --> 00:02:55.949
<v Speaker 1>bodies for soundproofing.

39
00:02:56.320 --> 00:03:00.160
<v Speaker 1>So if you could just provide a bit more
background just for our audience so they can

40
00:03:00.160 --> 00:03:01.500
<v Speaker 1>understand, that'd be great.

41
00:03:03.539 --> 00:03:09.289
<v Peter O'Hara>Yeah, start off just because I can do that
at the high level and Jan obviously knows

42
00:03:09.289 --> 00:03:11.240
<v Peter O'Hara>his equipment way better than me.

43
00:03:11.809 --> 00:03:18.449
<v Peter O'Hara>But it's an it's a really interesting use
case for us because I believe this was the

44
00:03:18.449 --> 00:03:24.889
<v Peter O'Hara>first time we've, uh, we've monitored a, a
pump injecting foam into the car body for

45
00:03:24.889 --> 00:03:25.610
<v Peter O'Hara>soundproofing.

46
00:03:26.059 --> 00:03:32.339
<v Peter O'Hara>And for one area when we are talking to
potentially new customers, one of their

47
00:03:32.339 --> 00:03:36.300
<v Peter O'Hara>first questions is always, have you
monitored this particular machine in this

48
00:03:36.300 --> 00:03:37.839
<v Peter O'Hara>particular environment before?

49
00:03:38.259 --> 00:03:42.789
<v Peter O'Hara>And I would always like to say, no, but
that's not necessarily important, unless of

50
00:03:42.789 --> 00:03:48.179
<v Peter O'Hara>course we have, because we monitor like
many thousands of machines across many

51
00:03:48.179 --> 00:03:49.779
<v Peter O'Hara>thousands um.

52
00:03:50.589 --> 00:03:55.360
<v Peter O'Hara>Uh, well, not many thousands, hundreds of
customers, and the main thing is we use

53
00:03:55.360 --> 00:04:00.240
<v Peter O'Hara>the same algorithms to monitor data that
means something to the condition of the

54
00:04:00.240 --> 00:04:00.380
<v Peter O'Hara>machine.

55
00:04:01.210 --> 00:04:08.089
<v Peter O'Hara>And those algorithms are self-learning,
and so it's about feeding them with the

56
00:04:08.089 --> 00:04:12.850
<v Peter O'Hara>correct data that will tell the
application something about the condition of the

57
00:04:12.850 --> 00:04:17.769
<v Peter O'Hara>machine and ideally removing the process
that the machine is undergoing, and that

58
00:04:17.769 --> 00:04:18.790
<v Peter O'Hara>can be a little bit tricky.

59
00:04:19.450 --> 00:04:22.170
<v Peter O'Hara>But getting something about the condition
of the machine.

60
00:04:22.410 --> 00:04:27.760
<v Peter O'Hara>So in this case, we're looking at a
process that has 3 stages to it.

61
00:04:27.809 --> 00:04:32.230
<v Peter O'Hara>We have a pumping mode, a circulation
mode, and then a pressurization stage.

62
00:04:32.690 --> 00:04:38.049
<v Peter O'Hara>And for the circulation, we are monitoring
uh flow rate, pressure, the RPM of the

63
00:04:38.049 --> 00:04:39.709
<v Peter O'Hara>pump, and temperature.

64
00:04:40.250 --> 00:04:45.529
<v Peter O'Hara>And then when we get to the pressurization
stage, we take the time to get to half

65
00:04:45.529 --> 00:04:47.489
<v Peter O'Hara>pressure and the time to get to full
pressure.

66
00:04:48.390 --> 00:04:54.940
<v Peter O'Hara>And one of the beauties of this
application is that Jan chose the

67
00:04:54.940 --> 00:05:00.290
<v Peter O'Hara>data himself and decided what to feed into
Sensi predictive maintenance, and then

68
00:05:00.989 --> 00:05:05.690
<v Peter O'Hara>very quickly within the deployment, we had
this first success that we are talking

69
00:05:05.690 --> 00:05:06.029
<v Peter O'Hara>about.

70
00:05:06.230 --> 00:05:12.750
<v Peter O'Hara>So perhaps I should hand over to Jan to
just tell me a few more things about this

71
00:05:12.750 --> 00:05:14.109
<v Peter O'Hara>um injection pump.

72
00:05:16.609 --> 00:05:22.890
<v Jan Trnovec>Yeah, as Peter already said, uh, there are
like high pressure pumps which are

73
00:05:22.890 --> 00:05:29.450
<v Jan Trnovec>feeding or pump pressurizing to medium to
mediums which are mixed together in jets

74
00:05:29.450 --> 00:05:31.730
<v Jan Trnovec>and then create the foam.

75
00:05:32.750 --> 00:05:39.709
<v Jan Trnovec>Uh, I think the most important thing, uh,
in this success was that we had

76
00:05:39.709 --> 00:05:46.670
<v Jan Trnovec>the right data which described, uh, Which
describes the machine,

77
00:05:47.250 --> 00:05:54.130
<v Jan Trnovec>uh, the, the failure mode of the machine
because we knew that the pumps

78
00:05:54.140 --> 00:06:00.929
<v Jan Trnovec>go wrong, uh, and these breakdowns cost us
or the customer some time

79
00:06:00.929 --> 00:06:02.570
<v Jan Trnovec>to replace the pump during production

80
00:06:02.570 --> 00:06:03.329
<v Jan Trnovec>time and so on.

81
00:06:04.230 --> 00:06:10.269
<v Jan Trnovec>And so therefore, we looked at the pumps
closely and found out that you can actually

82
00:06:10.269 --> 00:06:16.910
<v Jan Trnovec>see the, the washing out of the pump in
the way how the pump is building the

83
00:06:16.910 --> 00:06:17.309
<v Jan Trnovec>pressure.

84
00:06:17.470 --> 00:06:20.489
<v Jan Trnovec>So in the, pressurization stage.

85
00:06:21.670 --> 00:06:28.200
<v Jan Trnovec>And In this press pressurization stage,
uh, where the

86
00:06:28.200 --> 00:06:31.760
<v Jan Trnovec>circulation valve closed and the pump
starts to build pressure.

87
00:06:32.609 --> 00:06:39.500
<v Jan Trnovec>Uh, we can see that if the pressure isn't
rising linearly, but slows down as the

88
00:06:39.500 --> 00:06:46.260
<v Jan Trnovec>pressure rises, then we know that the pump
starts to wash out and which, which is

89
00:06:46.260 --> 00:06:47.779
<v Jan Trnovec>then followed by a breakdown.

90
00:06:48.959 --> 00:06:53.279
<v Jan Trnovec>Uh, so we simply started to monitor via
Sensi predictive maintenance.

91
00:06:53.320 --> 00:06:59.679
<v Jan Trnovec>The time, as Peter said, needed to reach
half of the set point and time needed to go

92
00:06:59.679 --> 00:07:02.739
<v Jan Trnovec>from from half of the set point to the set
point itself.

93
00:07:03.970 --> 00:07:08.170
<v Jan Trnovec>Uh, these two times we called half time
and full time.

94
00:07:09.000 --> 00:07:13.649
<v Jan Trnovec>And we programmed all the logic inside,
uh, in the PLC.

95
00:07:15.040 --> 00:07:21.059
<v Jan Trnovec>To, to get these two times regularly for
each pressurization stage,

96
00:07:21.510 --> 00:07:25.000
<v Jan Trnovec>uh, then we connect it to the sensei.

97
00:07:25.739 --> 00:07:27.059
<v Jan Trnovec>And voila.

98
00:07:28.320 --> 00:07:34.029
<v Jan Trnovec>After about 2 months, we get a warning
from Senzai.

99
00:07:35.000 --> 00:07:38.579
<v Jan Trnovec>Uh, that the, the time is changed.

100
00:07:39.160 --> 00:07:46.000
<v Jan Trnovec>The, the, the, the, uh, full time was
changed on one pump and we could see that,

101
00:07:46.440 --> 00:07:53.079
<v Jan Trnovec>uh, really the, the curvature of the, of
the rising of the pressure was

102
00:07:53.079 --> 00:07:53.720
<v Jan Trnovec>changed.

103
00:07:54.720 --> 00:07:57.279
<v Jan Trnovec>And we could see that the pump.

104
00:07:58.440 --> 00:08:03.859
<v Jan Trnovec>Is going to be wrong, like it's, it's,
it's going to break down.

105
00:08:04.440 --> 00:08:11.160
<v Jan Trnovec>Uh, what we didn't, we didn't, we didn't
know that, uh, how much

106
00:08:11.160 --> 00:08:12.540
<v Jan Trnovec>time we actually have.

107
00:08:13.470 --> 00:08:20.429
<v Jan Trnovec>Uh, so we saw that the bump is changing,
uh, but we didn't

108
00:08:20.429 --> 00:08:23.130
<v Jan Trnovec>know how, how long does it take.

109
00:08:24.040 --> 00:08:27.790
<v Jan Trnovec>To how much time we, we have to the actual
breakdown.

110
00:08:29.239 --> 00:08:33.159
<v Jan Trnovec>Uh, so in one case, we actually leave it.

111
00:08:34.520 --> 00:08:41.169
<v Jan Trnovec>And uh then we also knew the exact time
when the pump will go wrong, and when the,

112
00:08:41.250 --> 00:08:43.330
<v Jan Trnovec>when the pump will break down.

113
00:08:44.249 --> 00:08:50.648
<v Jan Trnovec>Uh, and now we like 100% know or 99% know
where is the

114
00:08:50.648 --> 00:08:57.448
<v Jan Trnovec>limit we can go, uh, to, to, to use the
pump as much

115
00:08:57.448 --> 00:08:58.609
<v Jan Trnovec>as, as possible.

116
00:09:00.330 --> 00:09:06.969
<v Jan Trnovec>And uh that's actually interesting because
uh we get a warning, uh, like a

117
00:09:06.969 --> 00:09:11.369
<v Jan Trnovec>month and a half before the actual
breakdown.

118
00:09:12.190 --> 00:09:13.109
<v Jan Trnovec>That's quite nice.

119
00:09:13.190 --> 00:09:19.390
<v Jan Trnovec>Like you have a month to plan the, the,
the, the, the replacing, the place of the

120
00:09:19.390 --> 00:09:19.770
<v Jan Trnovec>pump.

121
00:09:20.669 --> 00:09:22.510
<v Jan Trnovec>That was really, really nice.

122
00:09:22.750 --> 00:09:28.190
<v Peter O'Hara>And then, so you, you didn't add any more
sensors to, uh, to the pumps.

123
00:09:28.229 --> 00:09:31.159
<v Peter O'Hara>This is just harvesting data that exists.

124
00:09:31.710 --> 00:09:36.090
<v Jan Trnovec>Yes, of course, yeah, yeah, that was also
really nice that we didn't have to.

125
00:09:36.539 --> 00:09:42.299
<v Jan Trnovec>Mm, change something on the machine to add
sensors and so on.

126
00:09:42.700 --> 00:09:45.989
<v Jan Trnovec>We just used all the data we already had.

127
00:09:46.419 --> 00:09:51.659
<v Jan Trnovec>We used the data from frequency
converters, about the speed of the pump.

128
00:09:52.210 --> 00:09:58.619
<v Jan Trnovec>We used, uh, the pressures which are being
monitored, uh, also for the technology

129
00:09:58.619 --> 00:10:02.580
<v Jan Trnovec>purposes of the machine, for the
regulation and so on.

130
00:10:03.549 --> 00:10:05.909
<v Jan Trnovec>Uh, so we don't have to add anything.

131
00:10:06.030 --> 00:10:12.789
<v Jan Trnovec>We just collect it or, or, um, summarize
the data which we already had.

132
00:10:13.909 --> 00:10:20.390
<v Jan Trnovec>In the BLC and then we write uh the files
in the visualization and then send it.

133
00:10:21.770 --> 00:10:21.809
<v Jan Trnovec>To Sensi.

134
00:10:22.520 --> 00:10:28.960
<v Peter O'Hara>And so what process did you go through to
go from, obviously you had some data, but

135
00:10:28.960 --> 00:10:33.760
<v Peter O'Hara>then you'd have to look at it at fairly
high resolution to choose what to send to

136
00:10:33.760 --> 00:10:37.369
<v Peter O'Hara>Sensi, and then, so how did you go through
that process?

137
00:10:37.960 --> 00:10:44.479
<v Jan Trnovec>Yeah, we simply used an an PLC analyzer,
uh, to see in high resolution the

138
00:10:44.479 --> 00:10:45.059
<v Jan Trnovec>data.

139
00:10:45.659 --> 00:10:52.659
<v Jan Trnovec>Uh, where we, where we looked at, at the
behaving schemes of the, of

140
00:10:52.659 --> 00:10:59.659
<v Jan Trnovec>the machine, how the, how the pressure
rises, how, how it slow down, uh, how, how

141
00:10:59.659 --> 00:11:01.219
<v Jan Trnovec>the system works simply.

142
00:11:02.099 --> 00:11:09.059
<v Jan Trnovec>Uh, we tried to, to, uh, understand what's
going under the skin of the machine.

143
00:11:10.450 --> 00:11:16.179
<v Jan Trnovec>And then, uh, we found out that you can
simply, uh,

144
00:11:17.969 --> 00:11:19.390
<v Jan Trnovec>We, we, we, we.

145
00:11:20.590 --> 00:11:27.030
<v Jan Trnovec>We managed to find out the the simplest
way how to get the important data because

146
00:11:27.039 --> 00:11:32.979
<v Jan Trnovec>that's, I think from my point of view it's
the, it's the most important thing to

147
00:11:32.979 --> 00:11:39.979
<v Jan Trnovec>feed the right data because if you don't
have, if you have only some data, the

148
00:11:39.979 --> 00:11:46.710
<v Jan Trnovec>sensei uh is a perfect tool, but it, it
isn't some, some wizard

149
00:11:46.710 --> 00:11:49.909
<v Jan Trnovec>inside with a, with a glass.

150
00:11:50.750 --> 00:11:54.039
<v Jan Trnovec>Uh, sphere telling you now it will break.

151
00:11:54.549 --> 00:11:56.229
<v Jan Trnovec>It calculates, yeah.

152
00:11:57.469 --> 00:12:04.450
<v Peter O'Hara>And we often try to guide our, our
customers through a process of having

153
00:12:05.630 --> 00:12:10.690
<v Peter O'Hara>the correct people in the room at the
early stage of a deployment, so they know

154
00:12:11.070 --> 00:12:17.929
<v Peter O'Hara>maybe somebody who has uh ITOT type
experience who is able to get the data, but also

155
00:12:17.929 --> 00:12:20.679
<v Peter O'Hara>will have a maintainer there who will be
able to say.

156
00:12:21.539 --> 00:12:28.179
<v Peter O'Hara>what failure modes are important to the
customers and also some management to decide

157
00:12:28.179 --> 00:12:29.820
<v Peter O'Hara>which machines are important, etc.

158
00:12:29.979 --> 00:12:36.520
<v Peter O'Hara>It sounds to me that we have the perfect
person on the project because you can

159
00:12:37.299 --> 00:12:40.700
<v Peter O'Hara>analyze the data, you have the skills to
interact with the PLCs, you have the

160
00:12:40.700 --> 00:12:44.739
<v Peter O'Hara>skills, the analytic skills to look at the
data to decide what is important for

161
00:12:44.739 --> 00:12:47.099
<v Peter O'Hara>these failure modes, and then to pull that
out.

162
00:12:47.219 --> 00:12:54.000
<v Peter O'Hara>So this was a Beautiful deployment, but
if, uh, if we could, um,

163
00:12:54.010 --> 00:12:58.750
<v Peter O'Hara>scale yams and put them in every factory,
I think we would have a much easier life.

164
00:12:59.450 --> 00:13:01.090
<v Jan Trnovec>Yeah, but it wasn't just me.

165
00:13:01.130 --> 00:13:04.460
<v Jan Trnovec>Like I've had a, I had a colleagues that
helped me, of course.

166
00:13:04.929 --> 00:13:11.349
<v Jan Trnovec>I had in particular one colleague, uh, a
technician, which uh

167
00:13:11.849 --> 00:13:17.210
<v Jan Trnovec>knew the machine really, really well, and
he could, he was able to tell me some

168
00:13:17.210 --> 00:13:21.559
<v Jan Trnovec>insight, uh, what Where can I look?

169
00:13:22.570 --> 00:13:24.609
<v Jan Trnovec>And then when I looked, I found it.

170
00:13:24.809 --> 00:13:30.809
<v Jan Trnovec>Yeah, that's, that's the always the thing
that, um, you need colleagues, you need,

171
00:13:30.880 --> 00:13:37.609
<v Jan Trnovec>you need the experience, uh, of probably
more years, uh, you need also the

172
00:13:37.609 --> 00:13:44.250
<v Jan Trnovec>experience of, of the team behind Sensai,
of course, uh, which you can tell, uh,

173
00:13:44.260 --> 00:13:49.820
<v Jan Trnovec>who, who can tell you what is the Sensai
able to do, how you can monitor this and

174
00:13:49.820 --> 00:13:50.179
<v Jan Trnovec>this.

175
00:13:51.059 --> 00:13:57.760
<v Jan Trnovec>And also Um, as you mentioned, you didn't
have any experience with, uh, for example,

176
00:13:57.909 --> 00:14:04.440
<v Jan Trnovec>foaming machine or, uh, this type of
machinery, but, uh, when I

177
00:14:04.440 --> 00:14:11.159
<v Jan Trnovec>reached to you, um, uh, with, with, with
particular, uh, types of

178
00:14:11.159 --> 00:14:18.119
<v Jan Trnovec>breakdowns and what type of data I have,
uh, you helped me on the, on the other

179
00:14:18.119 --> 00:14:18.380
<v Jan Trnovec>side.

180
00:14:19.619 --> 00:14:25.260
<v Jan Trnovec>OK, we can extract it in this and this way
so we can monitor it back then and

181
00:14:25.260 --> 00:14:27.479
<v Jan Trnovec>process it and at the sensi site.

182
00:14:28.739 --> 00:14:32.599
<v Jan Trnovec>So that's always, um, it's not a one
person.

183
00:14:32.859 --> 00:14:33.580
<v Jan Trnovec>Uh,

184
00:14:33.979 --> 00:14:38.219
<v Peter O'Hara>I will agree it's not one person, but it's
the skills you need to make a successful

185
00:14:38.219 --> 00:14:44.479
<v Peter O'Hara>project involve knowing the, the failure
modes that you want to capture, but then.

186
00:14:45.219 --> 00:14:47.039
<v Peter O'Hara>Having the ability to get the data.

187
00:14:47.309 --> 00:14:54.020
<v Peter O'Hara>Now sometimes that is pulling it from a
PLC, sometimes it's, I need to add an

188
00:14:54.020 --> 00:14:59.500
<v Peter O'Hara>additional sensor here and having that bit
of experience as well, but yeah, it's um

189
00:14:59.979 --> 00:15:02.119
<v Peter O'Hara>it's nice to see.

190
00:15:03.010 --> 00:15:08.909
<v Peter O'Hara>Uh, a Siemens colleague, you know, fully
embracing the, the process of getting good

191
00:15:09.210 --> 00:15:13.859
<v Peter O'Hara>condition monitoring data into the
application and then letting the application kind

192
00:15:13.859 --> 00:15:20.809
<v Peter O'Hara>of run with it, and then I think with
early day success in a project, we

193
00:15:20.809 --> 00:15:25.669
<v Peter O'Hara>always need a couple of pieces of luck, so
one is, you know, an early.

194
00:15:26.669 --> 00:15:31.500
<v Peter O'Hara>Machine failure, obviously, you never want
the customers to have machine failures,

195
00:15:31.869 --> 00:15:33.409
<v Peter O'Hara>but when you look at.

196
00:15:34.299 --> 00:15:40.739
<v Peter O'Hara>The likelihood of a project succeeding,
you need to have machines, uh, enough of

197
00:15:40.739 --> 00:15:45.099
<v Peter O'Hara>them, so the chances of one of them
failing early on, uh, within the first couple of

198
00:15:45.099 --> 00:15:50.539
<v Peter O'Hara>months, um, that certainly helps, because
if everything runs as normal for quite a

199
00:15:50.539 --> 00:15:53.820
<v Peter O'Hara>while, people are like, yeah, well, why
are we monitoring this?

200
00:15:53.900 --> 00:15:57.280
<v Peter O'Hara>And then as soon as things start to
happen, then you can.

201
00:15:57.840 --> 00:16:01.760
<v Peter O'Hara>Uh, help people to see that actually,
yeah, this is, this is good, this could be

202
00:16:01.760 --> 00:16:05.700
<v Peter O'Hara>successful, and then you get them to
think, well, where else do we have a problem?

203
00:16:06.239 --> 00:16:10.280
<v Peter O'Hara>Where else can I harvest data or where can
I put some sensors, and then pull that

204
00:16:10.280 --> 00:16:12.659
<v Peter O'Hara>equipment into the application as well.

205
00:16:14.520 --> 00:16:15.299
<v Jan Trnovec>Yeah, exactly.

206
00:16:15.659 --> 00:16:22.390
<v Jan Trnovec>Like it doesn't make sense to, to, to
monitor a perfectly well and working uh

207
00:16:22.390 --> 00:16:24.659
<v Jan Trnovec>machine which doesn't have any breakdowns.

208
00:16:24.710 --> 00:16:28.359
<v Jan Trnovec>Like you can monitor it for years and
nothing happens.

209
00:16:28.869 --> 00:16:34.530
<v Jan Trnovec>Like that's, that's the also, that's also
an important thing to monitor things which

210
00:16:35.309 --> 00:16:40.760
<v Jan Trnovec>do break down and uh which are also
important because.

211
00:16:41.210 --> 00:16:47.690
<v Jan Trnovec>Uh, every, like in every facility, in
every, every plant, there are

212
00:16:47.690 --> 00:16:54.039
<v Jan Trnovec>machines which aren't important, and they
do break down by, but you just

213
00:16:55.090 --> 00:16:56.909
<v Jan Trnovec>Um, somehow no.

214
00:16:57.890 --> 00:16:59.250
<v Jan Trnovec>You do not care.

215
00:16:59.409 --> 00:17:01.739
<v Jan Trnovec>Like they are not important.

216
00:17:01.969 --> 00:17:07.290
<v Jan Trnovec>And then there are machines which are
really important because they stop the line,

217
00:17:07.369 --> 00:17:08.010
<v Jan Trnovec>for example.

218
00:17:08.779 --> 00:17:11.229
<v Jan Trnovec>Uh, and these are the important ones.

219
00:17:11.348 --> 00:17:16.659
<v Jan Trnovec>These are the bottlenecks which you have
to analyze also from the management, uh,

220
00:17:16.668 --> 00:17:19.509
<v Jan Trnovec>management, uh, uh, level.

221
00:17:20.250 --> 00:17:27.160
<v Jan Trnovec>Uh, and these need to inspect, needs to be
inspected properly and

222
00:17:27.380 --> 00:17:29.420
<v Jan Trnovec>if possible, then.

223
00:17:30.250 --> 00:17:33.810
<v Jan Trnovec>Um, also monitored via.

224
00:17:34.810 --> 00:17:36.609
<v Jan Trnovec>There are possible solutions.

225
00:17:37.530 --> 00:17:44.089
<v Jan Trnovec>Of sensor which is one of them and um from
my point of

226
00:17:44.089 --> 00:17:46.310
<v Jan Trnovec>view one of the best ones.

227
00:17:48.109 --> 00:17:53.089
<v Speaker 1>Um, that was really useful information
just to understand background around this

228
00:17:53.089 --> 00:17:54.060
<v Speaker 1>asset type.

229
00:17:54.500 --> 00:17:58.819
<v Speaker 1>And I just wanted to move into the next
question just around, um, you know, what

230
00:17:58.819 --> 00:18:05.020
<v Speaker 1>role does the foam injection pump play in
the automotive process.

231
00:18:08.150 --> 00:18:14.300
<v Jan Trnovec>Yeah, um, it's used for dampening noise,
noises or noise,

232
00:18:15.469 --> 00:18:17.650
<v Jan Trnovec>noise noise cancellation simply.

233
00:18:18.310 --> 00:18:24.229
<v Jan Trnovec>So the foaming is injected uh in uh
pillars of the vehicle.

234
00:18:25.209 --> 00:18:31.719
<v Jan Trnovec>Uh, for example, into A pillar or B pillar
in the pillars above the,

235
00:18:32.280 --> 00:18:34.739
<v Jan Trnovec>of the, uh, of the axels and so on.

236
00:18:35.780 --> 00:18:37.050
<v Jan Trnovec>To cancel noise.

237
00:18:38.189 --> 00:18:43.500
<v Jan Trnovec>So, and, um, one has to, has to mention
also that not every vehicle

238
00:18:44.900 --> 00:18:47.430
<v Jan Trnovec>does have the pillars formed.

239
00:18:48.069 --> 00:18:53.609
<v Jan Trnovec>Um It's not like the standard feature.

240
00:18:54.439 --> 00:19:00.420
<v Jan Trnovec>Uh, so that's also the, that's also
possible, the possibly the reason why,

241
00:19:00.760 --> 00:19:04.319
<v Jan Trnovec>uh, since I hadn't yet.

242
00:19:05.030 --> 00:19:11.310
<v Jan Trnovec>Uh, the, the experience with such, such a
machinery because it's not absolutely

243
00:19:11.310 --> 00:19:14.439
<v Jan Trnovec>common machine in automotive, uh, world.

244
00:19:15.349 --> 00:19:19.430
<v Jan Trnovec>But, uh, some, uh, customers use it.

245
00:19:20.660 --> 00:19:27.579
<v Jan Trnovec>Uh, and, uh, what I see, uh, it, it's used
more and more

246
00:19:27.579 --> 00:19:28.060
<v Jan Trnovec>often.

247
00:19:29.319 --> 00:19:32.640
<v Jan Trnovec>Exactly, because of noise cancellation,
it's quite a.

248
00:19:33.420 --> 00:19:35.930
<v Jan Trnovec>Good way to for for noise counseling.

249
00:19:36.949 --> 00:19:42.880
<v Jan Trnovec>Uh, and as the technology evolves, Uh.

250
00:19:45.319 --> 00:19:47.790
<v Jan Trnovec>It will be probably used more and more.

251
00:19:51.000 --> 00:19:51.800
<v Peter O'Hara>That sounds good.

252
00:19:54.920 --> 00:19:57.439
<v Peter O'Hara>So, One of the questions.

253
00:19:58.479 --> 00:20:00.510
<v Peter O'Hara>We had was.

254
00:20:01.819 --> 00:20:07.410
<v Peter O'Hara>How did sensor predictive maintenance, uh,
detect the trend and how reliable are the

255
00:20:07.410 --> 00:20:08.170
<v Peter O'Hara>alerts.

256
00:20:08.660 --> 00:20:14.660
<v Peter O'Hara>So, what I would say, I think I'm probably
repeating myself here, but we use a fleet

257
00:20:14.660 --> 00:20:19.900
<v Peter O'Hara>of algorithms that learn the normal
behavior for a machine, and I think that this

258
00:20:19.900 --> 00:20:20.359
<v Peter O'Hara>is.

259
00:20:20.890 --> 00:20:27.369
<v Peter O'Hara>Different to a lot of other products out
there, so we don't have people behind the

260
00:20:27.369 --> 00:20:33.530
<v Peter O'Hara>scenes making models for all the different
machine types that we, that we monitor,

261
00:20:33.930 --> 00:20:34.290
<v Peter O'Hara>so.

262
00:20:34.930 --> 00:20:41.500
<v Peter O'Hara>The early days learning is all
unsupervised, and what I found interesting in this

263
00:20:41.500 --> 00:20:47.500
<v Peter O'Hara>case is that the data for, I was looking
at this, yeah, yesterday or the day before,

264
00:20:47.579 --> 00:20:53.540
<v Peter O'Hara>you know, the data for this machine, it
stayed stable for about 4 days, and then it

265
00:20:53.540 --> 00:20:57.439
<v Peter O'Hara>started to trend up, and then as you said,
since I alerted the user.

266
00:20:57.859 --> 00:21:02.739
<v Peter O'Hara>Uh, after, uh, about a month that this
data was trending up, but I, I thought that

267
00:21:02.739 --> 00:21:03.420
<v Peter O'Hara>was quite interesting.

268
00:21:03.500 --> 00:21:09.979
<v Peter O'Hara>People often ask me, you know, how long is
the learning process, and the short

269
00:21:09.979 --> 00:21:12.479
<v Peter O'Hara>answer is it depends on the type of.

270
00:21:13.250 --> 00:21:17.979
<v Peter O'Hara>Algorithm alert that we're using, but
usually we quote about 48 hours, and I found

271
00:21:17.979 --> 00:21:22.630
<v Peter O'Hara>it interesting that, like I say, with
this, I could see the data was nice and stable

272
00:21:22.630 --> 00:21:27.660
<v Peter O'Hara>for the 1st 4 days and then just slowly
started to trend up, so I, I thought that

273
00:21:27.660 --> 00:21:29.520
<v Peter O'Hara>was quite interesting.

274
00:21:33.339 --> 00:21:37.939
<v Peter O'Hara>And then one of the other ones, sorry Gag
don't know if I'm uh stealing your thunder

275
00:21:37.939 --> 00:21:44.219
<v Peter O'Hara>with this one, is the role of the
co-pilot, so we haven't really mentioned that yet,

276
00:21:44.459 --> 00:21:50.020
<v Peter O'Hara>but with the use of co-pilot in industrial
applications now.

277
00:21:50.770 --> 00:21:54.530
<v Peter O'Hara>Uh, the co-pilot did spot this failure
mode.

278
00:21:54.650 --> 00:21:56.790
<v Peter O'Hara>Uh, I wondered what you thought of that.

279
00:22:00.150 --> 00:22:06.489
<v Jan Trnovec>Yeah, um, thanks to the expertise we have
done in the first stage,

280
00:22:06.989 --> 00:22:13.609
<v Jan Trnovec>the copilot, copilot actually didn't tell
me, told me anything which I Uh,

281
00:22:13.680 --> 00:22:20.439
<v Jan Trnovec>haven't already knew, but the thing is
that, uh, after replacing the

282
00:22:20.439 --> 00:22:27.380
<v Jan Trnovec>pump and so on, we, uh, insert the, the
input of the, the, the all the

283
00:22:28.119 --> 00:22:30.050
<v Jan Trnovec>necessary data into Sensi,

284
00:22:30.400 --> 00:22:35.599
<v Jan Trnovec>and then on the next case, the co-pilot
already.

285
00:22:36.270 --> 00:22:43.189
<v Jan Trnovec>Uh, pointed out that uh it's probably a
uh, a pump

286
00:22:43.510 --> 00:22:47.750
<v Jan Trnovec>which has to be replaced, which was also
the right answer.

287
00:22:48.569 --> 00:22:54.670
<v Jan Trnovec>Uh, so whenever the the the system will be
used by someone who

288
00:22:55.089 --> 00:23:01.770
<v Jan Trnovec>doesn't have as much experience as, for
example, I have, the co-pilot can

289
00:23:01.770 --> 00:23:07.709
<v Jan Trnovec>direct him what to do, and I think that's
a really, really, really good thing that

290
00:23:08.050 --> 00:23:14.770
<v Jan Trnovec>it can use the information which is in put
it in and to show it to the

291
00:23:14.770 --> 00:23:16.290
<v Jan Trnovec>next person.

292
00:23:17.280 --> 00:23:23.380
<v Jan Trnovec>Who Doesn't have to, doesn't have to

293
00:23:24.000 --> 00:23:29.699
<v Jan Trnovec>have uh such deep level of the knowledge
of the machine.

294
00:23:30.599 --> 00:23:32.680
<v Speaker 1>Yeah, I, I agree with that, Jan.

295
00:23:32.719 --> 00:23:36.119
<v Speaker 1>I just wanted to highlight, I think that's
one of the key benefits with the

296
00:23:36.119 --> 00:23:36.969
<v Speaker 1>co-pilot.

297
00:23:37.239 --> 00:23:43.380
<v Speaker 1>Um, you know, you're able, it's designed
to take that knowledge and you don't

298
00:23:43.380 --> 00:23:48.109
<v Speaker 1>necessarily need to have a high level, um,
you know, amount of knowledge to

299
00:23:48.109 --> 00:23:48.619
<v Speaker 1>understand.

300
00:23:49.109 --> 00:23:52.869
<v Speaker 1>So it helps to support and just break that
information down.

301
00:23:53.020 --> 00:23:55.420
<v Speaker 1>So, um, sorry, I just wanted to jump in.

302
00:23:55.790 --> 00:23:55.939
<v Speaker 1>Yeah,

303
00:23:56.109 --> 00:24:01.859
<v Peter O'Hara>Especially when we're scaling, this is the
thing, and in, so I've been in this

304
00:24:01.859 --> 00:24:05.910
<v Peter O'Hara>industry for 7 years with uh set aside
predictive maintenance, but.

305
00:24:06.619 --> 00:24:11.939
<v Peter O'Hara>You, early days of the project, you are
working with knowledgeable people about

306
00:24:11.939 --> 00:24:15.839
<v Peter O'Hara>their machines, and then you go to scale,
and then sometimes you go to a different

307
00:24:16.300 --> 00:24:18.079
<v Peter O'Hara>plant or a different site and.

308
00:24:19.119 --> 00:24:25.250
<v Peter O'Hara>That knowledge can be hard to replicate,
and so the fact that the, the co-pilot will

309
00:24:25.250 --> 00:24:31.770
<v Peter O'Hara>tell you exactly what's going on, but
also, you have the ability to ask it the, the

310
00:24:31.770 --> 00:24:37.569
<v Peter O'Hara>silly questions because we have seen in
other projects in the past where the project

311
00:24:37.569 --> 00:24:42.209
<v Peter O'Hara>gets moved on to people who are perhaps
not as data literate and perhaps don't

312
00:24:42.209 --> 00:24:47.689
<v Peter O'Hara>understand the, uh, what this data means
for their machines as much as the original.

313
00:24:48.229 --> 00:24:54.660
<v Peter O'Hara>Uh, people on the project, so it's really
nice that you can, uh, you can interrogate

314
00:24:54.660 --> 00:25:01.069
<v Peter O'Hara>it, you can ask it questions, as I often
say, it takes my role, customer success

315
00:25:01.069 --> 00:25:04.660
<v Peter O'Hara>management, it takes me a while to build
rapport with people, and I know I've

316
00:25:04.660 --> 00:25:07.910
<v Peter O'Hara>cracked it when they turned around to me
and say, I don't understand what this

317
00:25:07.910 --> 00:25:13.349
<v Peter O'Hara>means, but people will ask the co-pilot
that sort of question way sooner because

318
00:25:13.349 --> 00:25:14.050
<v Peter O'Hara>people don't mind.

319
00:25:14.760 --> 00:25:20.439
<v Peter O'Hara>Um, you know, asking the co-pilot the
questions that they really want to ask, and

320
00:25:20.439 --> 00:25:22.760
<v Peter O'Hara>sometimes they just say, oh, I was just
testing it to see if it knows, and it's

321
00:25:22.760 --> 00:25:26.119
<v Peter O'Hara>like, yay, that's, that's what it's there
for, just ask it away.

322
00:25:26.199 --> 00:25:26.900
<v Peter O'Hara>What does this mean?

323
00:25:27.000 --> 00:25:27.949
<v Peter O'Hara>Why is it doing that?

324
00:25:28.280 --> 00:25:32.680
<v Peter O'Hara>I think that's a really beautiful step
forward, and it's great to see it out in

325
00:25:32.680 --> 00:25:33.020
<v Peter O'Hara>industry.

326
00:25:33.520 --> 00:25:36.180
<v Peter O'Hara>And, yeah, nice and quick in this case as
well.

327
00:25:36.969 --> 00:25:40.489
<v Speaker 1>Yeah, I was going to say as well, Peter,
just touching on that point, I think we had

328
00:25:40.489 --> 00:25:46.609
<v Speaker 1>a discussion maybe a couple of weeks ago,
but you know, around the co-pilot, people

329
00:25:46.609 --> 00:25:51.229
<v Speaker 1>generally seem to be more open to asking
questions.

330
00:25:51.689 --> 00:25:54.589
<v Speaker 1>And like I said, it's so intusive.

331
00:25:54.609 --> 00:25:59.969
<v Speaker 1>Like it can give you that information
without talking to Um, you know, a human as

332
00:25:59.969 --> 00:26:00.930
<v Speaker 1>such, if that makes sense.

333
00:26:01.010 --> 00:26:03.680
<v Speaker 1>If you need the answers there and then,
it's very quick.

334
00:26:04.250 --> 00:26:07.790
<v Speaker 1>So, um, definitely helps with efficiency.

335
00:26:08.170 --> 00:26:12.439
<v Peter O'Hara>And as Jan said, you know, he closed the
case and then the next time the co-pilot

336
00:26:12.439 --> 00:26:17.530
<v Peter O'Hara>sees something, it's feeding back the
words that Jan has used to the users, and so

337
00:26:17.530 --> 00:26:18.310
<v Peter O'Hara>you can.

338
00:26:19.290 --> 00:26:22.290
<v Peter O'Hara>You know, you spread that knowledge
throughout the organization because the co-pilot

339
00:26:22.290 --> 00:26:28.849
<v Peter O'Hara>will tell other people and uh it's, yeah,
it's a very powerful tool for

340
00:26:28.849 --> 00:26:34.410
<v Peter O'Hara>um upskilling the, the users and the
workforce and sharing knowledge.

341
00:26:35.229 --> 00:26:40.290
<v Speaker 1>So moving on to the next question, I'm
really interested to understand how critical

342
00:26:40.910 --> 00:26:44.890
<v Speaker 1>was the 40 day early warning in avoiding
downtime.

343
00:26:48.369 --> 00:26:54.550
<v Jan Trnovec>Um, normally, we need like, mm,

344
00:26:54.849 --> 00:27:01.510
<v Jan Trnovec>only days to warn us because, uh, when we
get a warning during

345
00:27:01.510 --> 00:27:08.510
<v Jan Trnovec>the week, we Prob, we always are able to,

346
00:27:08.660 --> 00:27:15.189
<v Jan Trnovec>to manage the, the replace or the, the
repair uh during the weekend when the

347
00:27:15.189 --> 00:27:16.170
<v Jan Trnovec>line stops.

348
00:27:17.079 --> 00:27:24.010
<v Jan Trnovec>Uh, the thing is that if you have such an,
an, a long, long

349
00:27:24.010 --> 00:27:30.540
<v Jan Trnovec>time period, you are, you, you are able
to, to manage

350
00:27:30.540 --> 00:27:37.040
<v Jan Trnovec>it, uh, not only that you repair it, uh,
as the machine, but

351
00:27:37.040 --> 00:27:42.040
<v Jan Trnovec>you get into the stage that you can manage
how many stocks.

352
00:27:42.430 --> 00:27:49.010
<v Jan Trnovec>Do you have because all the things you
need to replace or or the or the spare parts

353
00:27:49.270 --> 00:27:50.430
<v Jan Trnovec>you, you want to.

354
00:27:51.510 --> 00:27:58.150
<v Jan Trnovec>Uh, have to do replace at some breakdown
and so on in, in such a, such a, for such a

355
00:27:58.150 --> 00:28:03.849
<v Jan Trnovec>machinery, uh, you need a spare part for
it and you have to manage the spare part

356
00:28:03.849 --> 00:28:10.469
<v Jan Trnovec>and you have to decide how many parts do I
want in my warehouse and that's

357
00:28:10.469 --> 00:28:12.510
<v Jan Trnovec>also in a cost way.

358
00:28:12.989 --> 00:28:19.339
<v Jan Trnovec>Like, uh, you can choose, OK, we have, uh,
8, high pressure pumps, and we would like

359
00:28:19.339 --> 00:28:24.969
<v Jan Trnovec>to have 8, spare parts, spare high
pressure pumps in our warehouse.

360
00:28:25.939 --> 00:28:32.300
<v Jan Trnovec>But um it will be, it would be really,
really high cost of the warehouse,

361
00:28:32.699 --> 00:28:34.640
<v Jan Trnovec>which doesn't make sense, of course.

362
00:28:35.260 --> 00:28:41.880
<v Jan Trnovec>So if you get warning like months before
the actual breakdown, you are able

363
00:28:42.459 --> 00:28:49.199
<v Jan Trnovec>to lower down the spare parts costs in a
way that you

364
00:28:49.199 --> 00:28:50.390
<v Jan Trnovec>don't need.

365
00:28:51.199 --> 00:28:57.180
<v Jan Trnovec>For example, 3 or 2 or 3 spare pumps.

366
00:28:57.989 --> 00:29:00.229
<v Jan Trnovec>You just need one or two.

367
00:29:00.709 --> 00:29:07.089
<v Jan Trnovec>Like if one goes wrong, you, you send it
away to the, to the, to the repair and one

368
00:29:07.550 --> 00:29:14.050
<v Jan Trnovec>rests in the, in the warehouse for some
immediate repair or whatever you need

369
00:29:14.949 --> 00:29:16.829
<v Jan Trnovec>in case something really goes bad.

370
00:29:17.560 --> 00:29:23.890
<v Jan Trnovec>Uh, but if you use it, Then you have time
to contact the,

371
00:29:24.000 --> 00:29:26.969
<v Jan Trnovec>the, the manufacturer.

372
00:29:27.189 --> 00:29:31.170
<v Jan Trnovec>OK, I don't have any spare, spare parts in
more.

373
00:29:31.469 --> 00:29:38.380
<v Jan Trnovec>I just got a warning that in 1 month my
pump breaks down,

374
00:29:38.869 --> 00:29:39.599
<v Jan Trnovec>do break down.

375
00:29:39.790 --> 00:29:42.369
<v Jan Trnovec>So can you please send me, you have 2
weeks.

376
00:29:43.380 --> 00:29:48.819
<v Jan Trnovec>And that's a really good thing because you
then, you can then manage the, the, the

377
00:29:48.819 --> 00:29:51.459
<v Jan Trnovec>warehouse really, really good.

378
00:29:52.180 --> 00:29:52.770
<v Speaker 1>Yeah.

379
00:29:53.099 --> 00:29:56.900
<v Speaker 1>No, that's great, Jan, really great to
hear because it just reinforces the whole

380
00:29:56.900 --> 00:30:03.719
<v Speaker 1>reduction of costs, you know, improving
efficiency, um, having that

381
00:30:03.780 --> 00:30:08.579
<v Speaker 1>information sooner rather than later just
helps to plan accordingly.

382
00:30:08.660 --> 00:30:12.229
<v Speaker 1>So at least you know what actions need to
be taken.

383
00:30:12.800 --> 00:30:19.219
<v Peter O'Hara>Yeah, there's also, if, if you know that
you need to, you have a month to uh replace

384
00:30:19.219 --> 00:30:25.670
<v Peter O'Hara>a part, then you can also perhaps plan
that in, so it's not necessarily during the

385
00:30:25.670 --> 00:30:30.449
<v Peter O'Hara>overtime period, so your labor costs can
be reduced as well as your inventory costs,

386
00:30:30.800 --> 00:30:35.119
<v Peter O'Hara>and we know that uh a large proportion of
a.

387
00:30:35.520 --> 00:30:40.469
<v Peter O'Hara>A part's carbon footprint comes from the
transportation of it, so if you are

388
00:30:40.469 --> 00:30:45.729
<v Peter O'Hara>expediting a part because of a, a failure,
then that carbon footprint is about twice

389
00:30:45.729 --> 00:30:49.469
<v Peter O'Hara>what it is normally, so having a uh.

390
00:30:50.819 --> 00:30:55.900
<v Peter O'Hara>That a long lead time to knowing when you
have to maintain things is obviously much

391
00:30:55.900 --> 00:30:59.380
<v Peter O'Hara>better for many parts of the organization,
for the customer.

392
00:31:01.540 --> 00:31:02.089
<v Speaker 1>Yeah, definitely.

393
00:31:03.849 --> 00:31:09.359
<v Speaker 1>And um I know we've we've highlighted
maybe some of the, you know, key benefits as

394
00:31:09.359 --> 00:31:14.410
<v Speaker 1>such, um, but it would be really great
just to understand in terms of this success

395
00:31:14.410 --> 00:31:20.689
<v Speaker 1>story, if there's any um main, you know,
benefits which have been achieved.

396
00:31:21.089 --> 00:31:26.989
<v Speaker 1>And also on the flip side, just to see if
there were any unexpected challenges um

397
00:31:27.010 --> 00:31:28.229
<v Speaker 1>during this process.

398
00:31:29.239 --> 00:31:34.489
<v Jan Trnovec>Yeah, but the benefits are clear, I think,
uh, the downsides.

399
00:31:36.449 --> 00:31:41.329
<v Jan Trnovec>I Actually, can't think of any like.

400
00:31:45.880 --> 00:31:48.689
<v Jan Trnovec>We just, we just did things.

401
00:31:49.660 --> 00:31:55.569
<v Jan Trnovec>Which uh helped the machine goes a little
bit better than before.

402
00:31:56.609 --> 00:32:02.969
<v Jan Trnovec>Like, I, I, I don't see any downsides to
it, uh.

403
00:32:04.609 --> 00:32:11.390
<v Jan Trnovec>Regarding um Some, some, uh, troubles or,
or,

404
00:32:11.400 --> 00:32:12.410
<v Jan Trnovec>or, um.

405
00:32:13.660 --> 00:32:17.089
<v Jan Trnovec>Uh How was the word, sorry?

406
00:32:18.300 --> 00:32:24.060
<v Jan Trnovec>Used, uh, challenges, challenges, yeah,
yeah, of course, thank you regarding

407
00:32:24.060 --> 00:32:31.060
<v Jan Trnovec>challenges, uh, the challenge was to, to
get to the data, to get

408
00:32:31.060 --> 00:32:37.160
<v Jan Trnovec>the data into Sensi, uh, not in terms of,
of, uh,

409
00:32:37.790 --> 00:32:42.540
<v Jan Trnovec>of, um, or in terms of, of infrastructure.

410
00:32:42.920 --> 00:32:49.390
<v Jan Trnovec>Because the machinery is 20 years old or
15 years old approximately.

411
00:32:49.800 --> 00:32:55.520
<v Jan Trnovec>So it doesn't have the proper connection
and also the, the legal basis of sending

412
00:32:55.520 --> 00:33:02.479
<v Jan Trnovec>data outside of the plant uh was always a
problem to get all, to, to

413
00:33:02.479 --> 00:33:03.800
<v Jan Trnovec>get, to, to.

414
00:33:04.729 --> 00:33:11.579
<v Jan Trnovec>Um, go with the data all, uh, with all
through, through these firewalls and so

415
00:33:11.579 --> 00:33:11.859
<v Jan Trnovec>on.

416
00:33:13.150 --> 00:33:17.989
<v Jan Trnovec>So, uh, that's one thing which, which, um.

417
00:33:18.719 --> 00:33:24.800
<v Jan Trnovec>Is the main challenge to, to send the, the
data away and that's always a problem

418
00:33:24.800 --> 00:33:31.219
<v Jan Trnovec>because um the, the, the IT infrastructure

419
00:33:31.640 --> 00:33:34.420
<v Jan Trnovec>is um sometimes.

420
00:33:35.630 --> 00:33:36.910
<v Jan Trnovec>Uh, not.

421
00:33:38.260 --> 00:33:42.739
<v Jan Trnovec>Um, Is uh.

422
00:33:44.339 --> 00:33:44.989
<v Jan Trnovec>I have to say it.

423
00:33:46.709 --> 00:33:53.469
<v Peter O'Hara>There's often challenges around data
security and what, uh, tools

424
00:33:53.469 --> 00:33:58.939
<v Peter O'Hara>are available to harness the data, capture
it and to transmit it as well, uh,

425
00:33:59.030 --> 00:33:59.750
<v Peter O'Hara>obviously.

426
00:34:00.430 --> 00:34:05.020
<v Peter O'Hara>The beauty of working for a company like
Siemens is that there are a number of

427
00:34:05.020 --> 00:34:10.189
<v Peter O'Hara>different technologies that allow the
customer to send data to the cloud and to

428
00:34:10.510 --> 00:34:16.550
<v Peter O'Hara>store it effectively, but we always have
to have agreement from the customer's IT

429
00:34:16.550 --> 00:34:22.409
<v Peter O'Hara>department about that cybersecurity, and
that can often, uh.

430
00:34:23.310 --> 00:34:28.790
<v Peter O'Hara>Slow down the initial process with a
customer because, of course, everybody wants to

431
00:34:28.790 --> 00:34:35.310
<v Peter O'Hara>make absolutely sure that their data's
secure and that they're not, um, leaving

432
00:34:35.310 --> 00:34:40.110
<v Peter O'Hara>any um intellectual property out there
that could be uh captured by anybody else,

433
00:34:40.229 --> 00:34:46.389
<v Peter O'Hara>so, we have plenty of technical data
sheets and security questionnaires and all of

434
00:34:46.389 --> 00:34:49.489
<v Peter O'Hara>that sort of thing that we can share with
potential customers.

435
00:34:50.610 --> 00:34:57.429
<v Jan Trnovec>Yeah, that's exactly the, the thing that
uh in, in our case we don't

436
00:34:57.429 --> 00:35:03.340
<v Jan Trnovec>even share like the technical data or
technological data of the machinery.

437
00:35:03.699 --> 00:35:09.159
<v Jan Trnovec>We don't need uh any data of the product
itself.

438
00:35:09.620 --> 00:35:10.770
<v Jan Trnovec>We just uh.

439
00:35:11.120 --> 00:35:17.919
<v Jan Trnovec>Uh, uh, monitor the physical data of the
machinery, not the like the technical

440
00:35:17.919 --> 00:35:19.590
<v Jan Trnovec>data, but not technological data.

441
00:35:21.479 --> 00:35:26.760
<v Peter O'Hara>So we know the pressure that the pump is
under at any one given moment, but we have

442
00:35:26.760 --> 00:35:30.199
<v Peter O'Hara>no idea what car model it is or anything
like

443
00:35:30.199 --> 00:35:30.610
<v Jan Trnovec>that.

444
00:35:31.129 --> 00:35:32.000
<v Jan Trnovec>Yeah, exactly, yeah.

445
00:35:35.639 --> 00:35:36.100
<v Speaker 1>Yeah.

446
00:35:36.919 --> 00:35:43.889
<v Speaker 1>And um Pete, from just uh just from your
perspective, um, how does this predictive

447
00:35:43.889 --> 00:35:47.199
<v Speaker 1>maintenance approach scale to other
equipment?

448
00:35:48.149 --> 00:35:48.159
<v Speaker 1>Well,

449
00:35:50.070 --> 00:35:51.929
<v Peter O'Hara>I would suggest easily.

450
00:35:52.139 --> 00:35:58.209
<v Peter O'Hara>Um, we've, with our other automotive
customers, we've, we've often started in

451
00:35:58.870 --> 00:36:03.870
<v Peter O'Hara>one shop on some critical pieces of
equipment, and then we've just scaled and scaled

452
00:36:03.870 --> 00:36:09.870
<v Peter O'Hara>and scaled, and now when you look, we have
many thousands of machines across

453
00:36:09.870 --> 00:36:10.669
<v Peter O'Hara>different sites.

454
00:36:10.790 --> 00:36:13.669
<v Peter O'Hara>We have all sorts of.

455
00:36:14.159 --> 00:36:18.560
<v Peter O'Hara>Different types of machinery will be in
the, the casting house at the beginning of

456
00:36:18.560 --> 00:36:24.169
<v Peter O'Hara>the process, we'll be in the um machine
shop, we'll be in the.

457
00:36:24.469 --> 00:36:29.929
<v Peter O'Hara>The Bo shop, as the, the cars, well, we've
got stamping, we'll be in the press shop,

458
00:36:30.370 --> 00:36:36.939
<v Peter O'Hara>you know, so we go through every, every
part of the process, um, right through to,

459
00:36:37.110 --> 00:36:41.189
<v Peter O'Hara>you know, from the paint shop, we'll be
monitoring the machines that's scrubbing

460
00:36:41.189 --> 00:36:45.350
<v Peter O'Hara>the, uh, the nasty chemicals out of the
air of the paint shop before releasing that

461
00:36:45.350 --> 00:36:48.889
<v Peter O'Hara>to the environment, so we can monitor.

462
00:36:49.500 --> 00:36:55.929
<v Peter O'Hara>Anything, the one thing we need is data
that means something about the condition of

463
00:36:55.929 --> 00:37:02.739
<v Peter O'Hara>the machine and ideally people to
understand what that data means to them, so we can

464
00:37:02.739 --> 00:37:09.489
<v Peter O'Hara>take some really nasty data if the
customer is happy to know,

465
00:37:09.540 --> 00:37:12.840
<v Peter O'Hara>OK, I don't want to know if it spikes all
the time, but if it's spiking.

466
00:37:13.620 --> 00:37:19.580
<v Peter O'Hara>You know, 20 times over the last 40 that I
want to know, or if it's, you know, if

467
00:37:19.580 --> 00:37:22.699
<v Peter O'Hara>this data's dropping, I don't care about
it, but if it goes really low, then I

468
00:37:22.699 --> 00:37:23.560
<v Peter O'Hara>really wanna know.

469
00:37:23.899 --> 00:37:30.659
<v Peter O'Hara>Um, so as long as they can tell us what
they want to see from the data, we

470
00:37:30.659 --> 00:37:36.500
<v Peter O'Hara>can make a few little tweaks, uh, and get
the system to tell them, uh, what it is

471
00:37:36.500 --> 00:37:39.040
<v Peter O'Hara>they've asked to know, and then scale.

472
00:37:39.139 --> 00:37:41.899
<v Peter O'Hara>So, we have some cust, well.

473
00:37:42.550 --> 00:37:48.270
<v Peter O'Hara>Some customers will have 1 or 2 sort of
critical machine types, and then many.

474
00:37:49.040 --> 00:37:55.979
<v Peter O'Hara>Uh, less critical machine types, and we
have things like temperature of electronic

475
00:37:56.030 --> 00:38:01.399
<v Peter O'Hara>cabinets, you know, we'll have hundreds of
those, and they just run weeks, weeks,

476
00:38:01.520 --> 00:38:05.840
<v Peter O'Hara>months, never, never really changes, you
know, you have little fluctuations, but

477
00:38:05.840 --> 00:38:09.760
<v Peter O'Hara>then as soon as somebody fixes the AC and
forgets to turn it back on, all the

478
00:38:09.760 --> 00:38:13.949
<v Peter O'Hara>cabinets heat up and sensor will go, whoa,
guys, you've got a problem over here.

479
00:38:14.320 --> 00:38:18.090
<v Peter O'Hara>So, um, it's not necessarily about.

480
00:38:18.770 --> 00:38:23.929
<v Peter O'Hara>The equipment that you need to look at
every week, but the equipment that will be

481
00:38:23.929 --> 00:38:25.989
<v Peter O'Hara>important to you if it stops working.

482
00:38:26.449 --> 00:38:26.689
<v Peter O'Hara>Yeah,

483
00:38:26.850 --> 00:38:31.570
<v Speaker 1>of course, and that just reinforces um
from a sense I predict a maintenance

484
00:38:31.570 --> 00:38:38.449
<v Speaker 1>perspective that there is a potential to
essentially monitor

485
00:38:38.449 --> 00:38:45.260
<v Speaker 1>any type of asset um within that um
industry, which is really interesting

486
00:38:45.610 --> 00:38:46.790
<v Speaker 1>and really good to know.

487
00:38:47.290 --> 00:38:47.850
<v Speaker 1>Um.

488
00:38:48.320 --> 00:38:52.679
<v Speaker 1>What lessons can be applied to similar
future issues?

489
00:38:54.540 --> 00:39:01.300
<v Peter O'Hara>Uh, well, I think we should, customers
should do what Jan did

490
00:39:01.300 --> 00:39:06.699
<v Peter O'Hara>really, so I understand why you want to
monitor a particular asset, that's always

491
00:39:06.699 --> 00:39:07.080
<v Peter O'Hara>important.

492
00:39:08.159 --> 00:39:14.919
<v Peter O'Hara>Don't just monitor something because it
has data, because chances are that data

493
00:39:14.919 --> 00:39:16.439
<v Peter O'Hara>won't necessarily tell you what.

494
00:39:17.290 --> 00:39:20.530
<v Peter O'Hara>Uh, you know, something about the
condition of the machine, so we always have to

495
00:39:20.530 --> 00:39:26.169
<v Peter O'Hara>look at the reason you're monitoring this
machine and the reasons why.

496
00:39:26.290 --> 00:39:32.129
<v Peter O'Hara>Now that can be preventing downtime, it
could be preventing over maintenance, uh, it

497
00:39:32.129 --> 00:39:35.929
<v Peter O'Hara>could be because of what we were saying
about the stock of equipment, that sort of

498
00:39:35.929 --> 00:39:40.929
<v Peter O'Hara>thing, so know why you're monitoring this
particular machine, look at the data that

499
00:39:40.929 --> 00:39:41.689
<v Peter O'Hara>you're monitoring.

500
00:39:42.469 --> 00:39:46.800
<v Peter O'Hara>Understand what that data means to you, to
highlight those particular issues.

501
00:39:47.620 --> 00:39:51.760
<v Peter O'Hara>I would Always want customers to.

502
00:39:52.659 --> 00:39:58.169
<v Peter O'Hara>Start at a reasonable scale because as I
say in other podcasts.

503
00:39:59.290 --> 00:40:05.889
<v Peter O'Hara>You, most customers intervene in between,
say, 2 and 12% of their equipment per

504
00:40:05.889 --> 00:40:06.360
<v Peter O'Hara>year.

505
00:40:06.689 --> 00:40:13.530
<v Peter O'Hara>So if you choose a small sample of
equipment and say, right, we're going to run a, a

506
00:40:13.530 --> 00:40:19.929
<v Peter O'Hara>3 month proof of concept on these 5, 10
motors, then what's the likelihood of

507
00:40:19.929 --> 00:40:26.350
<v Peter O'Hara>something failing in that time, and so
you're not giving yourself or the project.

508
00:40:26.800 --> 00:40:32.620
<v Peter O'Hara>Uh, a great possibility of success because
we, we want to try and take chance out of

509
00:40:32.620 --> 00:40:39.360
<v Peter O'Hara>this, so I would suggest a reasonable
scale, so you know, uh, that you

510
00:40:39.360 --> 00:40:41.100
<v Peter O'Hara>will see some issues with the machines.

511
00:40:41.399 --> 00:40:46.860
<v Peter O'Hara>Ideally you start with a more simple
machine type and.

512
00:40:47.699 --> 00:40:52.260
<v Peter O'Hara>Uh, that you can get data from, or you
start with a machine where you can add

513
00:40:52.260 --> 00:40:57.250
<v Peter O'Hara>sensors, so there are plenty of, um, you
know, vibration sensors and that sort of

514
00:40:57.250 --> 00:40:57.340
<v Peter O'Hara>thing.

515
00:40:57.419 --> 00:41:02.439
<v Peter O'Hara>We have plenty of, uh, customers who
monitor things with just current monitoring,

516
00:41:02.739 --> 00:41:09.659
<v Peter O'Hara>and particularly since the sensi
protective maintenance has become part

517
00:41:09.659 --> 00:41:13.399
<v Peter O'Hara>of Siemens, we are getting a lot more
information from.

518
00:41:13.840 --> 00:41:20.090
<v Peter O'Hara>Um, Siemens drives and PLCs and being able
to harvest, that is really good, and we

519
00:41:20.090 --> 00:41:26.469
<v Peter O'Hara>have lots of colleagues who can help us
with non-S Siemens PLCs, so if you've got

520
00:41:26.889 --> 00:41:32.629
<v Peter O'Hara>PLCs out there, you can get data to
Sensei, and you can start your journey, start

521
00:41:32.629 --> 00:41:37.050
<v Peter O'Hara>and then just never stop, just keep adding
machines, that's all I want you to do.

522
00:41:38.350 --> 00:41:40.489
<v Speaker 1>Mm Yeah, of course.

523
00:41:40.739 --> 00:41:46.459
<v Speaker 1>I mean it, it helps, doesn't it, in terms
of just from a machine perspective, asset

524
00:41:46.459 --> 00:41:51.669
<v Speaker 1>perspective, it's about fixing the issues.

525
00:41:52.159 --> 00:41:58.000
<v Speaker 1>Um, and I feel when it does come to a lot
of these different asset types, it's

526
00:41:58.000 --> 00:42:02.080
<v Speaker 1>having that education and knowledge to
understand what's going on with your

527
00:42:02.080 --> 00:42:02.610
<v Speaker 1>machines.

528
00:42:03.000 --> 00:42:07.919
<v Speaker 1>And that's where, um, you know, sensi
predictive maintenance really helps with that

529
00:42:07.919 --> 00:42:08.360
<v Speaker 1>process.

530
00:42:08.909 --> 00:42:12.699
<v Speaker 1>Um, and Jann, just touching on what you
mentioned before around this.

531
00:42:13.062 --> 00:42:17.982
<v Speaker 1>Collective approach, um, when it does come
to these type of projects, it's all of

532
00:42:17.982 --> 00:42:24.222
<v Speaker 1>these key stakeholders who come together
to support the process and assist with, you

533
00:42:24.222 --> 00:42:29.683
<v Speaker 1>know, the success cases of particular
assets within this industry.

534
00:42:30.562 --> 00:42:33.603
<v Speaker 1>Um, so just to wrap this discussion up.

535
00:42:33.805 --> 00:42:39.956
<v Speaker 1>I wanted to understand, you know, from a
Siemens perspective, Siemens provides

536
00:42:39.956 --> 00:42:42.865
<v Speaker 1>solutions to many other industries.

537
00:42:43.285 --> 00:42:47.855
<v Speaker 1>And it'll be really good just to
understand, just for the benefit of our audience,

538
00:42:47.926 --> 00:42:53.525
<v Speaker 1>um, what those solutions are and how we
can support when it comes to those

539
00:42:53.525 --> 00:42:54.345
<v Speaker 1>industries.

540
00:42:55.540 --> 00:43:02.219
<v Peter O'Hara>Yeah, so, obviously, I mean, Siemens
supply a lot of industrial equipment, so a lot

541
00:43:02.219 --> 00:43:04.379
<v Peter O'Hara>of the PLCs, a lot of the drives, etc.

542
00:43:04.500 --> 00:43:05.419
<v Peter O'Hara>a lot of the hardware.

543
00:43:05.909 --> 00:43:12.110
<v Peter O'Hara>And they are now linking that through
various different technologies, whether it's

544
00:43:12.110 --> 00:43:16.080
<v Peter O'Hara>Brownfield Connect, Industrial Edge,
Insights-Hub, etc.

545
00:43:16.239 --> 00:43:22.879
<v Peter O'Hara>to get the, the data that a product like
Sensive Predictive Maintenance needs to, to

546
00:43:22.879 --> 00:43:23.229
<v Peter O'Hara>work.

547
00:43:23.479 --> 00:43:26.080
<v Peter O'Hara>So they are helping, they are enabling.

548
00:43:26.520 --> 00:43:32.110
<v Peter O'Hara>Uh, us as a software to get the data from
our customers.

549
00:43:32.439 --> 00:43:38.639
<v Peter O'Hara>So that's the, the, I think the main area
where now we are part of Siemens, that's

550
00:43:38.639 --> 00:43:44.020
<v Peter O'Hara>really, really useful, and we support
customers in many different industries.

551
00:43:44.129 --> 00:43:47.600
<v Peter O'Hara>So if we focus on like with the
automotive.

552
00:43:48.389 --> 00:43:54.750
<v Peter O'Hara>We've got customers who will be uh you
know, mining the, the iron ore, etc.

553
00:43:54.870 --> 00:44:00.110
<v Peter O'Hara>that will then go and we have customers
who are smelting and rolling and producing

554
00:44:00.110 --> 00:44:02.310
<v Peter O'Hara>the metal, and then we have the.

555
00:44:03.139 --> 00:44:07.260
<v Peter O'Hara>Uh, the, the providers of parts for the
car industry, and then obviously the cars

556
00:44:07.260 --> 00:44:09.919
<v Peter O'Hara>out the end, and then for.

557
00:44:10.699 --> 00:44:15.300
<v Peter O'Hara>The customers of those who buy the cars,
you know, we are monitoring pet food, we're

558
00:44:15.300 --> 00:44:19.899
<v Peter O'Hara>monitoring ice cream, manufacturing, uh,
candies.

559
00:44:21.000 --> 00:44:27.110
<v Peter O'Hara>You know, different, uh, Uh, household
products and things like that, so, uh, yeah,

560
00:44:27.229 --> 00:44:31.909
<v Peter O'Hara>it, it doesn't really stop, we are not
limited to.

561
00:44:32.689 --> 00:44:39.530
<v Peter O'Hara>Any particular industries, any particular
sectors, we have our focus, but from

562
00:44:39.530 --> 00:44:41.030
<v Peter O'Hara>a, a Siemen's point of view.

563
00:44:42.250 --> 00:44:46.689
<v Peter O'Hara>I would say to anybody who is an account
manager out there, please talk to your

564
00:44:46.689 --> 00:44:52.429
<v Peter O'Hara>customers, if they have a problem that can
be solved by predictive maintenance, then

565
00:44:52.639 --> 00:44:55.360
<v Peter O'Hara>let's get talking because we would love to
help them.

566
00:44:55.850 --> 00:44:57.629
<v Peter O'Hara>They must have data out there.

567
00:44:57.810 --> 00:45:02.770
<v Peter O'Hara>If they don't have data out there, Siemens
provides the equipment to get that data.

568
00:45:03.229 --> 00:45:07.429
<v Peter O'Hara>And if they have machines or they can't
get the data from it, Siemens provide people

569
00:45:07.429 --> 00:45:08.979
<v Peter O'Hara>who can pull that data.

570
00:45:09.350 --> 00:45:13.389
<v Peter O'Hara>We have partners who will provide you with
certain sensing solutions, Siemens have

571
00:45:13.389 --> 00:45:19.510
<v Peter O'Hara>certain sensing solutions, so the world is
our oyster, and we are determined to

572
00:45:19.510 --> 00:45:21.250
<v Peter O'Hara>start monitoring every machine out there.

573
00:45:21.830 --> 00:45:22.070
<v Peter O'Hara>That's

574
00:45:22.070 --> 00:45:23.010
<v Speaker 1>what we want to hear.

575
00:45:24.229 --> 00:45:28.790
<v Speaker 1>Um, Jan, from your perspective, is there
anything that you wanted to add, um, just

576
00:45:28.790 --> 00:45:31.449
<v Speaker 1>as we wrap this uh episode up?

577
00:45:33.239 --> 00:45:35.229
<v Jan Trnovec>Mm, I don't know.

578
00:45:35.520 --> 00:45:38.879
<v Jan Trnovec>Peter stole, stole, stole all my words.

579
00:45:43.260 --> 00:45:49.459
<v Peter O'Hara>Well, I, I really do appreciate you
joining us, Jan, and I've loved interacting with

580
00:45:49.459 --> 00:45:50.239
<v Peter O'Hara>you on this project.

581
00:45:50.860 --> 00:45:56.899
<v Peter O'Hara>And, man, if, if I could clone you
throughout Siemens, I would have you in every

582
00:45:56.899 --> 00:46:02.340
<v Peter O'Hara>facility because, uh, the way you have
helped with this project is just fantastic.

583
00:46:02.459 --> 00:46:03.580
<v Peter O'Hara>So I really appreciate you.

584
00:46:04.860 --> 00:46:05.459
<v Jan Trnovec>Thank you.

585
00:46:06.459 --> 00:46:11.300
<v Jan Trnovec>I appreciate, I really appreciate the help
from you, Peter, and also in particular

586
00:46:11.300 --> 00:46:12.100
<v Jan Trnovec>from Jack Dutton.

587
00:46:12.340 --> 00:46:14.439
<v Jan Trnovec>It was really, really helpful.

588
00:46:15.060 --> 00:46:16.459
<v Peter O'Hara>Yeah, we all love Jack.

589
00:46:17.320 --> 00:46:18.439
<v Speaker 1>Yeah, definitely do.

590
00:46:19.239 --> 00:46:20.429
<v Speaker 1>He's a great asset, isn't he?

591
00:46:20.439 --> 00:46:22.449
<v Speaker 1>He's he's he's amazing at what he does.

592
00:46:22.879 --> 00:46:26.679
<v Speaker 1>But just wanted to say a big thank you to
both of you for joining me.

593
00:46:27.469 --> 00:46:32.199
<v Speaker 1>Really great valuable insights into the
processes and technology that have driven

594
00:46:32.199 --> 00:46:33.439
<v Speaker 1>this plant's success.

595
00:46:33.959 --> 00:46:40.320
<v Speaker 1>So, um, it's been a really great
discussion and, um, yeah, thank you once again.

596
00:46:40.879 --> 00:46:41.679
<v Peter O'Hara>Thanks for hosting.

597
00:46:43.050 --> 00:46:43.449
<v Peter O'Hara>Thank you.

598
00:46:51.159 --> 00:46:57.399
<v Speaker 1>Please like and subscribe to all major
podcast channels and find out more about

599
00:46:57.399 --> 00:47:00.600
<v Speaker 1>Sensi Predictive maintenance at
Siemens.com.
