WEBVTT
Language: en-US

1
00:00:01.860 --> 00:00:06.170
<v Speaker 1>Welcome podcast powered by Siemens, and
now

2
00:00:06.170 --> 00:00:06.699
<v Eric Rood>your host.

3
00:00:08.770 --> 00:00:09.369
<v Eric Rood>Good morning.

4
00:00:09.489 --> 00:00:10.029
<v Eric Rood>Good afternoon.

5
00:00:10.409 --> 00:00:14.050
<v Eric Rood>My name is Eric Rood, director with SIMMI,
and it is my honor to welcome you to

6
00:00:14.050 --> 00:00:18.409
<v Eric Rood>Siemens' webinar from Reactive to
Proactive Mastering Predictive Maintenance with

7
00:00:18.409 --> 00:00:19.010
<v Eric Rood>AI.

8
00:00:19.409 --> 00:00:22.770
<v Eric Rood>SEMI's committed to providing a platform
to bring technologies and efficiencies to

9
00:00:22.770 --> 00:00:26.950
<v Eric Rood>our members, and I hope you will find
Siemens' presentation valuable and actionable

10
00:00:26.950 --> 00:00:28.280
<v Eric Rood>for your own operations.

11
00:00:28.690 --> 00:00:33.130
<v Eric Rood>To begin, it is my pleasure to introduce
Ryan Mishin, global account manager for

12
00:00:33.130 --> 00:00:35.909
<v Eric Rood>Semiconductor at Siemens, and our
moderator today.

13
00:00:36.290 --> 00:00:37.759
<v Eric Rood>Ryan, could you please take it from here?

14
00:00:39.319 --> 00:00:40.750
<v Ryan>Appreciate the introduction, Eric.

15
00:00:41.209 --> 00:00:43.709
<v Ryan>Uh, as you mentioned, my name is Ryan, Ms.

16
00:00:43.729 --> 00:00:44.009
<v Ryan>Jan.

17
00:00:44.049 --> 00:00:50.240
<v Ryan>I work at Siemens, uh, in the
semiconductor space, and I'm happy to host this

18
00:00:50.240 --> 00:00:53.849
<v Ryan>conversation today with some of our, uh,
some of Siemens experts in the predictive

19
00:00:53.849 --> 00:00:59.810
<v Ryan>maintenance and AI, uh, technologies and
just to host this conversation about what

20
00:00:59.810 --> 00:01:05.250
<v Ryan>we've been seeing, ah, around
semiconductor with these predictive maintenance and AI

21
00:01:05.250 --> 00:01:05.970
<v Ryan>type strategies.

22
00:01:06.089 --> 00:01:12.459
<v Ryan>So, As the title suggests, this is really
a, it's a hybridization of predictive

23
00:01:12.459 --> 00:01:17.680
<v Ryan>maintenance, which is Not such a new
topic, ah, it's a little bit more mature.

24
00:01:17.779 --> 00:01:22.819
<v Ryan>Um, a lot of companies have a predictive
maintenance strategy or are leaning to

25
00:01:22.819 --> 00:01:25.519
<v Ryan>leaning that way, and we'll look at what
that looks like.

26
00:01:26.209 --> 00:01:32.309
<v Ryan>But then also with the, ah, emergence of
AI, ah, especially in the industrial field,

27
00:01:32.790 --> 00:01:38.220
<v Ryan>ah, how to couple that to really make a
more robust and impactful predictive

28
00:01:38.220 --> 00:01:38.980
<v Ryan>maintenance strategy.

29
00:01:39.080 --> 00:01:44.889
<v Ryan>So, Uh, we're happy to discuss this with
our experts and, and hopefully, ah, provide

30
00:01:44.889 --> 00:01:48.569
<v Ryan>some, some additional insights to, to
people out there in the semi industry.

31
00:01:50.949 --> 00:01:55.910
<v Ryan>So just a couple of table setting
statements um that we've heard from, number one,

32
00:01:55.949 --> 00:01:59.339
<v Ryan>there's a an an automotive manufacturer
that we've been working with for quite some

33
00:01:59.339 --> 00:01:59.849
<v Ryan>time.

34
00:02:00.209 --> 00:02:03.050
<v Ryan>Um, and I'm just gonna go ahead and read
it.

35
00:02:03.379 --> 00:02:09.789
<v Ryan>Ah, so we saved $45 million in labor and
unplanned downtime avoidance at a single

36
00:02:09.789 --> 00:02:11.669
<v Ryan>site since 2019.

37
00:02:11.830 --> 00:02:18.089
<v Ryan>So, cumulatively $45 million in, in this
labor and unplanned downtime.

38
00:02:18.600 --> 00:02:21.690
<v Ryan>Ah, just by implementing a predictive
maintenance strategy.

39
00:02:21.929 --> 00:02:28.139
<v Ryan>So, ah, by, by working with their
historical data and sensing and, and

40
00:02:28.320 --> 00:02:34.949
<v Ryan>tending to issues before they break, can
really have some, some huge

41
00:02:35.449 --> 00:02:38.690
<v Ryan>upsides in, in, ah, for your bottom line.

42
00:02:39.449 --> 00:02:44.089
<v Ryan>Ah, the next statement is really more of
an industry neutral statement, so it's from

43
00:02:44.089 --> 00:02:47.419
<v Ryan>a, uh, consultant that's, that we've
worked with called McKinsey Company.

44
00:02:47.490 --> 00:02:51.169
<v Ryan>I think a lot of people may be familiar
with them.

45
00:02:52.160 --> 00:02:57.800
<v Ryan>And again, this is more industry neutral,
um, but digital tools have dramatically

46
00:02:57.800 --> 00:02:59.750
<v Ryan>reduced unplanned outages.

47
00:03:00.240 --> 00:03:06.240
<v Ryan>Higher availability and a more efficient
workforce have increased profitability by 4

48
00:03:06.240 --> 00:03:08.539
<v Ryan>to 10% in some organizations.

49
00:03:09.179 --> 00:03:11.720
<v Ryan>So a lot of these digital tools are
available.

50
00:03:12.020 --> 00:03:18.119
<v Ryan>Um, it's really how you implement them to
best suit your application,

51
00:03:19.020 --> 00:03:22.360
<v Ryan>could yield that 4 to 10% um.

52
00:03:22.710 --> 00:03:25.289
<v Ryan>Ah, impact on your business.

53
00:03:25.389 --> 00:03:30.470
<v Ryan>So, again, that's more of the industry
neutral one, where that top one is, is really

54
00:03:30.470 --> 00:03:32.169
<v Ryan>pertinent to to automotive.

55
00:03:32.589 --> 00:03:37.350
<v Ryan>But you know, really there's, there's not
an industry out there that that couldn't

56
00:03:37.350 --> 00:03:43.429
<v Ryan>benefit from a predictive maintenance
strategy, most notably the semiconductor

57
00:03:43.429 --> 00:03:44.570
<v Ryan>industry which we're talking about today.

58
00:03:46.759 --> 00:03:50.020
<v Ryan>So just a quick introduction to the
speakers, panelists.

59
00:03:50.160 --> 00:03:54.720
<v Ryan>So I'll be moderating, ah, sharing some
kind of high-level information at the

60
00:03:54.720 --> 00:03:55.160
<v Ryan>beginning.

61
00:03:55.720 --> 00:03:58.750
<v Ryan>So, as Eric mentioned, my name's Ryan
Mischan.

62
00:03:58.880 --> 00:04:05.020
<v Ryan>I'm a global account manager at Siemens,
ah, servicing the semiconductor market.

63
00:04:05.520 --> 00:04:11.990
<v Ryan>I've been with Siemens almost 18 years, so
grew out of the automation drives, um, a

64
00:04:11.990 --> 00:04:17.778
<v Ryan>business, and now I support our whole
portfolio through our Through our software

65
00:04:17.778 --> 00:04:22.209
<v Ryan>portfolio, um, so yeah, uh, that's me.

66
00:04:22.459 --> 00:04:23.959
<v Ryan>I'll introduce Tim next.

67
00:04:24.750 --> 00:04:26.079
<v Tim Hunt>Hello, I'm Tim Hunt.

68
00:04:26.399 --> 00:04:30.679
<v Tim Hunt>Uh, I've been with Siemens, uh, Siemens
for 2.5 years now.

69
00:04:31.200 --> 00:04:34.040
<v Tim Hunt>Before that, I was with a company called
Logan Aluminum.

70
00:04:34.119 --> 00:04:36.600
<v Tim Hunt>It's an aluminum rolling mill out of
Russville, Kentucky.

71
00:04:36.880 --> 00:04:39.440
<v Tim Hunt>I was, uh, uh, I was there for 25 years.

72
00:04:39.480 --> 00:04:40.640
<v Tim Hunt>I retired from there.

73
00:04:41.000 --> 00:04:46.239
<v Tim Hunt>And uh I had the opportunity to really
sink my teeth into predictive maintenance for

74
00:04:46.239 --> 00:04:46.540
<v Tim Hunt>that.

75
00:04:46.839 --> 00:04:50.750
<v Tim Hunt>I was able to do it in one of the
departments, the finishing department, I was able

76
00:04:50.750 --> 00:04:55.559
<v Tim Hunt>to, I was there for 5 years and had so
much success by using all the predictive

77
00:04:55.559 --> 00:05:00.320
<v Tim Hunt>maintenance technologies that I was got
promoted to being able to do that for the

78
00:05:00.320 --> 00:05:04.059
<v Tim Hunt>plant and that's where I really grew a
passion for uh predictive maintenance.

79
00:05:04.589 --> 00:05:08.869
<v Tim Hunt>And that's why I was able to build
programs and, and, in the departments and train

80
00:05:09.109 --> 00:05:15.170
<v Tim Hunt>PDM coordinators and, and I just love this
journey of using all the uh PDM tools and

81
00:05:15.170 --> 00:05:19.350
<v Tim Hunt>then bringing on AI as being another part
of this PDM tool.

82
00:05:19.470 --> 00:05:23.630
<v Tim Hunt>So uh and I'm the uh it's a pleasure to be
here with y'all.

83
00:05:25.799 --> 00:05:27.869
<v Speaker 5>Um, good morning, everybody, and thank
you, Ryan.

84
00:05:27.950 --> 00:05:29.420
<v Speaker 5>Um, Todd Martin.

85
00:05:29.709 --> 00:05:35.820
<v Speaker 5>Uh, just recently joined Siemens myself,
but before Siemens, I, um, worked at Nissan

86
00:05:36.160 --> 00:05:41.000
<v Speaker 5>at their, um, automotive production
facility, and it was there that I began my

87
00:05:41.000 --> 00:05:42.940
<v Speaker 5>predictive maintenance, uh, journey.

88
00:05:43.559 --> 00:05:48.859
<v Speaker 5>Before that, I was heavily into automation
and machine building, um, industries.

89
00:05:49.440 --> 00:05:55.160
<v Speaker 5>So it was there at Nissan that I began,
um, Designing and implementing uh

90
00:05:55.160 --> 00:05:58.760
<v Speaker 5>condition-based monitoring systems and
then taking those and bringing on uh

91
00:05:58.760 --> 00:06:00.980
<v Speaker 5>predictive maintenance technologies.

92
00:06:01.399 --> 00:06:04.239
<v Speaker 5>Um, and we were quite successful at it.

93
00:06:04.279 --> 00:06:09.100
<v Speaker 5>So that's brought me to Siemens, um, to
continue my career.

94
00:06:10.670 --> 00:06:11.079
<v Ryan>Awesome.

95
00:06:11.239 --> 00:06:12.040
<v Ryan>Thanks, Todd.

96
00:06:13.540 --> 00:06:16.559
<v Ryan>So, today, today's agenda, what does it
look like?

97
00:06:17.019 --> 00:06:23.859
<v Ryan>Um, yeah, just, um, maybe just a, a more
of a, on a personal note, um, uh, a little

98
00:06:23.859 --> 00:06:26.779
<v Ryan>maintenance thing I've got going on at, at
my own property.

99
00:06:26.899 --> 00:06:30.820
<v Ryan>Uh, I have a gate latch out there that I
had to reinstall this morning.

100
00:06:31.320 --> 00:06:36.459
<v Ryan>Uh, we have some rather inclement weather
in Portland, Oregon, and, um, yeah, I'm

101
00:06:36.459 --> 00:06:41.540
<v Ryan>all of a sudden kind of wishing I had more
of a Proactive maintenance strategy for

102
00:06:41.540 --> 00:06:42.040
<v Ryan>it.

103
00:06:42.500 --> 00:06:47.579
<v Ryan>Uh, so, yeah, just on a personal note, um,
but yeah, what are we looking for?

104
00:06:47.660 --> 00:06:49.540
<v Ryan>What are we looking at in terms of the
agenda today?

105
00:06:49.619 --> 00:06:54.420
<v Ryan>So, we're gonna look at some of the
overall, so overarching challenges that the

106
00:06:54.420 --> 00:06:58.940
<v Ryan>semiconductor industry, uh, are facing
from, from our perspective.

107
00:06:59.100 --> 00:07:04.720
<v Ryan>Uh, we, we talk or we work regularly with
all sorts of, uh, end users out there.

108
00:07:05.320 --> 00:07:11.720
<v Ryan>And yeah, to no surprise, a lot of their
challenges ring true from customer to

109
00:07:11.720 --> 00:07:13.869
<v Ryan>customer, so we'll kind of share what
those look like.

110
00:07:14.790 --> 00:07:18.720
<v Ryan>Um, how can predictive maintenance address
those challenges?

111
00:07:18.959 --> 00:07:23.440
<v Ryan>Um, yeah, a lot of those challenges are
easily addressable or, or have some, some

112
00:07:23.440 --> 00:07:28.619
<v Ryan>big impacts if, if you implement a good,
ah, solid PDM strategy.

113
00:07:29.359 --> 00:07:32.070
<v Ryan>How is maintenance typically performed
today?

114
00:07:32.239 --> 00:07:36.239
<v Ryan>You know, we'll talk about, you know, the
classic, you know, the old school break

115
00:07:36.239 --> 00:07:37.559
<v Ryan>fix strategy.

116
00:07:38.130 --> 00:07:43.059
<v Ryan>Through a proactive maintenance strategy
all the way up through a predictive

117
00:07:43.059 --> 00:07:45.480
<v Ryan>maintenance, um, maintenance strategy.

118
00:07:46.209 --> 00:07:49.529
<v Ryan>So what is predictive maintenance and what
are some of the benefits that you could,

119
00:07:49.619 --> 00:07:52.399
<v Ryan>you could gain from implementing a
strategy like this.

120
00:07:52.980 --> 00:07:57.299
<v Ryan>And then I think a lot of what we're
interested in today is, is how can this

121
00:07:57.299 --> 00:08:00.019
<v Ryan>emerging AI artificial intelligence.

122
00:08:00.440 --> 00:08:05.769
<v Ryan>Um, technology really enhance those
benefits, um, that, that you may already be

123
00:08:05.769 --> 00:08:09.890
<v Ryan>using in a predictive maintenance
strategy, or how can you really accelerate that

124
00:08:09.890 --> 00:08:12.750
<v Ryan>predictive maintenance, um, that you
already have going.

125
00:08:12.970 --> 00:08:17.589
<v Ryan>Ah, implementation of PDM and then we'll
wrap it up with a little summary.

126
00:08:17.820 --> 00:08:22.970
<v Ryan>Uh, we'll get some time for Q&amp;A with,
with Tim and Todd, and, um, and we'll keep on

127
00:08:22.970 --> 00:08:23.309
<v Ryan>moving.

128
00:08:25.170 --> 00:08:30.880
<v Ryan>So, we have a little bit of an interactive
poll that we'd like to, to ask the

129
00:08:30.880 --> 00:08:31.989
<v Ryan>audience members.

130
00:08:33.010 --> 00:08:36.469
<v Ryan>And so Hole number one.

131
00:08:38.049 --> 00:08:44.068
<v Ryan>So what aspects of predictive maintenance
are you most curious about?

132
00:08:44.689 --> 00:08:44.968
<v Speaker 6>Let's

133
00:08:44.968 --> 00:08:48.549
<v Ryan>say #1, is it the implementation strategy?

134
00:08:49.088 --> 00:08:55.689
<v Ryan>Is it 2, cost savings that you can realize
by predictive maintenance, 3,

135
00:08:55.888 --> 00:09:01.028
<v Ryan>the use of AI, you know, the, the emerging
technology, or something else.

136
00:09:12.849 --> 00:09:16.590
<v Ryan>OK, I'm not, uh, not sure where the poll
questions are.

137
00:09:19.119 --> 00:09:20.179
<v Ryan>I have a chat here.

138
00:09:21.210 --> 00:09:22.950
<v Ryan>Does everyone see the results?

139
00:09:26.510 --> 00:09:29.159
<v Speaker 6>Or Ryan, please check the post section.

140
00:09:29.330 --> 00:09:30.669
<v Speaker 6>The results are coming in.

141
00:09:32.349 --> 00:09:33.109
<v Speaker 6>OK.

142
00:09:33.250 --> 00:09:34.650
<v Speaker 6>Um, let's see.

143
00:09:39.330 --> 00:09:44.559
<v Ryan>Imagine there's some real juicy statistics
out there for what we have coming in, but

144
00:09:44.559 --> 00:09:44.989
<v Ryan>uh.

145
00:09:47.969 --> 00:09:51.669
<v Ryan>I'm not sure if I can see the controls
here for poll showing.

146
00:09:52.669 --> 00:09:53.450
<v Ryan>Audience view.

147
00:09:56.869 --> 00:09:57.450
<v Speaker 6>OK.

148
00:09:59.719 --> 00:10:01.619
<v Ryan>At audio web sharing.

149
00:10:02.979 --> 00:10:03.599
<v Speaker 6>One second.

150
00:10:03.840 --> 00:10:06.080
<v Speaker 6>Let me share the results with you.

151
00:10:06.500 --> 00:10:06.940
<v Ryan>Oh, excellent.

152
00:10:07.020 --> 00:10:07.500
<v Ryan>Thank you.

153
00:10:13.969 --> 00:10:16.229
<v Ryan>Doing some live sharing of the results.

154
00:10:23.539 --> 00:10:24.119
<v Ryan>OK.

155
00:10:27.739 --> 00:10:28.760
<v Ryan>Scared attendance,

156
00:10:29.070 --> 00:10:29.840
<v Speaker 6>timer.

157
00:10:30.320 --> 00:10:30.919
<v Ryan>OK.

158
00:10:31.380 --> 00:10:32.119
<v Ryan>Questions.

159
00:10:33.590 --> 00:10:34.320
<v Ryan>Um,

160
00:10:35.070 --> 00:10:38.219
<v Speaker 6>If you could go to post, then you should
be able to see.

161
00:10:39.039 --> 00:10:40.299
<v Ryan>OK, here we go, there we go.

162
00:10:41.000 --> 00:10:45.599
<v Ryan>So, uh, to, to share this with the
audience, about half of, half of everyone said

163
00:10:45.599 --> 00:10:49.239
<v Ryan>the use of AI, that's the number one thing
that you're that they're, that you're all

164
00:10:49.239 --> 00:10:50.500
<v Ryan>here to, to learn about.

165
00:10:51.359 --> 00:10:55.880
<v Ryan>Second, ah, by a close margin was, uh, the
implementation strategies on how you

166
00:10:55.880 --> 00:11:02.130
<v Ryan>implement PDM and then, you know, roughly
11% said cost savings that you can realize

167
00:11:02.130 --> 00:11:03.760
<v Ryan>with with the predictive maintenance
strategy.

168
00:11:03.880 --> 00:11:08.880
<v Ryan>So, ah, we'll definitely hit heavy on the,
on the use of AI as we, as we talk

169
00:11:08.880 --> 00:11:11.799
<v Ryan>through some of these high points.

170
00:11:12.909 --> 00:11:13.520
<v Ryan>OK.

171
00:11:13.820 --> 00:11:14.390
<v Ryan>Thanks, Elle.

172
00:11:15.869 --> 00:11:22.599
<v Ryan>OK, so like we talked about, um, our teams
work with semiconductor manufacturers

173
00:11:22.599 --> 00:11:23.419
<v Ryan>across the board.

174
00:11:24.179 --> 00:11:30.700
<v Ryan>And to no surprise, a lot of these topics
should not be, um, should not be foreign,

175
00:11:30.780 --> 00:11:33.539
<v Ryan>I would, I would guess to a lot of the
people that are on this webinar.

176
00:11:33.659 --> 00:11:37.940
<v Ryan>So if you're on this webinar, you're
probably working directly with or around some

177
00:11:37.940 --> 00:11:44.859
<v Ryan>of these topics anyway, so, um, yeah,
let's see how How these might impact your

178
00:11:44.859 --> 00:11:45.320
<v Ryan>business.

179
00:11:45.460 --> 00:11:50.739
<v Ryan>So sustainability is, is one area that by
far, it's, it's one of the key pillars

180
00:11:50.739 --> 00:11:56.820
<v Ryan>behind all of our semiconductor
manufacturers, um, kind of their, their company

181
00:11:56.820 --> 00:11:57.489
<v Ryan>ethos.

182
00:11:57.609 --> 00:12:03.299
<v Ryan>So, that, that doesn't necessarily only
apply to environmental impacts, um, of

183
00:12:03.299 --> 00:12:08.619
<v Ryan>sustainability, but that also it means
sustainability of machines, um, the

184
00:12:08.619 --> 00:12:10.559
<v Ryan>sustainability of the business itself.

185
00:12:10.969 --> 00:12:16.169
<v Ryan>Um, but certainly you can implement a
predictive maintenance strategy to, to

186
00:12:16.169 --> 00:12:22.520
<v Ryan>optimize your energy bill, um, by, by
learning from historical data to, to, ah,

187
00:12:22.530 --> 00:12:24.669
<v Ryan>interact with machines in a certain way.

188
00:12:25.210 --> 00:12:28.559
<v Ryan>You can really, ah, lower your, your
energy bill.

189
00:12:28.929 --> 00:12:31.789
<v Ryan>So that's one use case of predictive
maintenance.

190
00:12:32.150 --> 00:12:37.049
<v Ryan>You know, of course, you know, by, by
properly taking care of machines and

191
00:12:37.049 --> 00:12:37.609
<v Ryan>processes.

192
00:12:38.210 --> 00:12:43.330
<v Ryan>You ultimately lower your waste of raw
goods and and lost products, so that's, you

193
00:12:43.330 --> 00:12:46.549
<v Ryan>know, inherently a sustainability topic
that we see all the time.

194
00:12:47.760 --> 00:12:52.219
<v Ryan>I think, you know, this, the next one goes
without without saying is is maximizing

195
00:12:52.679 --> 00:12:55.200
<v Ryan>your overall machine lifespan.

196
00:12:55.400 --> 00:12:59.539
<v Ryan>So that's probably the most understood
application of PDM.

197
00:13:00.280 --> 00:13:04.070
<v Ryan>And then of course, you know, we'd like to
help reduce the potential overall

198
00:13:04.070 --> 00:13:09.119
<v Ryan>environmental impacts by driving down the
waste and driving down energy consumption.

199
00:13:09.239 --> 00:13:15.780
<v Ryan>This as a, as a whole, ah, has, has a big,
has a big impact on sustainability.

200
00:13:17.440 --> 00:13:22.130
<v Ryan>Um, data usage and AI modeling is probably
another one that we're, like we talked

201
00:13:22.130 --> 00:13:24.580
<v Ryan>about, we're all here for, um.

202
00:13:25.520 --> 00:13:31.840
<v Ryan>So, your companies out there are sitting
on what I'm guessing is a huge stockpile of

203
00:13:31.840 --> 00:13:32.969
<v Ryan>data, OK.

204
00:13:33.080 --> 00:13:38.359
<v Ryan>So, you have historical data that's usable
in these predictive maintenance

205
00:13:38.359 --> 00:13:39.070
<v Ryan>strategies.

206
00:13:39.150 --> 00:13:45.260
<v Ryan>You can use the data to help model your
predictive maintenance by by inputting them

207
00:13:45.260 --> 00:13:46.500
<v Ryan>in a clever way.

208
00:13:46.830 --> 00:13:51.400
<v Ryan>Um, so, yeah, leveraging your data to to
better enhance your predictive maintenance

209
00:13:51.400 --> 00:13:52.140
<v Ryan>strategies.

210
00:13:53.500 --> 00:13:59.510
<v Ryan>I think a lot of people on the on the
webinar are probably well down the line of uh

211
00:13:59.510 --> 00:14:04.500
<v Ryan>being able to connect and talk to uh
different machines and different processes in

212
00:14:04.500 --> 00:14:06.880
<v Ryan>their in their fab or back end facility.

213
00:14:07.500 --> 00:14:13.640
<v Ryan>um, so that's kind of inherently
inherently critical for managing

214
00:14:14.460 --> 00:14:16.520
<v Ryan>sorry, I have a dog that's barking a lot.

215
00:14:17.299 --> 00:14:19.359
<v Ryan>It shouldn't last, um.

216
00:14:20.169 --> 00:14:25.289
<v Ryan>Yeah, being able to talk to these machines
is critical for um for for being able to

217
00:14:25.299 --> 00:14:28.669
<v Ryan>to manage them and send send schedules out
to them.

218
00:14:29.559 --> 00:14:36.039
<v Ryan>And, and then of course, implementing an
AI strategy based on that historical

219
00:14:36.039 --> 00:14:41.419
<v Ryan>data, that's sort of how we would enable
AI is, is a lot of that historical data and

220
00:14:41.419 --> 00:14:46.520
<v Ryan>being able to bring in ah other data
sources and data streams to make your processes

221
00:14:46.520 --> 00:14:47.320
<v Ryan>more intelligent.

222
00:14:49.190 --> 00:14:54.880
<v Ryan>Digitalization is, is probably another
more of like a company-wide ah strategy that

223
00:14:54.880 --> 00:14:58.760
<v Ryan>a lot of companies are using to transform
their business.

224
00:14:59.159 --> 00:15:05.630
<v Ryan>Ah, and we see this a lot in, in the way
that people are now starting to, ah, Use

225
00:15:05.630 --> 00:15:12.469
<v Ryan>digital tools to simulate different parts
of their processes or machines or figure

226
00:15:12.469 --> 00:15:15.250
<v Ryan>out how different parts of their company
work together.

227
00:15:15.630 --> 00:15:21.229
<v Ryan>Um, we're seeing a lot of this stuff, a
lot of digital tools implemented to, to, ah,

228
00:15:21.309 --> 00:15:22.690
<v Ryan>to start to address that stuff.

229
00:15:23.919 --> 00:15:30.659
<v Ryan>And Digital tools are also uh helping us
learn more about our

230
00:15:30.659 --> 00:15:36.580
<v Ryan>labor optimism or labor um and how we, how
we implement um uh or how we optimize

231
00:15:36.580 --> 00:15:37.599
<v Ryan>labor in the workforce.

232
00:15:38.710 --> 00:15:43.789
<v Ryan>So, before, ah, digital tools and and
predictive maintenance around those digital

233
00:15:43.789 --> 00:15:49.789
<v Ryan>tools were really just seen as the ability
to simulate and maybe um maybe optimize

234
00:15:49.789 --> 00:15:55.299
<v Ryan>throughput, but lately, what we're seeing
is, is using these digital tools,

235
00:15:55.549 --> 00:16:00.289
<v Ryan>predictive maintenance included, in ways
where we can optimize our labor.

236
00:16:00.710 --> 00:16:07.500
<v Ryan>So, um, One application would be ah the
scheduling of labor to

237
00:16:07.500 --> 00:16:12.520
<v Ryan>perform maintenance, um, you know, not
just sending people out on a, on a scheduled

238
00:16:12.520 --> 00:16:17.830
<v Ryan>time, but actually learning about machines
or processes in an intelligent enough way

239
00:16:18.289 --> 00:16:23.799
<v Ryan>to then say, hey, at this date, I need to
schedule this this resource out to make X,

240
00:16:23.809 --> 00:16:30.049
<v Ryan>Y, Z changes, um, really being much more
intelligent about how, how you deploy your

241
00:16:30.049 --> 00:16:30.349
<v Ryan>labor.

242
00:16:31.010 --> 00:16:37.140
<v Ryan>And then also you can use that as as a
training, um, ah, ah, a training platform as

243
00:16:37.140 --> 00:16:42.719
<v Ryan>well, um, using the predictive maintenance
platform to to gain information, to train

244
00:16:43.169 --> 00:16:49.099
<v Ryan>personnel, and of course, uh, inherently
making the um the operations more, more

245
00:16:49.099 --> 00:16:53.320
<v Ryan>safe and secure and, and easy to operate
for, for your employees.

246
00:16:56.659 --> 00:17:03.159
<v Ryan>So, at this point, we have a question that
I'd like to pose to Tim and Todd.

247
00:17:04.329 --> 00:17:07.349
<v Ryan>So, yeah, question for the for the
fellows.

248
00:17:07.810 --> 00:17:10.750
<v Ryan>How is maintenance typically performed
today?

249
00:17:11.670 --> 00:17:14.329
<v Ryan>And what are the challenges of doing it
this way?

250
00:17:17.119 --> 00:17:23.319
<v Tim Hunt>Well, you see a lot today and, and I was
surprised because the world I come from 30

251
00:17:23.319 --> 00:17:28.119
<v Tim Hunt>years ago we done reactive maintenance,
but we moved on to that from that to

252
00:17:28.119 --> 00:17:31.949
<v Tim Hunt>predictive maintenance and I thought
that's what everybody in the world was doing

253
00:17:31.949 --> 00:17:38.839
<v Tim Hunt>until I came to work for Siemens and uh
2.5 years into this uh being the customer

254
00:17:38.839 --> 00:17:41.800
<v Tim Hunt>success manager, there are so many folks
doing.

255
00:17:41.974 --> 00:17:42.895
<v Tim Hunt>Active maintenance.

256
00:17:42.974 --> 00:17:44.175
<v Tim Hunt>They're running into failure.

257
00:17:44.454 --> 00:17:48.334
<v Tim Hunt>They're run into this thing just melts
down, then they go out there and take a crane

258
00:17:48.334 --> 00:17:52.854
<v Tim Hunt>and pull it out and put something else in
and, and there's a better way to do it.

259
00:17:52.935 --> 00:17:54.494
<v Tim Hunt>We don't have to do that no more.

260
00:17:54.614 --> 00:17:57.895
<v Tim Hunt>That was good back in the 80s, but we
don't have to run like that no more.

261
00:17:58.214 --> 00:18:01.295
<v Tim Hunt>We can predict now uh uh and.

262
00:18:01.989 --> 00:18:04.989
<v Tim Hunt>You can predict 30 days, 60 days, 90 days.

263
00:18:05.150 --> 00:18:09.310
<v Tim Hunt>I mean you can depend on the line speed
and, and the pressures and torques and all,

264
00:18:09.589 --> 00:18:14.660
<v Tim Hunt>but I mean we can see way out in time to
give you plenty of time to order the parts,

265
00:18:14.739 --> 00:18:17.069
<v Tim Hunt>get the manpower and get everything
together.

266
00:18:17.349 --> 00:18:19.250
<v Tim Hunt>You don't have to tear all this.

267
00:18:19.474 --> 00:18:23.974
<v Tim Hunt>Equipment up anymore because secondary
failures is our biggest cost anyway when it

268
00:18:23.974 --> 00:18:24.614
<v Tim Hunt>fails.

269
00:18:24.935 --> 00:18:29.734
<v Tim Hunt>Um, this slide right here, I love talking
to this slide because it, it brings you a

270
00:18:29.734 --> 00:18:33.614
<v Tim Hunt>big picture because there might be some
folks here on the call uh that, that don't

271
00:18:33.614 --> 00:18:37.035
<v Tim Hunt>understand really how PM works, what the,
what the technologies are.

272
00:18:37.449 --> 00:18:42.089
<v Tim Hunt>Uh, you'll see this is the PDF curve, the
prevention to failure, prevention being on

273
00:18:42.089 --> 00:18:42.589
<v Tim Hunt>your left.

274
00:18:43.089 --> 00:18:47.170
<v Tim Hunt>Uh, this is when your machine is starting
to fail that goes into a failure mode like

275
00:18:47.170 --> 00:18:52.569
<v Tim Hunt>you got a bearing that creates a small on
the, uh, say the outer race and now it's

276
00:18:52.569 --> 00:18:53.810
<v Tim Hunt>going into a failure mode.

277
00:18:53.849 --> 00:18:55.170
<v Tim Hunt>That's your top left corner.

278
00:18:55.520 --> 00:19:02.089
<v Tim Hunt>Well, we can use, uh, ultrasound or we can
use, uh, uh, lubrication, uh, oil

279
00:19:02.089 --> 00:19:04.849
<v Tim Hunt>analysis to pick that up real early.

280
00:19:05.349 --> 00:19:11.040
<v Tim Hunt>But as it goes on, now, then we can use
vibration to pick up that failure and we're

281
00:19:11.040 --> 00:19:16.160
<v Tim Hunt>actually at that point, we're picking it
up with AI and as it trends on down and

282
00:19:16.160 --> 00:19:18.750
<v Tim Hunt>then you'll start getting the audible
noise.

283
00:19:18.839 --> 00:19:24.319
<v Tim Hunt>At this point, uh, before vibration, you
can't even see these faults with the human

284
00:19:24.319 --> 00:19:24.619
<v Tim Hunt>eye.

285
00:19:24.719 --> 00:19:26.550
<v Tim Hunt>I've split these bearings open before.

286
00:19:26.800 --> 00:19:28.170
<v Tim Hunt>Hey, there, there is no fault.

287
00:19:28.319 --> 00:19:31.839
<v Tim Hunt>Well, it's there in the oil analysis
because your iron counts is spiking.

288
00:19:32.290 --> 00:19:36.699
<v Tim Hunt>But as it continues to fail, then the
operators and maintenance will say, hey, I

289
00:19:36.699 --> 00:19:39.530
<v Tim Hunt>hear something, something don't sound
right on this line.

290
00:19:39.780 --> 00:19:43.339
<v Tim Hunt>Well, that's that bearing, the spa is
getting bigger and bigger and as the rolling

291
00:19:43.339 --> 00:19:46.800
<v Tim Hunt>elements roll over, it's creating the
impact with a frequency.

292
00:19:47.180 --> 00:19:53.099
<v Tim Hunt>Uh, then you'll bring in your
thermography, well, you'll see it then the temperature

293
00:19:53.099 --> 00:19:53.859
<v Tim Hunt>is rising.

294
00:19:54.239 --> 00:19:57.300
<v Tim Hunt>Well, then the next point is failure,
ultimate failure.

295
00:19:57.439 --> 00:19:58.959
<v Tim Hunt>This is where you melt the bearings.

296
00:19:59.000 --> 00:20:03.060
<v Tim Hunt>This is when you score the shaft, probably
destroy your fan housing.

297
00:20:03.430 --> 00:20:05.199
<v Tim Hunt>Uh, so there's a better way to do it.

298
00:20:05.280 --> 00:20:09.189
<v Tim Hunt>There, there's a system here and that we
can really make a difference and I get to

299
00:20:09.189 --> 00:20:13.400
<v Tim Hunt>work with customers every week and see how
we're making a difference with our, with

300
00:20:13.400 --> 00:20:14.000
<v Tim Hunt>our equipment.

301
00:20:17.630 --> 00:20:22.069
<v Ryan>So, so Tim, so then what's the, what's the
alternative then?

302
00:20:24.890 --> 00:20:30.280
<v Speaker 5>Well, I mean, Ryan, the alternative is,
is, is, and we've been doing this for a long

303
00:20:30.280 --> 00:20:36.079
<v Speaker 5>time, you know, reactive went from
reactive to condition monitoring and

304
00:20:36.079 --> 00:20:37.150
<v Speaker 5>condition-based maintenance.

305
00:20:37.569 --> 00:20:42.250
<v Speaker 5>We've all added technologies to our
equipment to measure key factors on that

306
00:20:42.250 --> 00:20:49.119
<v Speaker 5>equipment, temperature, motor current,
torque, um, You know, vibration,

307
00:20:50.300 --> 00:20:57.020
<v Speaker 5>time, you know, time to activation, things
like that, and those technologies were a

308
00:20:57.020 --> 00:21:02.900
<v Speaker 5>big, huge improvement in, you know,
predicting failures ahead of time, but not

309
00:21:02.900 --> 00:21:06.900
<v Speaker 5>scalable because you would have to know
what to set the limits that you were going

310
00:21:06.900 --> 00:21:07.459
<v Speaker 5>to react

311
00:21:07.459 --> 00:21:08.530
<v Speaker 5>to those conditions.

312
00:21:08.729 --> 00:21:13.819
<v Speaker 5>Um, that means If you've got a few assets
or a few pieces of equipment, you can keep

313
00:21:13.819 --> 00:21:18.199
<v Speaker 5>up with it, but as I said, not scalable to
thousands of assets on the floor.

314
00:21:18.780 --> 00:21:24.650
<v Speaker 5>So, That led to what do we, you know, how
do we scale and be able to monitor all our

315
00:21:24.650 --> 00:21:30.180
<v Speaker 5>pieces of equipment in the facility and
leading into predictive maintenance and, you

316
00:21:30.180 --> 00:21:32.670
<v Speaker 5>know, data analytics techniques and all
that.

317
00:21:32.969 --> 00:21:38.900
<v Speaker 5>And this predictive maintenance
technologies, very mature, um, you know, I myself

318
00:21:38.900 --> 00:21:45.400
<v Speaker 5>have used them for, you know, coming up
close on 7 years and with great success.

319
00:21:46.449 --> 00:21:51.900
<v Speaker 5>Um, So, of course, a lot of people here
are going to be familiar that yes, we have

320
00:21:51.900 --> 00:21:56.380
<v Speaker 5>these predictive maintenance technologies,
um, but what's the new and up and coming,

321
00:21:56.500 --> 00:21:58.900
<v Speaker 5>what are we, and that's, it's, that's AI.

322
00:21:59.780 --> 00:22:03.459
<v Speaker 5>Now I imagine a lot of people think, OK,
we're going to actually use the AI to look

323
00:22:03.459 --> 00:22:08.119
<v Speaker 5>at those data signatures um to predict the
failures.

324
00:22:08.699 --> 00:22:14.050
<v Speaker 5>In some cases, yes, but one of the biggest
exciting things about using AI in this is

325
00:22:14.390 --> 00:22:20.949
<v Speaker 5>the ability to AI to ingest not just those
um data signals and measurement

326
00:22:20.949 --> 00:22:26.349
<v Speaker 5>values, but also your contextual data that
you've been entering into your

327
00:22:26.349 --> 00:22:32.630
<v Speaker 5>maintenance monitoring systems, you know,
work histories, um, oh, we've seen this on

328
00:22:32.630 --> 00:22:36.969
<v Speaker 5>another one and we input this because Tim
said the, the bearing blew up.

329
00:22:37.510 --> 00:22:37.959
<v Speaker 5>um.

330
00:22:38.719 --> 00:22:45.599
<v Speaker 5>So now with, with the AI integrated with
PDM it is able to tap that reservoir

331
00:22:45.599 --> 00:22:50.760
<v Speaker 5>of knowledge that us as human beings have
put in and spent a lot of time, you know,

332
00:22:50.880 --> 00:22:51.680
<v Speaker 5>accumulating.

333
00:22:53.030 --> 00:22:59.170
<v Speaker 5>Another portion of this AI is now
conversational PDM.

334
00:22:59.430 --> 00:23:04.670
<v Speaker 5>And what is that is that rather than just
say, you know, we saw this X thing happen

335
00:23:04.670 --> 00:23:10.010
<v Speaker 5>on this piece of equipment and this, this
and this, it now starts speaking to us in

336
00:23:10.109 --> 00:23:13.750
<v Speaker 5>conversational tones that we've become
familiar with in other, um, you know,

337
00:23:14.699 --> 00:23:16.849
<v Speaker 5>conversational AI tools.

338
00:23:17.510 --> 00:23:24.359
<v Speaker 5>So it's, it really Expands the um the
group of people that can use

339
00:23:24.359 --> 00:23:27.199
<v Speaker 5>predictive maintenance to maintain their
equipment.

340
00:23:28.079 --> 00:23:32.760
<v Speaker 5>Even the advent for a long time, but you
still had to have your, um, your, your

341
00:23:32.760 --> 00:23:36.280
<v Speaker 5>subject matter experts, you know, people
who were familiar with the equipment that

342
00:23:36.280 --> 00:23:41.079
<v Speaker 5>could say, oh yes, this predictive is
pointed at something's going on here, but then

343
00:23:41.079 --> 00:23:42.020
<v Speaker 5>it was still.

344
00:23:42.890 --> 00:23:47.709
<v Speaker 5>Almost 100% the human being had to look at
what it was telling you, take that data.

345
00:23:48.660 --> 00:23:53.079
<v Speaker 5>Investigate, interpret, to get really to
the root issue.

346
00:23:53.699 --> 00:23:58.079
<v Speaker 5>Now with this conversational and
generational AI included with it.

347
00:24:00.420 --> 00:24:05.180
<v Speaker 5>You don't have to have that specific
subject matter expert as much.

348
00:24:05.300 --> 00:24:07.329
<v Speaker 5>I'm not say anymore, but as much.

349
00:24:07.619 --> 00:24:14.380
<v Speaker 5>And now you, you brought it out, um,
you've, you've maybe probably feed up, freed up

350
00:24:14.380 --> 00:24:19.439
<v Speaker 5>one of your top experts on your equipment
to probably do more value-added on work.

351
00:24:23.079 --> 00:24:23.869
<v Ryan>Nice, thanks, Todd.

352
00:24:23.959 --> 00:24:29.119
<v Ryan>So, so what would you say that some of
the, some of the key value statements, um,

353
00:24:29.380 --> 00:24:32.300
<v Ryan>the key benefits of PDM just in general
would be?

354
00:24:34.050 --> 00:24:39.250
<v Speaker 5>Well, I, I know, as Tim will agree, the
biggest one since PDM started, and, you

355
00:24:39.250 --> 00:24:42.810
<v Speaker 5>know, predictive analytics has started is,
is downtime avoidance.

356
00:24:43.530 --> 00:24:46.469
<v Speaker 5>It's on everybody's mind and it is a big
dollar amount.

357
00:24:46.890 --> 00:24:52.979
<v Speaker 5>Um, and it's impressive, but, um, there's
labor savings, um.

358
00:24:53.640 --> 00:24:59.619
<v Speaker 5>You know, how much time is, um, are your,
your maintenance resources spending

359
00:24:59.859 --> 00:25:03.839
<v Speaker 5>repairing a piece of equipment once it
breaks versus how much it would have taken

360
00:25:03.839 --> 00:25:10.349
<v Speaker 5>just to um Just adjust it, you know, do
some preventative maintenance on it to avoid

361
00:25:10.349 --> 00:25:11.369
<v Speaker 5>this situation.

362
00:25:13.810 --> 00:25:17.050
<v Tim Hunt>One of the big things when the operator
goes out there and hits the button, that

363
00:25:17.050 --> 00:25:18.030
<v Tim Hunt>line should run.

364
00:25:18.520 --> 00:25:22.930
<v Tim Hunt>It that they shouldn't be going through
the struggles of lines tearing up, machines

365
00:25:22.930 --> 00:25:26.569
<v Tim Hunt>are down, the production folks, they're,
you know, they're planning ahead.

366
00:25:26.650 --> 00:25:30.650
<v Tim Hunt>They, they know how many, how many, how
much production we need to run this month

367
00:25:30.969 --> 00:25:32.790
<v Tim Hunt>and that downtime is terrible.

368
00:25:33.239 --> 00:25:34.170
<v Tim Hunt>That's where we can.

369
00:25:34.425 --> 00:25:39.665
<v Tim Hunt>come in with the downtime avoidance of
saying, hey, you've got 60 days to change out

370
00:25:39.665 --> 00:25:44.295
<v Tim Hunt>this $200 bearing and you can schedule
that on the next available outage.

371
00:25:44.464 --> 00:25:48.425
<v Tim Hunt>You can have all the manpower, the pieces,
the parts, the maintenance folks get

372
00:25:48.425 --> 00:25:49.964
<v Tim Hunt>their breaks, they get their lunch.

373
00:25:50.665 --> 00:25:54.625
<v Tim Hunt>Wait a minute, that means that they're
working safer because it's not in an upset

374
00:25:54.625 --> 00:25:54.824
<v Tim Hunt>condition.

375
00:25:55.369 --> 00:26:00.520
<v Tim Hunt>Because that brings in the safety side of
it of because when it's an upset

376
00:26:00.520 --> 00:26:04.290
<v Tim Hunt>condition, everybody's wanting to know how
much longer is it gonna be down now

377
00:26:04.290 --> 00:26:07.310
<v Tim Hunt>everything's everything's a pry bar and
everything's a hammer.

378
00:26:07.410 --> 00:26:12.239
<v Tim Hunt>I know I lived that life and PDM allowed
us to get to an outage.

379
00:26:12.479 --> 00:26:14.469
<v Tim Hunt>Now you're in a safe work environment.

380
00:26:14.849 --> 00:26:19.260
<v Tim Hunt>Everything is a lot smoother and there's
gonna, you're gonna have less, less issues

381
00:26:19.260 --> 00:26:21.989
<v Tim Hunt>by just getting it to an outage and being
able to schedule.

382
00:26:22.589 --> 00:26:27.270
<v Tim Hunt>Uh, you're, you're, uh, uh, it just, it
just, I've watched it, I just watched it

383
00:26:27.270 --> 00:26:30.630
<v Tim Hunt>happen and I've seen it happen so many
times, um.

384
00:26:32.160 --> 00:26:32.170
<v Tim Hunt>So

385
00:26:34.109 --> 00:26:34.949
<v Speaker 5>Yeah, absolutely.

386
00:26:35.109 --> 00:26:37.689
<v Speaker 5>Safety, big, big, big thing here.

387
00:26:38.359 --> 00:26:44.130
<v Speaker 5>Another thing I don't consider is, um, the
loss of, um, product or product.

388
00:26:44.229 --> 00:26:48.150
<v Speaker 5>If you've got a piece of, it may not break
completely and stop operating, but it may

389
00:26:48.150 --> 00:26:52.670
<v Speaker 5>not be producing product that's within the
tolerances or the quality limits that you

390
00:26:52.670 --> 00:26:58.189
<v Speaker 5>require, and you don't know about it for a
while and you're producing, uh, less, you

391
00:26:58.189 --> 00:26:59.930
<v Speaker 5>know, lower quality product.

392
00:27:00.469 --> 00:27:07.310
<v Speaker 5>Um, or when we have a breakdown of in one
piece of equipment, if it catastrophically

393
00:27:07.310 --> 00:27:12.349
<v Speaker 5>fails, you know, how much damage does that
do to, um, connected equipment?

394
00:27:13.260 --> 00:27:18.319
<v Speaker 5>Uh, potentials there and then you've got
even more downtime repairing that.

395
00:27:18.699 --> 00:27:18.959
<v Speaker 5>Yeah.

396
00:27:19.939 --> 00:27:24.829
<v Tim Hunt>And your purchasing folks will love you a
lot more when you can schedule and plan to

397
00:27:24.829 --> 00:27:27.770
<v Tim Hunt>have a bearing ordered in instead of
hotshotted in.

398
00:27:28.020 --> 00:27:32.699
<v Tim Hunt>They hate spending that money to have
stuff hotshotted in and hey, you don't have to

399
00:27:32.699 --> 00:27:33.459
<v Tim Hunt>do that anymore.

400
00:27:33.540 --> 00:27:37.770
<v Tim Hunt>I mean it's, it's There's a, there's a
better way of doing things.

401
00:27:39.109 --> 00:27:40.229
<v Tim Hunt>A good, good point.

402
00:27:40.699 --> 00:27:47.260
<v Ryan>Uh, what, so now when you pair PDM
together with AI, so that's, that's gotta mean

403
00:27:47.260 --> 00:27:50.380
<v Ryan>something more powerful when those, when
those two things are coupled together.

404
00:27:50.739 --> 00:27:52.199
<v Ryan>So can you talk a little bit about that?

405
00:27:53.829 --> 00:27:59.239
<v Speaker 5>Um, sure, yeah, so, the, you know, just,
just a little bit earlier and all that, it

406
00:27:59.239 --> 00:28:05.380
<v Speaker 5>just, it lets you be more efficient, spend
less time investigating, um, and digging

407
00:28:05.380 --> 00:28:11.469
<v Speaker 5>in there when, I mean, many of you on
this, uh, call might be, um, are familiar with

408
00:28:11.469 --> 00:28:15.540
<v Speaker 5>vibration analysis, so you understand
that.

409
00:28:16.359 --> 00:28:21.829
<v Speaker 5>There's a certain amount of time taken in,
in going in there and um looking at those

410
00:28:21.829 --> 00:28:26.800
<v Speaker 5>vibration signatures um to determine pro
you know, what the root cause of the

411
00:28:26.800 --> 00:28:32.000
<v Speaker 5>problems may be, where to, you know, focus
your efforts on um with AI it's learned

412
00:28:32.000 --> 00:28:32.300
<v Speaker 5>that.

413
00:28:33.400 --> 00:28:38.479
<v Speaker 5>Um, you have these defined industry-wide
standards of what vibration looks like, and

414
00:28:38.479 --> 00:28:40.459
<v Speaker 5>that's been inputted in there.

415
00:28:40.959 --> 00:28:46.780
<v Speaker 5>And AI can provide that insight back, you
know, I use the predictive maintenance

416
00:28:46.949 --> 00:28:47.699
<v Speaker 5>every day.

417
00:28:48.449 --> 00:28:50.119
<v Speaker 5>Um, that's part of my role.

418
00:28:50.329 --> 00:28:56.359
<v Speaker 5>And, um, with the advent of this AI and
conversational and generative AI in here,

419
00:28:56.479 --> 00:29:01.959
<v Speaker 5>it, it's really sped up my, um, and
allowed me to um monitor more.

420
00:29:02.500 --> 00:29:05.510
<v Speaker 5>Do it more efficiently, spend less time,
um.

421
00:29:06.300 --> 00:29:09.199
<v Speaker 5>Investigating those types of things.

422
00:29:10.400 --> 00:29:10.989
<v Speaker 5>Um,

423
00:29:17.319 --> 00:29:24.040
<v Speaker 5>So Some of the other things that, you
know, and as I said, yes, you know,

424
00:29:24.199 --> 00:29:27.579
<v Speaker 5>just, it opens it up to a whole lot more
people.

425
00:29:29.930 --> 00:29:35.410
<v Speaker 5>So one of the tools that we use is of
course a um Siemens product Sensei and

426
00:29:35.410 --> 00:29:41.250
<v Speaker 5>predictive maintenance and um I've used it
for many, many years and have seen it

427
00:29:41.250 --> 00:29:46.300
<v Speaker 5>from, you know, close to its infancy to
the state it's in now.

428
00:29:46.390 --> 00:29:52.770
<v Speaker 5>Um, and with this tool, you know, I have
been able to monitor thousands and

429
00:29:52.770 --> 00:29:59.459
<v Speaker 5>thousands of assets and probably only
spend, um, You know, 11 to 2

430
00:29:59.459 --> 00:30:00.569
<v Speaker 5>hours a day doing it.

431
00:30:02.030 --> 00:30:07.469
<v Speaker 5>Um, able to train others who are, weren't
so much, you know, predictive maintenance

432
00:30:07.469 --> 00:30:14.260
<v Speaker 5>specialists into how to use a tool in
doing it and able to Um, perform

433
00:30:14.260 --> 00:30:15.500
<v Speaker 5>the same function.

434
00:30:16.369 --> 00:30:19.930
<v Speaker 5>But as part of it, you know, the thing is,
is that with this thing, you can monitor

435
00:30:19.930 --> 00:30:20.560
<v Speaker 5>your whole um.

436
00:30:22.229 --> 00:30:26.069
<v Speaker 5>Your, your whole site infrastructure, you
know, potentially globally.

437
00:30:27.449 --> 00:30:33.060
<v Speaker 5>Um, it will feedback information like, you
know, what pieces of equipment that you

438
00:30:33.060 --> 00:30:36.020
<v Speaker 5>should focus your attention on, um.

439
00:30:37.339 --> 00:30:41.089
<v Speaker 5>And then once it does point you to
something, it gives you that generative AI um

440
00:30:41.859 --> 00:30:42.099
<v Speaker 5>information.

441
00:30:42.930 --> 00:30:48.310
<v Speaker 5>Um, you know, it's a, it's a cloud-based
application, so it's, it's always evolving,

442
00:30:48.390 --> 00:30:52.670
<v Speaker 5>getting better very quickly, and it does
incorporate that AI.

443
00:30:57.109 --> 00:31:03.939
<v Ryan>Um So OK, so let's move on to the
implementation of,

444
00:31:04.000 --> 00:31:05.650
<v Ryan>of a PDM strategy.

445
00:31:06.069 --> 00:31:10.709
<v Ryan>So, what do you actually have to think
about when it comes to implementing a tool

446
00:31:10.709 --> 00:31:13.089
<v Ryan>like this, or, or a processes like this?

447
00:31:13.589 --> 00:31:19.270
<v Ryan>Are there any, are there any advice that
you might have on pitfalls to avoid, ah, or

448
00:31:19.270 --> 00:31:23.530
<v Ryan>things to absolutely take uh take in mind
when implementing?

449
00:31:25.239 --> 00:31:29.199
<v Speaker 5>Um, well, for sure, you need to know your
equipment and you need to know how it

450
00:31:29.199 --> 00:31:29.859
<v Speaker 5>breaks.

451
00:31:30.520 --> 00:31:35.839
<v Speaker 5>Um, so, you know, to get to know where to
focus your efforts, you know, of course,

452
00:31:35.880 --> 00:31:41.520
<v Speaker 5>you're probably, you know, running, um,
FMEA analysis on your equipment, you know,

453
00:31:41.640 --> 00:31:46.040
<v Speaker 5>defining its, uh, failure points, what are
the causes of those failures, you know,

454
00:31:46.199 --> 00:31:48.359
<v Speaker 5>what are the possible remedies, um.

455
00:31:49.780 --> 00:31:54.469
<v Speaker 5>You need to know those things so that you
know where to point and get your data for

456
00:31:54.469 --> 00:31:55.530
<v Speaker 5>predictive maintenance.

457
00:31:56.989 --> 00:32:01.160
<v Speaker 5>Um, That's probably the key.

458
00:32:02.890 --> 00:32:07.189
<v Speaker 5>The key to, to having a, a lot of us may
have data out there and it's stored, but

459
00:32:07.530 --> 00:32:11.010
<v Speaker 5>whether it's relevant or not for a piece
of equipment, you know, you just gotta,

460
00:32:11.150 --> 00:32:12.430
<v Speaker 5>you've got to make that determination.

461
00:32:13.760 --> 00:32:19.920
<v Speaker 5>Um, One of the things that, you know, made
initial

462
00:32:19.920 --> 00:32:26.319
<v Speaker 5>implementation of, of a predictive
maintenance strategy is Go for low-hanging fruit.

463
00:32:26.739 --> 00:32:33.020
<v Speaker 5>And what I mean by that is you are
probably already collecting um a lot of data and

464
00:32:33.020 --> 00:32:35.439
<v Speaker 5>you already have it stored in some sort of
historian.

465
00:32:37.609 --> 00:32:39.180
<v Speaker 5>You know, you can start with that.

466
00:32:39.890 --> 00:32:45.530
<v Speaker 5>Um, you can get a lot of, um, a lot of
pieces of equipment and assets onboarded into

467
00:32:45.530 --> 00:32:47.750
<v Speaker 5>the system quickly that way.

468
00:32:48.250 --> 00:32:51.810
<v Speaker 5>Um, and then from there, you just start
adding to it, you know, oh hey, we've got

469
00:32:51.810 --> 00:32:52.400
<v Speaker 5>this piece of equipment.

470
00:32:52.479 --> 00:32:53.130
<v Speaker 5>I don't have a data.

471
00:32:53.250 --> 00:32:54.410
<v Speaker 5>Well, what data do I need?

472
00:32:54.689 --> 00:32:55.729
<v Speaker 5>Well, I need vibration.

473
00:32:55.810 --> 00:33:01.050
<v Speaker 5>Well, OK, well, so we'll start designing
in, um, vibration monitoring or any other

474
00:33:01.050 --> 00:33:02.810
<v Speaker 5>technologies that might be relevant.

475
00:33:04.989 --> 00:33:05.810
<v Speaker 5>Deploy it.

476
00:33:06.390 --> 00:33:11.209
<v Speaker 5>It's always an evolving um uh process with
the company.

477
00:33:11.420 --> 00:33:14.229
<v Speaker 5>You want to keep adding until you cover as
much as you can.

478
00:33:15.150 --> 00:33:19.349
<v Speaker 5>And then you'll still find you're, you're
tweaking and adjusting and, you know, to

479
00:33:19.349 --> 00:33:21.050
<v Speaker 5>get the best results possible.

480
00:33:31.790 --> 00:33:34.449
<v Ryan>Uh, so, OK, so I'm sure it's, it's not
all.

481
00:33:35.199 --> 00:33:41.420
<v Ryan>It's not all a bed of roses here, guys,
so, uh, what, what uh challenges do

482
00:33:41.420 --> 00:33:46.880
<v Ryan>organizations typically have, uh, when
transitioning to uh to a predictive

483
00:33:46.880 --> 00:33:49.020
<v Ryan>maintenance uh implementation like this?

484
00:33:50.130 --> 00:33:56.380
<v Tim Hunt>A lot of times you run into the maturity
of the workforce, uh, the, a lot of the old

485
00:33:56.380 --> 00:34:00.380
<v Tim Hunt>school folks, you know, they, I don't
know, we're, we'll take a screwdriver and

486
00:34:00.380 --> 00:34:02.569
<v Tim Hunt>we'll put it to our ear and we'll listen
to this bearing.

487
00:34:02.619 --> 00:34:04.180
<v Tim Hunt>I can tell you when it's gonna fail.

488
00:34:04.540 --> 00:34:06.140
<v Tim Hunt>Yeah, but we've got predictive
maintenance.

489
00:34:06.219 --> 00:34:10.979
<v Tim Hunt>We can do a vibration analysis on it and
tell you it's a spa on the outer race and

490
00:34:10.979 --> 00:34:11.959
<v Tim Hunt>you've got this many days.

491
00:34:12.020 --> 00:34:13.520
<v Tim Hunt>You can't do that with a screwdriver.

492
00:34:14.139 --> 00:34:15.830
<v Tim Hunt>Well, we've evolved from there.

493
00:34:16.100 --> 00:34:19.540
<v Tim Hunt>Now those technologies have really proven
their sale.

494
00:34:19.939 --> 00:34:26.689
<v Tim Hunt>Now we're bringing AI in and it's just
another useful technology and you, you still,

495
00:34:26.780 --> 00:34:31.719
<v Tim Hunt>you've got to have folks willing to learn,
to bring to, to learn this AI part of it,

496
00:34:32.020 --> 00:34:35.739
<v Tim Hunt>uh, because there's still a lot of folks
out there that see it as a black box.

497
00:34:35.820 --> 00:34:36.860
<v Tim Hunt>I mean they're scared of it.

498
00:34:36.939 --> 00:34:38.409
<v Tim Hunt>They don't want nothing to do with it.

499
00:34:38.620 --> 00:34:43.860
<v Tim Hunt>They're not willing to think through it
and, and, and this AI is just.

500
00:34:44.350 --> 00:34:49.949
<v Tim Hunt>It's overwhelming because uh you, you
still have a lot of folks that push only

501
00:34:49.949 --> 00:34:50.709
<v Tim Hunt>vibration.

502
00:34:50.830 --> 00:34:52.989
<v Tim Hunt>I'm gonna just do everything with
vibration.

503
00:34:53.379 --> 00:34:55.969
<v Tim Hunt>Well, yeah, and I, and I'm, I love
vibration.

504
00:34:56.110 --> 00:35:00.709
<v Tim Hunt>I mean that's what I cut my teeth on 17
years ago and doing vibration.

505
00:35:01.159 --> 00:35:07.449
<v Tim Hunt>But this adds to it because with
vibration, I can't see everything else to where AI

506
00:35:07.449 --> 00:35:12.889
<v Tim Hunt>brings in flows, pressure, differential
pressure, uh, current, amperage.

507
00:35:13.169 --> 00:35:20.169
<v Tim Hunt>I can get a 3D view by using AI that I
never could get with using just the uh

508
00:35:20.169 --> 00:35:22.810
<v Tim Hunt>thermography, ultrasound or or vibration.

509
00:35:23.229 --> 00:35:28.429
<v Tim Hunt>So I love the 3D view that I'm able to get
and I'm able to teach that uh as I work

510
00:35:28.429 --> 00:35:34.189
<v Tim Hunt>with these customers uh and once they get
that picture, then they're like, oh yeah,

511
00:35:34.310 --> 00:35:38.090
<v Tim Hunt>yeah, because it's just like Todd said,
you've already got the data.

512
00:35:38.270 --> 00:35:39.830
<v Tim Hunt>All these plants have got the data.

513
00:35:39.909 --> 00:35:42.020
<v Tim Hunt>They've got the motor current, they got
the average.

514
00:35:42.389 --> 00:35:46.580
<v Tim Hunt>It don't cost you nothing at that point to
stick that in there and if a motor starts

515
00:35:46.580 --> 00:35:48.469
<v Tim Hunt>loading up, the current's gonna spike.

516
00:35:48.590 --> 00:35:51.010
<v Tim Hunt>Well, that's something down the drive
train line.

517
00:35:51.459 --> 00:35:56.060
<v Tim Hunt>We've had so much success just looking at
things without you having to spend any

518
00:35:56.060 --> 00:35:58.300
<v Tim Hunt>extra money on other stuff.

519
00:35:58.620 --> 00:36:02.860
<v Tim Hunt>Now the other stuff's good, adding the
vibration sensors, we love that part of it

520
00:36:02.860 --> 00:36:07.830
<v Tim Hunt>too, but uh, so that's kinda, that's kind
of one of the things is, you gotta have

521
00:36:07.830 --> 00:36:11.000
<v Tim Hunt>the right culture, willing to folks that
are willing to learn.

522
00:36:11.739 --> 00:36:12.100
<v Speaker 5>Yeah.

523
00:36:12.760 --> 00:36:17.000
<v Ryan>I, I do wanna touch on something that uh
that you guys said, uh, a little bit ago

524
00:36:17.000 --> 00:36:23.659
<v Ryan>about uh where to manage the, the, the
data being on-site versus on off-site,

525
00:36:24.159 --> 00:36:24.719
<v Ryan>um.

526
00:36:25.989 --> 00:36:31.510
<v Ryan>So, so there, there is, I think a lot of
the people on this call here, um, you know,

527
00:36:31.629 --> 00:36:36.540
<v Ryan>they, they have a lot of preference right
now for, for on-premise data storage and,

528
00:36:36.550 --> 00:36:42.439
<v Ryan>and implementation of this type of a, um,
you know, maybe PDM data type strategy.

529
00:36:42.750 --> 00:36:47.010
<v Ryan>Are, are you guys able to touch on, you
know, what a cloud-based strategy versus an

530
00:36:47.010 --> 00:36:48.889
<v Ryan>on-prem strategy might look like?

531
00:36:51.439 --> 00:36:58.020
<v Speaker 5>Um, Well, Well, no, you know, the, the
data storage requirements and the

532
00:36:58.020 --> 00:37:02.679
<v Speaker 5>processing power for AI is, you know, is
quite large.

533
00:37:03.219 --> 00:37:08.379
<v Speaker 5>So maintaining that kind of infrastructure
on a on-premise, you know, has its own

534
00:37:08.379 --> 00:37:09.239
<v Speaker 5>challenges.

535
00:37:09.729 --> 00:37:10.280
<v Speaker 5>Um.

536
00:37:11.669 --> 00:37:16.669
<v Speaker 5>So you would probably most likely, well,
and it also requires a very specific set of

537
00:37:16.669 --> 00:37:21.949
<v Speaker 5>um, uh, you know, experts to be, to
maintain that kind of um system.

538
00:37:22.250 --> 00:37:22.689
<v Speaker 5>Um.

539
00:37:23.770 --> 00:37:27.659
<v Speaker 5>Whereas, you know, the cloud-base works,
we're, we're seeing most of the AI

540
00:37:27.659 --> 00:37:30.459
<v Speaker 5>functionality resides, um.

541
00:37:31.929 --> 00:37:34.659
<v Speaker 5>So there is that plus, um.

542
00:37:38.000 --> 00:37:42.770
<v Speaker 5>You know, cloud-base is going to have
faster response times for updates, um, and

543
00:37:42.770 --> 00:37:48.330
<v Speaker 5>upgrades and evolving, um, technologies,
algorithms, and those things being added

544
00:37:48.330 --> 00:37:53.530
<v Speaker 5>will be more real-time added in there
where on-premise you're gonna have to go

545
00:37:53.530 --> 00:37:57.629
<v Speaker 5>through the update schedules and, you
know, keep that and then, um.

546
00:38:02.540 --> 00:38:04.909
<v Speaker 5>It's, it's a good point.

547
00:38:06.500 --> 00:38:13.419
<v Speaker 5>You know, security, uh, we're all
concerned about security, um, but you know, all

548
00:38:13.419 --> 00:38:17.379
<v Speaker 5>the ones that I've dealt with cloud-based
predictive maintenance is, they're their

549
00:38:17.379 --> 00:38:18.280
<v Speaker 5>own little world.

550
00:38:18.739 --> 00:38:23.060
<v Speaker 5>Even with the AI, they're not sharing
resources outside that.

551
00:38:23.080 --> 00:38:28.379
<v Speaker 5>They're protected, um, and nothing goes in
without, you know, without the proper

552
00:38:28.379 --> 00:38:28.899
<v Speaker 5>credentials.

553
00:38:28.939 --> 00:38:30.239
<v Speaker 5>The AI does not.

554
00:38:31.000 --> 00:38:35.290
<v Speaker 5>Send any of its data out to these other AI
systems out there in the cloud, it is all

555
00:38:35.290 --> 00:38:37.550
<v Speaker 5>local for that particular application.

556
00:38:39.810 --> 00:38:40.510
<v Ryan>Yeah, good point.

557
00:38:41.030 --> 00:38:44.389
<v Ryan>And, and maybe a moment about, uh, about
safety and security.

558
00:38:44.429 --> 00:38:47.209
<v Ryan>I guess you, you just talked on security
there, right?

559
00:38:50.250 --> 00:38:55.290
<v Speaker 5>Yeah, I, of course, I'm not a
cybersecurity expert, so I can't speak too much, but

560
00:38:55.290 --> 00:39:00.800
<v Speaker 5>it, you know, it uses all the industry
standard um technologies, you know, TLS 1.2,

561
00:39:00.969 --> 00:39:05.969
<v Speaker 5>data encryption, you know, cyber testing,
um.

562
00:39:07.290 --> 00:39:11.570
<v Speaker 5>And you know, the one that we use since Ia
Siemens and all that has been sports

563
00:39:11.570 --> 00:39:18.100
<v Speaker 5>single sign-on, um, been audited by many,
many different companies and

564
00:39:18.689 --> 00:39:21.070
<v Speaker 5>has a very high um app security rating.

565
00:39:21.939 --> 00:39:22.469
<v Ryan>Awesome.

566
00:39:24.260 --> 00:39:28.479
<v Ryan>Maybe a second, uh, because you guys have
so much experience, uh, directly

567
00:39:28.479 --> 00:39:30.899
<v Ryan>implementing these PDM type strategies.

568
00:39:31.020 --> 00:39:34.679
<v Ryan>Could you talk a little bit about, a
little bit more about your specific use case?

569
00:39:35.699 --> 00:39:42.350
<v Speaker 5>Sure, um, and this particular slide is
actually one that I was involved with, um,

570
00:39:43.219 --> 00:39:47.139
<v Speaker 5>uh, the global, um, automotive
manufacturer, um.

571
00:39:48.139 --> 00:39:52.889
<v Speaker 5>One of, you know, the first, the first
site to use the, the Sensei product was in

572
00:39:52.889 --> 00:39:54.179
<v Speaker 5>um, Europe and the UK.

573
00:39:54.770 --> 00:40:01.770
<v Speaker 5>Um, I came from one of the um US sites
and, you know, we, I

574
00:40:01.770 --> 00:40:06.449
<v Speaker 5>worked in the, um, the body shop, you
know, where all the, the body parts are welded

575
00:40:06.449 --> 00:40:11.909
<v Speaker 5>together, you know, over 1300, um, welding
robots in there.

576
00:40:12.169 --> 00:40:12.649
<v Speaker 5>Um.

577
00:40:13.729 --> 00:40:17.169
<v Speaker 5>Welding robots were the primary source of
downtime there.

578
00:40:17.330 --> 00:40:20.590
<v Speaker 5>It's just they're under heavy loads and
they wear out.

579
00:40:22.729 --> 00:40:28.729
<v Speaker 5>We'd had condition-based monitoring
systems on them for many, many years and we knew

580
00:40:28.729 --> 00:40:34.169
<v Speaker 5>that we could see these failures on coming
with the data, the tort data, but before

581
00:40:34.169 --> 00:40:37.810
<v Speaker 5>predictive maintenance, the only way to do
it is that you had to look at a, um, tort

582
00:40:37.810 --> 00:40:38.250
<v Speaker 5>graph.

583
00:40:39.689 --> 00:40:45.729
<v Speaker 5>And that could take anywhere from 15 to 20
minutes or longer to do.

584
00:40:45.860 --> 00:40:52.560
<v Speaker 5>So you can imagine, multiply that by 1300
robots, the amount of time taken to it.

585
00:40:52.840 --> 00:40:56.949
<v Speaker 5>And even if we threw everybody in the
maintenance department in that shop in there,

586
00:40:57.229 --> 00:41:04.080
<v Speaker 5>you still cannot return, return around to
a specific robot to catch it potentially

587
00:41:04.080 --> 00:41:04.679
<v Speaker 5>in time.

588
00:41:05.639 --> 00:41:09.209
<v Speaker 5>So we knew there that we had to have some
sort of predictive maintenance.

589
00:41:09.330 --> 00:41:16.310
<v Speaker 5>So, you know, we, we contacted um Uh,
Sensei is the other, you know, UK plant

590
00:41:16.310 --> 00:41:22.750
<v Speaker 5>was um using and began a uh proof of
concept with them plus a few other um offerings

591
00:41:22.750 --> 00:41:23.409
<v Speaker 5>at the time.

592
00:41:24.270 --> 00:41:25.669
<v Speaker 5>And um.

593
00:41:26.669 --> 00:41:32.850
<v Speaker 5>We implemented, um, once through the proof
of concept and testing that we, you know,

594
00:41:34.030 --> 00:41:39.550
<v Speaker 5>did to each one of them, we chose the
Sensii product and at that point, we

595
00:41:39.949 --> 00:41:46.889
<v Speaker 5>onboarded, um, you know, 2000 as robots
and their weld guns and then

596
00:41:46.979 --> 00:41:53.479
<v Speaker 5>began adding more and more and more over
the years and, and that's where Since with

597
00:41:53.479 --> 00:41:56.310
<v Speaker 5>all the additional assets on boarded, we
had that

598
00:41:56.310 --> 00:42:03.120
<v Speaker 5>$45 million savings, um, we were actually
able to justify

599
00:42:03.120 --> 00:42:09.159
<v Speaker 5>the cost of the system in labor savings
alone, not, not even taking downtime into

600
00:42:09.159 --> 00:42:09.840
<v Speaker 5>consideration.

601
00:42:13.159 --> 00:42:13.780
<v Ryan>Nice.

602
00:42:14.879 --> 00:42:19.840
<v Ryan>So, although this is ah an automotive ah
application, I mean, we could certainly

603
00:42:19.840 --> 00:42:26.780
<v Ryan>craft strategies like this for, for the
subfab in a semiconductor facility, ah, also

604
00:42:26.780 --> 00:42:29.899
<v Ryan>relate this to, you know, areas inside the
clean room.

605
00:42:30.020 --> 00:42:35.219
<v Ryan>So, ah, there's lots and lots of ways that
you can apply this type of a strategy to,

606
00:42:35.580 --> 00:42:40.340
<v Ryan>you know, to the motors in, um, you know,
in an air handling unit, or, you know,

607
00:42:40.459 --> 00:42:44.020
<v Ryan>more of this, more of the sophisticated
tooling inside the cleaning room.

608
00:42:44.139 --> 00:42:48.979
<v Ryan>So, Ah, there is adaptability of, of these
predictive maintenance type strategies

609
00:42:48.979 --> 00:42:54.199
<v Ryan>for different various areas of the, of a,
uh, semi facility.

610
00:42:54.620 --> 00:42:59.139
<v Ryan>Um, I did want to ask another question,
ah, to the group.

611
00:42:59.219 --> 00:43:04.540
<v Ryan>I know that we kind of touched on this,
uh, a second ago, um, about some of the

612
00:43:04.540 --> 00:43:10.419
<v Ryan>challenges that organizations have maybe
more culturally about adopting a predictive

613
00:43:10.419 --> 00:43:12.419
<v Ryan>maintenance strategy, um.

614
00:43:13.070 --> 00:43:17.840
<v Ryan>Uh, so Tim, uh, could you elaborate maybe
a little bit more about that, um, on the

615
00:43:17.840 --> 00:43:21.100
<v Ryan>cultural side of things of predictive
maintenance transformations?

616
00:43:21.560 --> 00:43:22.060
<v Tim Hunt>Sure.

617
00:43:22.600 --> 00:43:27.560
<v Tim Hunt>We've seen the greatest success come at,
at the plants when they have what's called

618
00:43:27.560 --> 00:43:28.709
<v Tim Hunt>a plant champion.

619
00:43:29.080 --> 00:43:33.879
<v Tim Hunt>When there's somebody there on that site
that owns it, and that's who I get to work

620
00:43:33.879 --> 00:43:39.010
<v Tim Hunt>with and, and Todd works with those folks
and And, uh, though, if we can get a plant

621
00:43:39.010 --> 00:43:44.209
<v Tim Hunt>champion, hey, this thing will move
forward and we will have true success because

622
00:43:44.209 --> 00:43:50.010
<v Tim Hunt>those individuals are what, uh, say they
have, they catch a fan that's, uh, a dirty

623
00:43:50.010 --> 00:43:53.040
<v Tim Hunt>fan blade and, uh, and, and we've watched
it.

624
00:43:53.129 --> 00:43:58.370
<v Tim Hunt>They are triggered and created a case and
we, we'll, we'll actually go in and create

625
00:43:58.370 --> 00:44:01.189
<v Tim Hunt>a success case, a slide for that.

626
00:44:01.639 --> 00:44:05.600
<v Tim Hunt>The plant champion will take it and show
it to his boss, which will in turn end up

627
00:44:05.600 --> 00:44:09.580
<v Tim Hunt>going up the food chain and hey, this is
what winning looks like.

628
00:44:09.600 --> 00:44:11.110
<v Tim Hunt>They saved us this much money.

629
00:44:11.439 --> 00:44:16.290
<v Tim Hunt>So the a plant champion, plants that don't
have a good solid champion, but we, we've

630
00:44:16.290 --> 00:44:20.750
<v Tim Hunt>struggled with those plants because if, if
it's somebody that just don't care, not

631
00:44:20.750 --> 00:44:25.320
<v Tim Hunt>willing to learn, but we, it just, it's
hard to get any traction even though we're,

632
00:44:25.399 --> 00:44:30.239
<v Tim Hunt>we're selling the story, we're, we're
showing it, but, uh, so a plant champion is

633
00:44:30.239 --> 00:44:30.899
<v Tim Hunt>priceless.

634
00:44:31.629 --> 00:44:37.959
<v Tim Hunt>Um, Another part of it is, as far as the
cultural side of it

635
00:44:38.239 --> 00:44:43.459
<v Tim Hunt>is, uh, we really pushed to get these
teams being data-driven decision-making.

636
00:44:43.949 --> 00:44:45.419
<v Tim Hunt>That is so important.

637
00:44:45.719 --> 00:44:50.560
<v Tim Hunt>And, and the and here's an example I can
give is uh one of the steel steel mills I

638
00:44:50.560 --> 00:44:54.979
<v Tim Hunt>was working with, they had a vibration
jump on their uh motor and pump combination.

639
00:44:55.879 --> 00:45:00.159
<v Tim Hunt>He said, typically we would go out there,
we would change out the motor, they didn't

640
00:45:00.159 --> 00:45:01.679
<v Tim Hunt>fix it, we'd change the pump out.

641
00:45:02.060 --> 00:45:06.409
<v Tim Hunt>But he said we had, we had just put the
differential pressure off of the heat

642
00:45:06.409 --> 00:45:09.580
<v Tim Hunt>exchanger plates that this motor pump
feeds through.

643
00:45:10.060 --> 00:45:15.620
<v Tim Hunt>We just put that on, on Sensa and so we
looked at that and we saw the trend and it

644
00:45:15.620 --> 00:45:20.300
<v Tim Hunt>actually had generated a case telling us,
hey, the pressure's going up, which means

645
00:45:20.300 --> 00:45:23.199
<v Tim Hunt>these heat exchanger plates are are
getting stopped up.

646
00:45:23.679 --> 00:45:27.159
<v Tim Hunt>He said in the past we would have started
changing pieces and parts out.

647
00:45:27.409 --> 00:45:31.600
<v Tim Hunt>He said, once we saw that we knew all we
had to do was clean out, clean the heat

648
00:45:31.600 --> 00:45:32.560
<v Tim Hunt>exchanger plates.

649
00:45:32.929 --> 00:45:37.090
<v Tim Hunt>Once they cleaned those, changed them, the
vibration went down on the motor and

650
00:45:37.090 --> 00:45:37.590
<v Tim Hunt>pumps.

651
00:45:37.969 --> 00:45:43.030
<v Tim Hunt>So that tells you, let the data drive you
to make decisions and through that they've

652
00:45:43.030 --> 00:45:47.649
<v Tim Hunt>been able to do PM optimization because
now they've got it set up to where this

653
00:45:47.649 --> 00:45:53.770
<v Tim Hunt>thing, it will generate a case uh whenever
it's uh these filters are starting to get

654
00:45:53.770 --> 00:45:54.389
<v Tim Hunt>stopped up.

655
00:45:54.770 --> 00:45:55.810
<v Tim Hunt>So that, that's huge.

656
00:45:55.889 --> 00:45:59.850
<v Tim Hunt>They're not, they're not changing them out
on a time basis anymore.

657
00:46:00.050 --> 00:46:03.570
<v Tim Hunt>They're changing them out when the data
says the filters are stopped up.

658
00:46:04.699 --> 00:46:06.760
<v Tim Hunt>So, those are a couple of examples.

659
00:46:09.149 --> 00:46:13.889
<v Ryan>Yeah, before we, before we get to the, the
Q&amp;A, we've got a couple of questions

660
00:46:13.889 --> 00:46:18.110
<v Ryan>going in here, um, but we did wanna just
kind of wrap this up from a, from a, a

661
00:46:18.110 --> 00:46:19.100
<v Ryan>presentation standpoint.

662
00:46:19.189 --> 00:46:23.489
<v Ryan>We do have a couple of questions though,
um, but I mean, as we heard today,

663
00:46:23.709 --> 00:46:24.959
<v Ryan>predictive maintenance is, is a.

664
00:46:25.024 --> 00:46:27.415
<v Ryan>Really great way to improve lots of
things.

665
00:46:27.465 --> 00:46:32.655
<v Ryan>Improve safety, improve your improve your
overall throughput, your machine, ah,

666
00:46:32.705 --> 00:46:39.304
<v Ryan>uptime, product quality, ah, you can train
workforce has so many upsides to

667
00:46:39.304 --> 00:46:43.024
<v Ryan>predictive maintenance, which has been
around for, for a while now.

668
00:46:43.304 --> 00:46:50.225
<v Ryan>But really, it's, it's taking ah AI
interfacing with different AI tools

669
00:46:50.225 --> 00:46:53.764
<v Ryan>to enhance your ah predictive maintenance
strategies.

670
00:46:54.399 --> 00:46:59.959
<v Ryan>Um, taking also safety and security into
mind when we interface with with AI engines

671
00:46:59.959 --> 00:47:01.699
<v Ryan>like that as we kind of looked at.

672
00:47:02.639 --> 00:47:09.360
<v Ryan>Um, and then, yeah, you should look to to
partner with with a company that can look

673
00:47:09.360 --> 00:47:13.320
<v Ryan>at a PDM strategy across the whole
company.

674
00:47:13.439 --> 00:47:16.219
<v Ryan>So, how do you partner with the right
company?

675
00:47:16.520 --> 00:47:21.919
<v Ryan>Um, Siemens is, we, we would hope that you
would consider Siemens as one of these

676
00:47:21.919 --> 00:47:28.479
<v Ryan>companies with the ability to have, ah, to
provide end to end solutions

677
00:47:28.479 --> 00:47:35.080
<v Ryan>from chip design, all the way to the final
integration of the sensor at the

678
00:47:35.080 --> 00:47:42.080
<v Ryan>factory level, and then using ah ah
predictive maintenance platform ah to to tie it

679
00:47:42.080 --> 00:47:44.379
<v Ryan>all together and and make sure that your
stuff.

680
00:47:44.760 --> 00:47:50.800
<v Ryan>You know, is, is maintained properly, so,
um, you know, with that, we do have a, a

681
00:47:50.800 --> 00:47:53.179
<v Ryan>really pretty poignant question here.

682
00:47:54.260 --> 00:48:01.030
<v Ryan>Ah, Tim and Todd, could you give an
example of, of, uh, predictive maintenance and

683
00:48:01.030 --> 00:48:04.209
<v Ryan>how conversational PDM might work?

684
00:48:04.500 --> 00:48:06.800
<v Ryan>So you mentioned that conversational PDM.

685
00:48:07.219 --> 00:48:10.500
<v Ryan>Could you explain, uh, what, what that,
what that would look like?

686
00:48:11.570 --> 00:48:16.729
<v Tim Hunt>I just got to see this, uh, like just
actually this week, uh, and, and it's

687
00:48:16.729 --> 00:48:17.770
<v Tim Hunt>happening more and more.

688
00:48:17.850 --> 00:48:23.629
<v Tim Hunt>The conversational part is, uh, Say, uh,
you get a case triggered, a fan is out of

689
00:48:23.629 --> 00:48:27.469
<v Tim Hunt>balance, and they go in and they go in and
they balance the fan.

690
00:48:27.590 --> 00:48:29.229
<v Tim Hunt>The amplitudes drop back down.

691
00:48:29.469 --> 00:48:33.590
<v Tim Hunt>Life is good, but then they go whenever,
but right before they close out a case,

692
00:48:33.669 --> 00:48:37.229
<v Tim Hunt>they'll enter what's called a work event
and that gives all the details.

693
00:48:37.350 --> 00:48:39.169
<v Tim Hunt>Hey, we watched this trend go up.

694
00:48:39.469 --> 00:48:41.189
<v Tim Hunt>We realized that it was out of balance.

695
00:48:41.389 --> 00:48:45.399
<v Tim Hunt>We went in and we cleaned the fan blades
and, and now that fixed it.

696
00:48:45.669 --> 00:48:47.149
<v Tim Hunt>So they, they typed that in.

697
00:48:47.659 --> 00:48:54.540
<v Tim Hunt>Well, the uh then this week, there was a
uh uh an instance where the co-pilot

698
00:48:54.540 --> 00:48:58.929
<v Tim Hunt>actually, there was a case generated and
co-pilot said you need to go out and clean

699
00:48:58.929 --> 00:49:02.639
<v Tim Hunt>your fan blade and, and, but the, the all
the details was in it.

700
00:49:02.959 --> 00:49:07.280
<v Tim Hunt>The gentleman that actually entered that
information in months ago because it had

701
00:49:07.280 --> 00:49:09.919
<v Tim Hunt>gotten dirty before, he said that's my
writing.

702
00:49:10.260 --> 00:49:11.149
<v Tim Hunt>I, that's how I write.

703
00:49:11.239 --> 00:49:12.530
<v Tim Hunt>I'm the one that put that in.

704
00:49:12.760 --> 00:49:17.600
<v Tim Hunt>He was so excited because it actually
learns from the information you're putting in,

705
00:49:18.000 --> 00:49:20.199
<v Tim Hunt>uh, what and so I, and I keep seeing.

706
00:49:20.274 --> 00:49:24.784
<v Tim Hunt>That at different customers of, hey, I put
that in there uh uh last year I put that,

707
00:49:25.114 --> 00:49:29.635
<v Tim Hunt>uh, you know, and so they're getting
excited about this because the AI is saying,

708
00:49:29.794 --> 00:49:34.155
<v Tim Hunt>hey, and it, the co-pilot's going in and
looking at all this and it's making

709
00:49:34.155 --> 00:49:38.675
<v Tim Hunt>decisions like that and, and telling them
go out there and look at this and so that,

710
00:49:38.794 --> 00:49:41.114
<v Tim Hunt>that's not just an example that I've seen
recently.

711
00:49:43.010 --> 00:49:48.530
<v Ryan>It's great, yeah, the, the AI learning
from input from some of the colleagues doing

712
00:49:48.530 --> 00:49:51.820
<v Ryan>the maintenance is, that's a, that's an
awesome application of AI.

713
00:49:53.409 --> 00:49:53.570
<v Ryan>Yeah,

714
00:49:53.729 --> 00:49:58.290
<v Tim Hunt>let me add to that just a little bit
because that information that the individual

715
00:49:58.290 --> 00:49:58.810
<v Tim Hunt>put in.

716
00:49:59.310 --> 00:50:01.879
<v Tim Hunt>Hey, in 5 years this guy's gonna be
retired out.

717
00:50:02.030 --> 00:50:06.889
<v Tim Hunt>There's gonna be new people come in that's
not gonna know what this looks like.

718
00:50:07.149 --> 00:50:12.189
<v Tim Hunt>And so this is a knowledge platform that
you can go in and save this information

719
00:50:12.469 --> 00:50:15.590
<v Tim Hunt>from a skilled labor, the, uh, the, the.

720
00:50:15.735 --> 00:50:19.145
<v Tim Hunt>Folks before they retire out, they've got
a place to put this information in.

721
00:50:19.385 --> 00:50:23.985
<v Tim Hunt>Hey, we align this motor and you gotta be,
watch for the soft foot on this foot.

722
00:50:24.185 --> 00:50:28.584
<v Tim Hunt>I mean there, there's, you can put in all
the details you want so that's another

723
00:50:28.584 --> 00:50:31.104
<v Tim Hunt>powerful is knowledge, is capturing

724
00:50:31.104 --> 00:50:31.385
<v Ryan>that

725
00:50:31.385 --> 00:50:31.804
<v Tim Hunt>knowledge.

726
00:50:35.689 --> 00:50:37.040
<v Ryan>That that's another great point, Tim.

727
00:50:37.090 --> 00:50:41.409
<v Ryan>I mean, yeah, we're we're, we're talking
about, you know, you know, almost infinite

728
00:50:41.409 --> 00:50:46.530
<v Ryan>databases out there where, you know, the
continuity of, of knowledge, you know, more

729
00:50:46.530 --> 00:50:51.929
<v Ryan>so than, than what we can keep or upkeep
as, as our own, you know, written records.

730
00:50:52.370 --> 00:50:55.350
<v Ryan>That's a, that's an awesome application of
of of AI there.

731
00:50:57.530 --> 00:51:03.219
<v Ryan>Uh, we had another cool question come in,
um, and maybe, maybe you guys can answer

732
00:51:03.219 --> 00:51:03.459
<v Ryan>this.

733
00:51:03.659 --> 00:51:10.479
<v Ryan>So what what impact to the overall
equipment effectiveness has been seen

734
00:51:10.739 --> 00:51:13.379
<v Ryan>due to the application of AI for PDM?

735
00:51:13.500 --> 00:51:20.199
<v Ryan>So we're looking for, you know, like an an
impact to the OEE uh by implementing AI

736
00:51:20.199 --> 00:51:21.159
<v Ryan>and PDM.

737
00:51:22.590 --> 00:51:29.510
<v Speaker 5>Well, and we know that PDM has improved
OEE, you know, a lot, um, over

738
00:51:29.510 --> 00:51:34.590
<v Speaker 5>time, and then just anything that makes
that PDM more effective is also going to

739
00:51:34.590 --> 00:51:36.610
<v Speaker 5>make your OEE, um, better.

740
00:51:37.010 --> 00:51:43.709
<v Speaker 5>Um, you know, you're gonna, you'll get
these repairs done quicker because we may

741
00:51:43.709 --> 00:51:46.510
<v Speaker 5>tell you something's going on with the,
with the AI.

742
00:51:46.629 --> 00:51:50.929
<v Speaker 5>It makes it, makes you, you can find the
root cause problem quicker.

743
00:51:51.820 --> 00:51:56.320
<v Speaker 5>Maybe you have, you have a single day of
downtime, you know, for maintenance, uh,

744
00:51:56.489 --> 00:51:59.719
<v Speaker 5>you know, one weekend, and you know you
got a problem going on, but you cannot find

745
00:51:59.719 --> 00:52:00.449
<v Speaker 5>it that way.

746
00:52:00.489 --> 00:52:02.909
<v Speaker 5>So now you got to wait a whole another
week.

747
00:52:05.159 --> 00:52:07.699
<v Speaker 5>You know, before you can work on it again.

748
00:52:08.389 --> 00:52:10.409
<v Speaker 5>AI is just gonna compress that.

749
00:52:11.340 --> 00:52:16.479
<v Speaker 5>Investigation repair cycle, even smaller,
get it done quicker,

750
00:52:17.770 --> 00:52:22.679
<v Speaker 5>reducing the, the potential for those
breakdowns to happen before you find the um

751
00:52:23.570 --> 00:52:25.080
<v Speaker 5>find the root cause problem.

752
00:52:28.149 --> 00:52:28.939
<v Ryan>Yeah, good call.

753
00:52:31.080 --> 00:52:35.189
<v Ryan>Ah, we have another question that came in,
uh, into the chat.

754
00:52:35.679 --> 00:52:42.360
<v Ryan>So, it's about building up the AI driven
insights, ah, from the

755
00:52:42.360 --> 00:52:45.540
<v Ryan>data that, that's coming from the
equipment.

756
00:52:45.959 --> 00:52:52.520
<v Ryan>Are these data insights, these data
inputs, supervised or unsupervised

757
00:52:52.520 --> 00:52:53.260
<v Ryan>models?

758
00:52:53.520 --> 00:52:57.040
<v Ryan>And I think we talked about this, we
showed a little bit about this with the

759
00:52:57.040 --> 00:52:58.850
<v Ryan>security part, um.

760
00:52:59.449 --> 00:53:06.219
<v Ryan>But again, this is ah what's the process
to build up the AI driven insights from the

761
00:53:06.219 --> 00:53:08.159
<v Ryan>data that's coming from the equipment?

762
00:53:08.489 --> 00:53:10.969
<v Ryan>Is that supervised or unsupervised?

763
00:53:13.669 --> 00:53:18.229
<v Speaker 5>Um, I'm trying to understand the question
between supervised and unsupervised is

764
00:53:18.229 --> 00:53:25.040
<v Speaker 5>that, um, It's supervising the point of
that is, it's localized for

765
00:53:25.040 --> 00:53:29.919
<v Speaker 5>your facility, your, your instance of this
um application, um.

766
00:53:31.449 --> 00:53:34.800
<v Speaker 5>Of course, you know, talking about what
can be input in there.

767
00:53:36.020 --> 00:53:38.780
<v Speaker 5>There's two ways that this can, you know,
be done.

768
00:53:38.939 --> 00:53:43.620
<v Speaker 5>One, you're manually entering in, as Tim
said, work events, or you manually enter in

769
00:53:43.620 --> 00:53:49.020
<v Speaker 5>other means in there through notes, you
know, results of the, um, of this, the

770
00:53:49.020 --> 00:53:49.879
<v Speaker 5>incidents.

771
00:53:50.129 --> 00:53:55.919
<v Speaker 5>Um, there's also a potential where, you
know, you, you can provide exports.

772
00:53:56.750 --> 00:54:02.260
<v Speaker 5>Out of your uh CMMS systems where probably
the majority of your historical

773
00:54:02.260 --> 00:54:08.860
<v Speaker 5>contextual information in regards is
residing so it can take that so you would

774
00:54:08.860 --> 00:54:14.070
<v Speaker 5>export that, run it through an import
model to put it in the appropriate um areas in

775
00:54:14.070 --> 00:54:17.729
<v Speaker 5>the, in the knowledge areas of the system
and then it becomes part of it.

776
00:54:18.550 --> 00:54:20.610
<v Speaker 5>Hopefully that answers the question.

777
00:54:22.340 --> 00:54:25.409
<v Ryan>Yeah, yeah, I think, I think we're going
in the right direction.

778
00:54:25.419 --> 00:54:29.500
<v Ryan>I mean, obviously there should be the,
the, you know, the watchdog of security kind

779
00:54:29.500 --> 00:54:35.719
<v Ryan>of governing the channels, um, but yeah,
we don't certainly don't want just any, um,

780
00:54:36.300 --> 00:54:40.199
<v Ryan>yeah, any, any input going into the, into
the AI models.

781
00:54:40.989 --> 00:54:44.159
<v Ryan>Uh, so, there's another question that came
in.

782
00:54:44.760 --> 00:54:49.810
<v Ryan>Have you applied condition monitoring data
with process yield?

783
00:54:50.860 --> 00:54:53.929
<v Ryan>To provide further tuning and financial
benefit.

784
00:54:55.350 --> 00:54:59.030
<v Speaker 5>Um, that's a good question that's been
asked a lot in the, in the past, and, you

785
00:54:59.030 --> 00:55:05.310
<v Speaker 5>know, you know, my previous positions and
all that using predictive maintenance and

786
00:55:08.219 --> 00:55:12.860
<v Speaker 5>The machine AI and machine learning
required to do that really outside of my scope

787
00:55:12.860 --> 00:55:13.370
<v Speaker 5>on there.

788
00:55:13.590 --> 00:55:18.669
<v Speaker 5>It's been looked at, but since the tool
that I use is primarily is really its focus

789
00:55:18.669 --> 00:55:21.570
<v Speaker 5>is predictive maintenance, I have not, um.

790
00:55:23.169 --> 00:55:26.580
<v Speaker 5>Use that for, you know, process
adjustments.

791
00:55:27.679 --> 00:55:33.020
<v Tim Hunt>I've been able, well, one thing that
we've, I've been able to do in the past was uh

792
00:55:33.389 --> 00:55:40.199
<v Tim Hunt>use the process, the AI and taking the
vibration and process data and the like uh

793
00:55:40.199 --> 00:55:45.100
<v Tim Hunt>like alloy, what is the alloy, aluminum
alloy cause the machines will run different

794
00:55:45.100 --> 00:55:49.719
<v Tim Hunt>on different alloys because some of them
are harder, some are softer and you can

795
00:55:49.719 --> 00:55:54.909
<v Tim Hunt>actually let the AI run and you can uh
like this alloy runs different than this one.

796
00:55:54.959 --> 00:55:56.159
<v Tim Hunt>You'll have more vibration.

797
00:55:56.510 --> 00:56:01.750
<v Tim Hunt>The sheet will, will chatter more at this
speed, so we, we can't run at this speed,

798
00:56:01.830 --> 00:56:02.110
<v Tim Hunt>right?

799
00:56:02.520 --> 00:56:07.830
<v Tim Hunt>So uh we've been able to do that and that,
that's pretty exciting to see is uh by

800
00:56:07.830 --> 00:56:12.709
<v Tim Hunt>bringing in all the process information
with it and when you overlay that, it'll

801
00:56:12.709 --> 00:56:16.860
<v Tim Hunt>tell a story that, uh, you know, that it's
just unbelievable.

802
00:56:17.030 --> 00:56:21.669
<v Tim Hunt>You're actually getting such a better view
of what this machine is doing from the

803
00:56:21.669 --> 00:56:22.889
<v Tim Hunt>start to the finish.

804
00:56:27.209 --> 00:56:31.229
<v Ryan>There's another maybe, maybe one more
interesting one that came in, guys, um.

805
00:56:32.209 --> 00:56:38.469
<v Ryan>Ah, and this is about the data coming in,
if it's, if it's insufficient or not.

806
00:56:38.610 --> 00:56:44.989
<v Ryan>So, what if the data we have is not
sufficient, so the, the, the historical data, or

807
00:56:44.989 --> 00:56:48.750
<v Ryan>if the equipment is old and doesn't have
sensors and logs.

808
00:56:49.330 --> 00:56:52.469
<v Ryan>Are there any AI models that we can look
at?

809
00:56:52.959 --> 00:56:58.280
<v Ryan>Now, I think this kind of goes into the,
you know, bottoms up learning versus top

810
00:56:58.280 --> 00:56:58.639
<v Ryan>down.

811
00:56:58.800 --> 00:57:03.350
<v Ryan>I know that the the the portfolio or the
platform that we work with at Siemens is

812
00:57:03.350 --> 00:57:08.199
<v Ryan>more of a a machine learning, and then
have the predictive maintenance software kind

813
00:57:08.199 --> 00:57:09.340
<v Ryan>of react on it.

814
00:57:09.919 --> 00:57:14.000
<v Ryan>The question is, do we have like a
top-down approach that we can recommend a model

815
00:57:14.000 --> 00:57:19.719
<v Ryan>from the top, and then push that down ah
to to the predictive maintenance ah

816
00:57:19.719 --> 00:57:20.419
<v Ryan>platform.

817
00:57:23.439 --> 00:57:24.409
<v Ryan>Um,

818
00:57:24.969 --> 00:57:26.689
<v Speaker 5>I can't answer that question myself.

819
00:57:26.770 --> 00:57:30.090
<v Speaker 5>I mean, as far as old equipment and all
that, um.

820
00:57:30.899 --> 00:57:35.100
<v Speaker 5>It really depends on the equipment that
you got, you know, if, if it's a relay

821
00:57:35.100 --> 00:57:37.879
<v Speaker 5>machine, we're probably not going to get
anywhere with that.

822
00:57:38.219 --> 00:57:44.129
<v Speaker 5>Um, but if it's PLC controlled and it's a
cycling piece of equipment, you know, one

823
00:57:44.129 --> 00:57:48.820
<v Speaker 5>of the things they use many times is use
the PLC to measure a time of flight.

824
00:57:49.689 --> 00:57:55.290
<v Speaker 5>For an operation and then use that
measured value as a predictive maintenance or a

825
00:57:55.290 --> 00:57:56.620
<v Speaker 5>condition on there.

826
00:57:57.050 --> 00:58:02.610
<v Speaker 5>If it takes too long, um, maybe there's,
there's friction included, you know, load

827
00:58:02.610 --> 00:58:07.449
<v Speaker 5>increases and all that and if that's
occurring over time, you know, by transmitting

828
00:58:07.449 --> 00:58:10.949
<v Speaker 5>that data, you'll see that increase in
time.

829
00:58:12.310 --> 00:58:15.850
<v Speaker 5>And, and that has been successful in the
past, at least for me.

830
00:58:17.379 --> 00:58:17.689
<v Ryan>Yeah.

831
00:58:22.739 --> 00:58:28.389
<v Ryan>Hey, so I think we're kind of running up
against the, the top of the hour here, um.

832
00:58:29.590 --> 00:58:32.510
<v Ryan>Yeah, so, uh, we thank you guys.

833
00:58:32.590 --> 00:58:35.669
<v Ryan>Thanks to everyone out there in the
audience for joining.

834
00:58:35.790 --> 00:58:41.149
<v Ryan>Uh, if you want to, please scan these, uh,
these two QR codes if you'd like to have

835
00:58:41.149 --> 00:58:47.050
<v Ryan>some follow-up about any other
consultation around predictive maintenance AI topics.

836
00:58:47.429 --> 00:58:52.459
<v Ryan>Um, if you wanna have a conversation with
us offline, or, or see a demo of this, uh,

837
00:58:52.510 --> 00:58:57.129
<v Ryan>AI, uh, PDM collaboration, hybridization
in work.

838
00:58:57.669 --> 00:59:03.830
<v Ryan>Um, so again, I've been Ryan, and thank
you to Tim and Todd, uh, for your expertise

839
00:59:03.830 --> 00:59:04.290
<v Ryan>today.

840
00:59:04.590 --> 00:59:09.310
<v Ryan>And, um, yeah, we hope that you found this
found this helpful, and maybe we'll talk

841
00:59:09.310 --> 00:59:10.790
<v Ryan>to you all sometime soon.

842
00:59:12.060 --> 00:59:14.300
<v Speaker 1>Please like and subscribe on all major.

843
00:59:15.600 --> 00:59:18.760
<v Speaker 1>And find out more about Sensi predictive
maintenance.

844
00:59:20.149 --> 00:59:20.489
<v Speaker 1>com
