WEBVTT
Language: en-US

1
00:00:07.320 --> 00:00:08.020
<v Liz McGinn>Welcome, everyone.

2
00:00:08.199 --> 00:00:11.189
<v Liz McGinn>I hope everyone is excited to learn how to
get started with Sensai Predictive

3
00:00:11.189 --> 00:00:11.899
<v Liz McGinn>Maintenance.

4
00:00:12.689 --> 00:00:17.659
<v Liz McGinn>As Dia said, I'm Liz McGinn, the demand
generation lead here at Sensai in the

5
00:00:17.659 --> 00:00:17.959
<v Liz McGinn>Americas.

6
00:00:18.040 --> 00:00:23.110
<v Liz McGinn>I'm joined by Doctor Natalie Kergen, the
the delivery training lead for Sensai

7
00:00:23.110 --> 00:00:24.819
<v Liz McGinn>Predictive maintenance at Siemens.

8
00:00:25.319 --> 00:00:29.020
<v Liz McGinn>Today, we've pretty much been explaining
Natalie's role with the help of companies,

9
00:00:29.290 --> 00:00:33.209
<v Liz McGinn>um, With helping companies get started
with predictive maintenance.

10
00:00:35.590 --> 00:00:38.340
<v Liz McGinn>Before we get started, let's kind of go
over our agenda.

11
00:00:38.580 --> 00:00:42.139
<v Liz McGinn>So we'll start, start with what is
predictive maintenance, then we'll go into what

12
00:00:42.139 --> 00:00:46.180
<v Liz McGinn>is Sensii predictive maintenance, and then
a little bit into our Senseii knowledge

13
00:00:46.180 --> 00:00:46.639
<v Liz McGinn>platform.

14
00:00:48.270 --> 00:00:52.319
<v Liz McGinn>Um, but before we get started, um, with
Natalie's full job, uh, Natalie, do you want

15
00:00:52.319 --> 00:00:55.020
<v Liz McGinn>to explain a little bit more about Sunset
Predictive maintenance?

16
00:00:55.830 --> 00:00:56.240
<v Liz McGinn>Sure.

17
00:00:57.439 --> 00:01:01.310
<v Speaker 2>Um, before we even talk about what Sensai
is, I think we should cover a little bit

18
00:01:01.310 --> 00:01:04.569
<v Speaker 2>about what is predictive maintenance
first, because sometimes there can be some

19
00:01:04.569 --> 00:01:06.209
<v Speaker 2>different definitions floating around.

20
00:01:06.330 --> 00:01:08.589
<v Speaker 2>So let's just make sure we're on the same
page.

21
00:01:09.330 --> 00:01:12.949
<v Speaker 2>Now, the precursor to predictive
maintenance is condition monitoring.

22
00:01:13.639 --> 00:01:18.680
<v Speaker 2>Uh, which is using sensor data from
machines to support decision making in your

23
00:01:18.680 --> 00:01:19.489
<v Speaker 2>maintenance journey.

24
00:01:20.230 --> 00:01:24.669
<v Speaker 2>Condition-based maintenance builds on that
and uses the data that you're collecting

25
00:01:24.669 --> 00:01:26.489
<v Speaker 2>to replace planned maintenance.

26
00:01:27.150 --> 00:01:34.069
<v Speaker 2>Um, predictive maintenance furthers that
more by using online monitoring so you can

27
00:01:34.069 --> 00:01:38.750
<v Speaker 2>monitor thousands of machines rather than
having a single person, um, analyzing

28
00:01:38.750 --> 00:01:39.410
<v Speaker 2>data.

29
00:01:39.830 --> 00:01:44.099
<v Speaker 2>And the difference as well is that
predictive maintenance is going to be using AI

30
00:01:44.099 --> 00:01:46.550
<v Speaker 2>and machine learning, different programs
to.

31
00:01:47.599 --> 00:01:51.400
<v Speaker 2>See how data is trending rather than just
telling you right when a machine is

32
00:01:51.400 --> 00:01:56.839
<v Speaker 2>experiencing a degradation or a failure,
something like Sensi would tell you when

33
00:01:56.839 --> 00:02:00.040
<v Speaker 2>it's on its way towards experiencing a
degradation.

34
00:02:00.870 --> 00:02:05.069
<v Speaker 2>Um, so this way it can minimize disruption
to operations and your maintenance by

35
00:02:05.069 --> 00:02:09.639
<v Speaker 2>making sure you have everything you need
to perform, uh, maintenance, and to even

36
00:02:09.639 --> 00:02:13.279
<v Speaker 2>know if your machine will make it to the
next planned maintenance event.

37
00:02:15.610 --> 00:02:17.509
<v Speaker 2>So, let's get into what is Sensii.

38
00:02:18.710 --> 00:02:23.789
<v Speaker 2>Now, even though predictive maintenance is
in Sensai's name, it's important to not

39
00:02:23.789 --> 00:02:24.449
<v Speaker 2>mix up.

40
00:02:25.470 --> 00:02:28.720
<v Speaker 2>And conflate Sensei with being predictive
maintenance.

41
00:02:28.830 --> 00:02:30.399
<v Speaker 2>Predictive maintenance is a strategy.

42
00:02:30.720 --> 00:02:34.500
<v Speaker 2>Senseii is a tool to help a customer
achieve that strategy.

43
00:02:35.710 --> 00:02:41.720
<v Speaker 2>Um, so what it's doing is it's using asset
intelligence, um, to

44
00:02:42.559 --> 00:02:43.059
<v Speaker 2>To help.

45
00:02:44.619 --> 00:02:48.350
<v Speaker 2>Increase like efficiency, essentially, I
like to say that one.

46
00:02:49.130 --> 00:02:51.880
<v Speaker 2>it helps increase uptime of your machines.

47
00:02:52.080 --> 00:02:58.169
<v Speaker 2>It reduces planned and reactive
maintenance, which reduces costs, um, supports

48
00:02:58.169 --> 00:03:01.729
<v Speaker 2>remote work, and increases sustainability.

49
00:03:04.190 --> 00:03:08.830
<v Speaker 2>So let's just get into how it works a
little bit before we get into how we onboard

50
00:03:08.830 --> 00:03:11.669
<v Speaker 2>and start somebody with using a, a
technology like Sensai.

51
00:03:13.460 --> 00:03:16.020
<v Speaker 2>So this is sort of the roadmap of what
Sensii is.

52
00:03:16.259 --> 00:03:17.660
<v Speaker 2>I have this laser turned on.

53
00:03:17.740 --> 00:03:20.880
<v Speaker 2>I'm gonna turn that off so I can click and
you don't get distracted.

54
00:03:21.660 --> 00:03:26.380
<v Speaker 2>So the way Sensii works is the first thing
that happens is we have to take data from

55
00:03:26.380 --> 00:03:27.080
<v Speaker 2>your machines.

56
00:03:27.220 --> 00:03:31.380
<v Speaker 2>Now, this could include a sensor data that
is coming directly from the machine

57
00:03:31.380 --> 00:03:32.619
<v Speaker 2>itself while it's running.

58
00:03:33.039 --> 00:03:37.800
<v Speaker 2>It can include information from the PLC
like, you know, control signals, and it can

59
00:03:37.800 --> 00:03:40.360
<v Speaker 2>also include information from maintenance.

60
00:03:40.479 --> 00:03:43.880
<v Speaker 2>So as you're interacting with the machine,
if there are hands on the machine, you're

61
00:03:43.880 --> 00:03:47.669
<v Speaker 2>gonna want to tell that information to
Sensi so that it knows to see changes in the

62
00:03:47.669 --> 00:03:48.070
<v Speaker 2>data.

63
00:03:48.559 --> 00:03:52.320
<v Speaker 2>All of this comes into uh our cloud-based
application.

64
00:03:53.039 --> 00:03:55.500
<v Speaker 2>And it's fed into our pattern engines.

65
00:03:56.449 --> 00:04:01.770
<v Speaker 2>So the pattern engines are, I would say
half of the, you know, the AI and the, the

66
00:04:01.770 --> 00:04:02.750
<v Speaker 2>magic of Sensii.

67
00:04:02.830 --> 00:04:06.770
<v Speaker 2>I kind of hesitate to use that word
because there's a lot of logic that goes into

68
00:04:06.770 --> 00:04:07.190
<v Speaker 2>it.

69
00:04:07.479 --> 00:04:11.250
<v Speaker 2>Um, so the pattern engines take all of the
data that you're feeding in.

70
00:04:12.210 --> 00:04:17.649
<v Speaker 2>They establish what a normal baseline
looks like for your data, and it takes minimum

71
00:04:17.649 --> 00:04:20.910
<v Speaker 2>2 days to do that for some of our more
basic patterns.

72
00:04:22.019 --> 00:04:24.899
<v Speaker 2>Uh, so they establish what a normal sort
of fingerprint looks like.

73
00:04:25.059 --> 00:04:29.100
<v Speaker 2>And then as your data and, as it moves
forward in time, deviates from that

74
00:04:29.100 --> 00:04:34.130
<v Speaker 2>fingerprint, The app will generate
patterns to categorize those changes in behavior.

75
00:04:35.589 --> 00:04:40.279
<v Speaker 2>So these pattern engines can run
automatically with no additional input.

76
00:04:41.399 --> 00:04:44.920
<v Speaker 2>Um, but you can also feed in, if you've
seen failures in the past that are

77
00:04:44.920 --> 00:04:48.170
<v Speaker 2>consistent and you have the right data for
them, you can feed that in.

78
00:04:48.980 --> 00:04:52.899
<v Speaker 2>Uh, to match to them in the future, and
you can feed in expert knowledge as well,

79
00:04:53.140 --> 00:04:56.540
<v Speaker 2>additional information, whatever else you
know about this machine and how it fails

80
00:04:56.540 --> 00:05:00.359
<v Speaker 2>and how it looks in the data, you can give
that all to the system in various ways.

81
00:05:02.429 --> 00:05:06.149
<v Speaker 2>So, coming out from the pattern engines
are all these different types of potential

82
00:05:06.149 --> 00:05:07.829
<v Speaker 2>patterns that can be detected.

83
00:05:08.820 --> 00:05:14.760
<v Speaker 2>Um, I'll show a couple examples of the
most, you know, default ones.

84
00:05:14.959 --> 00:05:16.880
<v Speaker 2>So the top is anomaly.

85
00:05:17.079 --> 00:05:21.799
<v Speaker 2>Now, anomaly is a short-term, short
duration change.

86
00:05:21.880 --> 00:05:25.839
<v Speaker 2>So there's a step change or a spike or a
burst of spikes like what we're seeing

87
00:05:25.839 --> 00:05:26.220
<v Speaker 2>here.

88
00:05:27.940 --> 00:05:32.799
<v Speaker 2>Uh, the next one is a trend, which is a
long-term baseline shift in the data.

89
00:05:32.880 --> 00:05:34.079
<v Speaker 2>It can be going up or down.

90
00:05:34.369 --> 00:05:38.959
<v Speaker 2>Um, those two are the ones that are gonna
be run automatically out of the box.

91
00:05:39.040 --> 00:05:43.540
<v Speaker 2>You feed your data into Sensa, it
automatically starts looking for these with no

92
00:05:43.959 --> 00:05:44.880
<v Speaker 2>configuration.

93
00:05:45.440 --> 00:05:50.420
<v Speaker 2>Um, the caveat is you've got to be
providing the app with condition indicator data,

94
00:05:51.040 --> 00:05:54.519
<v Speaker 2>uh, which we work with our customers to
help derive those and figure out how to do

95
00:05:54.519 --> 00:05:54.920
<v Speaker 2>that.

96
00:05:55.440 --> 00:06:01.359
<v Speaker 2>Um, what a condition indicator is, it's
data that in normal operation of the

97
00:06:01.359 --> 00:06:05.119
<v Speaker 2>machine, this is oversimplified, but in
normal operation of the machine, the data is

98
00:06:05.119 --> 00:06:05.609
<v Speaker 2>steady.

99
00:06:06.380 --> 00:06:10.510
<v Speaker 2>And when the machine is in a state of
degradation or failure, it's not steady.

100
00:06:10.750 --> 00:06:12.100
<v Speaker 2>It's doing something different.

101
00:06:12.470 --> 00:06:17.769
<v Speaker 2>So a change in the data represents a
change in the, the condition of the machine.

102
00:06:18.809 --> 00:06:22.489
<v Speaker 2>So we help our customers to drive those
and make sure we're feeding that into the

103
00:06:22.489 --> 00:06:25.529
<v Speaker 2>system in order to get these anomalies and
trends.

104
00:06:27.399 --> 00:06:32.149
<v Speaker 2>Uh, the other most basic type of pattern
that we detect does require some expert

105
00:06:32.149 --> 00:06:33.790
<v Speaker 2>knowledge, which would be a threshold.

106
00:06:34.000 --> 00:06:38.959
<v Speaker 2>So if you see this gray line going across
here, I'm not sure how easy that is to

107
00:06:38.959 --> 00:06:39.339
<v Speaker 2>see.

108
00:06:40.170 --> 00:06:42.190
<v Speaker 2>That indicates sort of like an alarm
level.

109
00:06:42.839 --> 00:06:47.489
<v Speaker 2>Um, if your data crosses a threshold, you
will get a notification every time.

110
00:06:48.690 --> 00:06:51.760
<v Speaker 2>Um, you only get one for as long as it's
violating the threshold.

111
00:06:51.820 --> 00:06:57.089
<v Speaker 2>So what we see here in green is basically
the notification that you would be given

112
00:06:57.089 --> 00:06:58.290
<v Speaker 2>for that duration of time.

113
00:07:00.429 --> 00:07:04.709
<v Speaker 2>Um, we do have a bunch of other patterns
that we can detect.

114
00:07:04.869 --> 00:07:08.829
<v Speaker 2>Uh, another one related to threshold, I
don't have another image, but it's called a

115
00:07:08.829 --> 00:07:14.450
<v Speaker 2>forecast, where you can enable Um, an
alert to be given to you if your data starts

116
00:07:14.450 --> 00:07:16.279
<v Speaker 2>trending towards a threshold.

117
00:07:16.959 --> 00:07:20.399
<v Speaker 2>And it will tell you approximately how
much time you have until it gets to the

118
00:07:20.399 --> 00:07:23.799
<v Speaker 2>threshold, you know, depending on whether
or not you're interacting with the machine

119
00:07:23.799 --> 00:07:25.239
<v Speaker 2>to get it to go away again.

120
00:07:26.130 --> 00:07:27.619
<v Speaker 2>Uh, so those are 4.

121
00:07:28.720 --> 00:07:31.410
<v Speaker 2>Um, basic types of patterns that are
detected.

122
00:07:31.769 --> 00:07:37.609
<v Speaker 2>Now, aside from thresholds, when an
anomaly or a trend is detected, you don't

123
00:07:37.609 --> 00:07:41.450
<v Speaker 2>necessarily automatically get a
notification, which in Sens that we call a case.

124
00:07:42.010 --> 00:07:46.769
<v Speaker 2>Um, instead, all of this pattern
information is fed into our attention engine.

125
00:07:48.290 --> 00:07:52.399
<v Speaker 2>So this information comes in here along
with configurations such as asset

126
00:07:52.399 --> 00:07:53.149
<v Speaker 2>criticality.

127
00:07:54.070 --> 00:07:58.100
<v Speaker 2>So if you set an asset as being highly
critical, you'll get a case a lot sooner than

128
00:07:58.100 --> 00:08:00.570
<v Speaker 2>an asset that you said is low criticality.

129
00:08:01.429 --> 00:08:04.179
<v Speaker 2>Um, there's additional configuration we
can do as well.

130
00:08:05.019 --> 00:08:10.619
<v Speaker 2>So all of that comes in, and if it's
determined that the change in your asset is

131
00:08:10.619 --> 00:08:12.519
<v Speaker 2>critical enough, it will raise a case.

132
00:08:14.130 --> 00:08:14.730
<v Speaker 2>Over here.

133
00:08:15.859 --> 00:08:21.420
<v Speaker 2>Um, this case opening is dependent on a
value of attention index, which is just a

134
00:08:21.420 --> 00:08:24.339
<v Speaker 2>value of how how severe.

135
00:08:25.149 --> 00:08:28.440
<v Speaker 2>Um, the change in the asset is, how much
it needs your attention.

136
00:08:29.989 --> 00:08:35.429
<v Speaker 2>Each case will come along with evidence,
uh, which is basically just a, a chart of

137
00:08:35.429 --> 00:08:38.390
<v Speaker 2>the data that has a pattern on it.

138
00:08:38.590 --> 00:08:42.549
<v Speaker 2>So you can review the evidence and provide
feedback.

139
00:08:43.599 --> 00:08:49.320
<v Speaker 2>In the form of case evidence and case
closure, and, and even in just interacting

140
00:08:49.320 --> 00:08:54.159
<v Speaker 2>with someone like me who can go in and
tune the pattern engines and the attention

141
00:08:54.159 --> 00:08:57.280
<v Speaker 2>engine to improve uh detections in the
future.

142
00:08:59.239 --> 00:09:00.700
<v Speaker 2>I forgot to click, sorry.

143
00:09:02.809 --> 00:09:07.280
<v Speaker 2>Um, I hope, hopefully that all makes
sense.

144
00:09:07.359 --> 00:09:11.640
<v Speaker 2>That's the, a kind of oversimplification
of what's going on here, but I think it's a

145
00:09:11.640 --> 00:09:12.559
<v Speaker 2>pretty good roadmap.

146
00:09:14.469 --> 00:09:14.950
<v Speaker 2>OK.

147
00:09:17.179 --> 00:09:18.780
<v Speaker 2>There's no questions, right, Liz?

148
00:09:19.260 --> 00:09:20.479
<v Liz McGinn>Nope, no questions so far.

149
00:09:20.820 --> 00:09:21.099
<v Liz McGinn>Okey

150
00:09:21.099 --> 00:09:21.880
<v Speaker 2>doke, thanks.

151
00:09:22.219 --> 00:09:26.380
<v Speaker 2>In that case, uh, if you have any
questions on how it works, drop them in the, the

152
00:09:26.380 --> 00:09:28.859
<v Speaker 2>questions chat, and we can get to them.

153
00:09:29.599 --> 00:09:30.390
<v Speaker 2>In a bit.

154
00:09:30.679 --> 00:09:33.020
<v Speaker 2>Otherwise, let's just get into our
process.

155
00:09:34.190 --> 00:09:40.390
<v Speaker 2>Now, it's all like since I Um, We, we want
to take a lot of care when we onboard a

156
00:09:40.390 --> 00:09:46.520
<v Speaker 2>customer with Sensei because the, you
know, how well we onboard can affect the

157
00:09:46.520 --> 00:09:48.020
<v Speaker 2>amount of success that we have.

158
00:09:48.799 --> 00:09:53.020
<v Speaker 2>So, we have something called the Senseii
Knowledge Platform, and this was built by

159
00:09:53.020 --> 00:09:58.039
<v Speaker 2>us using our best practices that we've
developed over years of, of doing these

160
00:09:58.039 --> 00:09:58.940
<v Speaker 2>deployments.

161
00:09:59.320 --> 00:10:03.840
<v Speaker 2>It has 5 phases, scope, design, deploy,
operate, and refine.

162
00:10:04.690 --> 00:10:08.570
<v Speaker 2>Scope is pretty much everything that's
done prior to a project charter.

163
00:10:08.950 --> 00:10:11.669
<v Speaker 2>So that's before someone like me would get
involved.

164
00:10:12.010 --> 00:10:17.369
<v Speaker 2>Um, That's done to figure out, you know,
what assets are involved, um, what sort of

165
00:10:17.369 --> 00:10:19.130
<v Speaker 2>benefits might a customer have.

166
00:10:19.859 --> 00:10:22.320
<v Speaker 2>And as soon as the project charter is
signed.

167
00:10:23.320 --> 00:10:25.229
<v Speaker 2>Someone like me in delivery will come in.

168
00:10:25.969 --> 00:10:28.590
<v Speaker 2>And start designing the project.

169
00:10:28.799 --> 00:10:29.130
<v Speaker 2>OK?

170
00:10:29.250 --> 00:10:32.289
<v Speaker 2>So, design phase is exactly what it sounds
like.

171
00:10:32.369 --> 00:10:37.869
<v Speaker 2>We will do a number of asset discovery
sessions to figure out what are your assets.

172
00:10:38.369 --> 00:10:42.539
<v Speaker 2>Um, a lot of times we have to talk with
customers to figure out what do you define

173
00:10:42.539 --> 00:10:46.090
<v Speaker 2>as an asset and what do we define as an
asset, and how can we sort of consolidate

174
00:10:46.090 --> 00:10:46.770
<v Speaker 2>those two.

175
00:10:48.039 --> 00:10:51.760
<v Speaker 2>We'll talk about what data you have
available, how do they fail, you know, what are

176
00:10:51.760 --> 00:10:54.960
<v Speaker 2>the pain points that you have with these
assets, and how best can we configure them

177
00:10:54.960 --> 00:10:57.239
<v Speaker 2>in the app to get you the results that
you're looking for.

178
00:10:59.080 --> 00:11:02.840
<v Speaker 2>As we're doing these asset discovery
sessions, we're also having integration

179
00:11:02.840 --> 00:11:05.179
<v Speaker 2>discussions to figure out how are you
gonna get your data to us.

180
00:11:06.520 --> 00:11:11.679
<v Speaker 2>Deploy phase is all about getting things
in the app, um, get the customer account

181
00:11:11.679 --> 00:11:16.640
<v Speaker 2>configured, start getting data in,
configure the assets, and do training.

182
00:11:18.239 --> 00:11:23.900
<v Speaker 2>Operate phase is just using the app, that
is your regular daily use.

183
00:11:24.200 --> 00:11:28.919
<v Speaker 2>Um, and we have different things to help
with that, and we, we also help with case

184
00:11:28.919 --> 00:11:30.599
<v Speaker 2>review sessions as we go.

185
00:11:30.679 --> 00:11:36.039
<v Speaker 2>I think one of the really important things
to note about Sensai is that we don't

186
00:11:36.039 --> 00:11:40.159
<v Speaker 2>just hand you the app in the, the, you
know, these steps here and say, all right,

187
00:11:40.479 --> 00:11:41.239
<v Speaker 2>see you, good luck.

188
00:11:41.400 --> 00:11:42.340
<v Speaker 2>Figure it out yourself.

189
00:11:42.679 --> 00:11:44.419
<v Speaker 2>We work with our customers.

190
00:11:46.429 --> 00:11:50.690
<v Speaker 2>For, forever and help them with their case
reviews, help, you know, get feedback,

191
00:11:50.830 --> 00:11:51.940
<v Speaker 2>what do you like, what do you not like?

192
00:11:51.950 --> 00:11:52.590
<v Speaker 2>How is this working?

193
00:11:52.710 --> 00:11:53.710
<v Speaker 2>How can we improve it?

194
00:11:54.109 --> 00:11:59.270
<v Speaker 2>And that feeds into refined phase, which
is when um the account relationship is in

195
00:11:59.270 --> 00:12:01.289
<v Speaker 2>the hands of a customer success manager.

196
00:12:02.530 --> 00:12:04.690
<v Speaker 2>And really, it continues the process.

197
00:12:04.770 --> 00:12:06.250
<v Speaker 2>It's really a cyclical.

198
00:12:06.690 --> 00:12:10.929
<v Speaker 2>So you might go back to redesign some
assets if they're not performing very well.

199
00:12:11.010 --> 00:12:16.909
<v Speaker 2>You might scale and add on more assets,
um, And, and have those sorts of

200
00:12:16.909 --> 00:12:17.590
<v Speaker 2>discussions.

201
00:12:20.570 --> 00:12:23.929
<v Speaker 2>So this is sort of an example of what a
project deployment could look like.

202
00:12:24.039 --> 00:12:26.520
<v Speaker 2>Again, I'm not gonna read all this again
because I kind of went over that.

203
00:12:27.010 --> 00:12:32.270
<v Speaker 2>But I wanna draw your attention to the,
the timeline here with all these little

204
00:12:32.320 --> 00:12:36.650
<v Speaker 2>cyclical aspects, um, because it's not
linear, and it's kind of really hard to

205
00:12:36.650 --> 00:12:37.390
<v Speaker 2>represent that.

206
00:12:38.229 --> 00:12:43.030
<v Speaker 2>And each of the steps that we have for
design, deploy, operate, and refine, we just

207
00:12:43.030 --> 00:12:43.940
<v Speaker 2>use as needed.

208
00:12:44.270 --> 00:12:49.710
<v Speaker 2>So there's a step in deploy, for instance,
for procuring sensors.

209
00:12:49.780 --> 00:12:52.619
<v Speaker 2>Well, if a customer already has sensors,
we're not going to do that step, right?

210
00:12:52.789 --> 00:12:59.190
<v Speaker 2>So this deployment plan is really made
custom per customer as we start working with

211
00:12:59.190 --> 00:12:59.469
<v Speaker 2>them.

212
00:13:01.080 --> 00:13:02.580
<v Speaker 2>Um, so you'll notice here.

213
00:13:03.320 --> 00:13:07.460
<v Speaker 2>That right when the kickoff meeting
happens, that is when delivery gets involved.

214
00:13:08.210 --> 00:13:11.289
<v Speaker 2>Um, and we start setting up a timeline.

215
00:13:11.409 --> 00:13:12.250
<v Speaker 2>What do we do next?

216
00:13:12.369 --> 00:13:13.929
<v Speaker 2>Who are people involved?

217
00:13:14.330 --> 00:13:17.049
<v Speaker 2>And then we start getting assets
onboarded.

218
00:13:17.859 --> 00:13:21.929
<v Speaker 2>Um, and even as we're onboarding assets,
we might still be going through the design

219
00:13:21.929 --> 00:13:23.960
<v Speaker 2>phase of other assets.

220
00:13:24.750 --> 00:13:28.710
<v Speaker 2>Um, as we're working and operate with
some, we may still be deploying some others,

221
00:13:28.789 --> 00:13:29.969
<v Speaker 2>so it's very, very cyclical.

222
00:13:31.270 --> 00:13:36.549
<v Speaker 2>Um, until we get into sort of a
steady-state period, um, in which we would just be

223
00:13:36.549 --> 00:13:38.309
<v Speaker 2>doing sort of periodic check-ins.

224
00:13:38.559 --> 00:13:41.789
<v Speaker 2>Customers don't need us so much, but we
are there when they do need us.

225
00:13:46.700 --> 00:13:47.440
<v Speaker 2>All right, let's.

226
00:13:48.440 --> 00:13:48.729
<v Liz McGinn>Awesome.

227
00:13:48.770 --> 00:13:50.830
<v Liz McGinn>So that will pull us into our first poll.

228
00:13:51.289 --> 00:13:55.549
<v Liz McGinn>Um, so our poll question will, uh, ask it
so everyone can answer.

229
00:13:55.729 --> 00:13:58.549
<v Liz McGinn>Natalie will go into a little bit more and
then we'll come back to the answer.

230
00:13:58.809 --> 00:14:02.729
<v Liz McGinn>So if you only had 50 assets you could
onboard into Sensai, what would you choose?

231
00:14:02.919 --> 00:14:07.320
<v Liz McGinn>The complex assets that cause the most
issues in downtime, simpler assets of the

232
00:14:07.320 --> 00:14:11.799
<v Liz McGinn>same type, simpler assets of different
types, or assets that are interesting to end

233
00:14:11.799 --> 00:14:13.450
<v Liz McGinn>users regardless of the type.

234
00:14:16.640 --> 00:14:20.239
<v Liz McGinn>And while you guys are answering that,
Natalie will go into the design phase.

235
00:14:21.260 --> 00:14:21.909
<v Speaker 2>Yes, ma'am.

236
00:14:23.210 --> 00:14:23.369
<v Speaker 2>All right.

237
00:14:23.489 --> 00:14:24.429
<v Speaker 2>So take your time with that.

238
00:14:24.489 --> 00:14:26.789
<v Speaker 2>We will get to your answers in just a
minute.

239
00:14:28.710 --> 00:14:32.330
<v Speaker 2>As we pull up our description of what we
consider design phase.

240
00:14:32.590 --> 00:14:37.849
<v Speaker 2>So design phase consists of 4 steps, and
again, they might not happen in this order.

241
00:14:38.219 --> 00:14:40.830
<v Speaker 2>Um, I'm working with a new customer right
now.

242
00:14:40.909 --> 00:14:45.270
<v Speaker 2>We are doing machine selection, failure
mode, and data selection as one piece, and

243
00:14:45.270 --> 00:14:48.109
<v Speaker 2>data architecture is happening in parallel
with another group of people.

244
00:14:48.270 --> 00:14:51.929
<v Speaker 2>So just as an example of how we might
custom deliver a project.

245
00:14:53.349 --> 00:14:58.869
<v Speaker 2>So the whole point of design is to choose
what assets we want to onboard, understand

246
00:14:58.869 --> 00:14:59.950
<v Speaker 2>their failure modes.

247
00:15:00.840 --> 00:15:05.200
<v Speaker 2>Um, understand how their data relates to
those failure modes and figure out how

248
00:15:05.200 --> 00:15:06.559
<v Speaker 2>we're going to configure it in the app.

249
00:15:06.640 --> 00:15:10.590
<v Speaker 2>It might require filtering the data, it
might require transformations, which we can

250
00:15:10.590 --> 00:15:11.299
<v Speaker 2>do in-app.

251
00:15:12.210 --> 00:15:16.090
<v Speaker 2>Um, and from there we, we decide how we
want to integrate.

252
00:15:16.140 --> 00:15:18.159
<v Speaker 2>It's really how the customer wants to
integrate.

253
00:15:18.500 --> 00:15:21.460
<v Speaker 2>And very importantly, we determine success
criteria.

254
00:15:21.539 --> 00:15:26.460
<v Speaker 2>We want to do this early on, um, so that
we don't get to, you know, renewal a year,

255
00:15:26.539 --> 00:15:30.059
<v Speaker 2>a year later, and, you know, they're,
they're not sure if they've had success or

256
00:15:30.059 --> 00:15:30.219
<v Speaker 2>not.

257
00:15:30.340 --> 00:15:35.080
<v Speaker 2>So we start that early on to make sure
that we know what we're looking out for.

258
00:15:36.429 --> 00:15:41.429
<v Speaker 2>In design phase, typically the people
involved would be the machine experts, because

259
00:15:41.429 --> 00:15:44.590
<v Speaker 2>we need to talk to somebody who knows the
machines and knows what the data might

260
00:15:44.590 --> 00:15:49.289
<v Speaker 2>look like for them and um so we can really
work together to figure out how to

261
00:15:49.289 --> 00:15:50.049
<v Speaker 2>monitor them.

262
00:15:50.760 --> 00:15:53.070
<v Speaker 2>And the Siemens delivery team.

263
00:15:53.409 --> 00:15:58.640
<v Speaker 2>I will say in design, I don't want to
confuse anyone and have them thinking that we

264
00:15:58.640 --> 00:16:01.750
<v Speaker 2>are building custom models per asset.

265
00:16:02.169 --> 00:16:05.090
<v Speaker 2>Um, the AI and machine learning builds the
model.

266
00:16:05.130 --> 00:16:06.340
<v Speaker 2>It does it on its own.

267
00:16:06.729 --> 00:16:10.190
<v Speaker 2>All we are trying to figure out is what is
the condition indicator.

268
00:16:10.380 --> 00:16:14.409
<v Speaker 2>And for instance, if you're onboarding,
let's say vibration data.

269
00:16:15.109 --> 00:16:20.280
<v Speaker 2>OK, it's a very common condition
indicator, but as it is, coming straight from the

270
00:16:20.280 --> 00:16:23.520
<v Speaker 2>machine, it might not be a condition
indicator because it might be including data

271
00:16:23.520 --> 00:16:24.700
<v Speaker 2>when the machine is off.

272
00:16:25.400 --> 00:16:29.440
<v Speaker 2>So, the first easy step is, OK, remove
data when the machine is off, so we can

273
00:16:29.440 --> 00:16:32.919
<v Speaker 2>filter that out and be providing a steady
signal that's not dropping all over the

274
00:16:32.919 --> 00:16:33.140
<v Speaker 2>place.

275
00:16:34.190 --> 00:16:38.349
<v Speaker 2>Um, the other thing to consider with
vibration is, is it dependent on any external

276
00:16:38.349 --> 00:16:39.250
<v Speaker 2>factors?

277
00:16:39.739 --> 00:16:44.590
<v Speaker 2>If you have a motor running on a VFD every
time your speed change, changes, your

278
00:16:44.590 --> 00:16:47.659
<v Speaker 2>vibration is going to change, but it
doesn't necessarily indicate failure.

279
00:16:48.020 --> 00:16:51.880
<v Speaker 2>So we can set up that correlation in the
app between speed and vibration.

280
00:16:52.229 --> 00:16:56.809
<v Speaker 2>That's what we're talking about, um,
rather than actually going in and configuring,

281
00:16:57.349 --> 00:17:01.909
<v Speaker 2>um, the actual algorithms that are looking
for patterns, we just try to drive

282
00:17:01.909 --> 00:17:03.349
<v Speaker 2>condition indicators early on.

283
00:17:05.699 --> 00:17:07.869
<v Speaker 2>Liz, do you want to read the results of
the poll?

284
00:17:09.229 --> 00:17:10.609
<v Speaker 2>Or do you want to give it a minute?

285
00:17:11.849 --> 00:17:12.520
<v Liz McGinn>Nope, it's up.

286
00:17:12.650 --> 00:17:18.569
<v Liz McGinn>Um, so the audience has chosen the complex
assets that cause the most issues in

287
00:17:18.569 --> 00:17:22.160
<v Liz McGinn>downtime as the, um, choice for the 50
assets.

288
00:17:22.239 --> 00:17:23.630
<v Liz McGinn>What do you think about that, Natalie?

289
00:17:24.489 --> 00:17:24.760
<v Speaker 2>OK.

290
00:17:24.839 --> 00:17:29.300
<v Speaker 2>I think that's a really common route that
our customers usually want to go down.

291
00:17:29.839 --> 00:17:32.280
<v Speaker 2>And we actually try to steer them away
from that.

292
00:17:32.439 --> 00:17:39.189
<v Speaker 2>Um, those most problematic assets, are not
the best choice usually for two

293
00:17:39.189 --> 00:17:39.839
<v Speaker 2>reasons.

294
00:17:40.189 --> 00:17:42.839
<v Speaker 2>Number one, if they're all, if they're
really problematic, they're probably being

295
00:17:42.839 --> 00:17:44.930
<v Speaker 2>monitored really heavily already.

296
00:17:45.510 --> 00:17:47.349
<v Speaker 2>So what value is Sensi going to show?

297
00:17:47.469 --> 00:17:50.349
<v Speaker 2>It might just prove that it works, but we
know it, it, it works.

298
00:17:50.500 --> 00:17:55.459
<v Speaker 2>So, um, it might not be actually adding
much value cause you're already monitoring

299
00:17:55.459 --> 00:17:55.819
<v Speaker 2>it.

300
00:17:56.270 --> 00:18:00.829
<v Speaker 2>The other reason we wouldn't want to start
with that one is because it might be a

301
00:18:00.829 --> 00:18:02.209
<v Speaker 2>very complex asset.

302
00:18:02.599 --> 00:18:07.020
<v Speaker 2>And in the very beginning, we are coming
from very separate spheres of knowledge.

303
00:18:07.439 --> 00:18:11.270
<v Speaker 2>I know the app really, really well, but I
don't know anything about your machines.

304
00:18:11.680 --> 00:18:14.439
<v Speaker 2>You know your machines really, really
well, but you don't know anything about the

305
00:18:14.439 --> 00:18:14.660
<v Speaker 2>app.

306
00:18:14.750 --> 00:18:18.099
<v Speaker 2>At some point, we have to kind of learn
about each other.

307
00:18:18.719 --> 00:18:22.599
<v Speaker 2>And once you kind of know what the app
needs to look like, what it needs to see in

308
00:18:22.599 --> 00:18:29.040
<v Speaker 2>the data to be successful, then we can
really work together a lot better to derive

309
00:18:29.040 --> 00:18:30.339
<v Speaker 2>those condition indicators.

310
00:18:31.189 --> 00:18:36.750
<v Speaker 2>Um, so we would usually recommend, let's
start with simpler assets that are more

311
00:18:36.750 --> 00:18:37.270
<v Speaker 2>basic.

312
00:18:37.310 --> 00:18:40.989
<v Speaker 2>And honestly, if you start with simpler
assets of the same type, get them all on the

313
00:18:40.989 --> 00:18:45.569
<v Speaker 2>app, get them generating value, while we
then go back and revisit your more complex

314
00:18:45.569 --> 00:18:49.910
<v Speaker 2>assets, so the app is working for you
while we try to figure out those condition

315
00:18:49.910 --> 00:18:50.790
<v Speaker 2>indicators.

316
00:18:51.750 --> 00:18:52.579
<v Speaker 2>Does that make sense?

317
00:18:53.489 --> 00:18:54.390
<v Liz McGinn>That makes great sense.

318
00:18:54.489 --> 00:18:57.569
<v Liz McGinn>Could you explain a little bit on like
what is an asset, because I know you spoke

319
00:18:57.569 --> 00:19:00.449
<v Liz McGinn>about it, um, that we have two different
definitions usually from our customers.

320
00:19:00.530 --> 00:19:04.589
<v Liz McGinn>So maybe it's best to kind of explain what
an asset is in our eyes.

321
00:19:04.810 --> 00:19:07.109
<v Speaker 2>I'm really glad you reminded me to do
that.

322
00:19:07.569 --> 00:19:12.869
<v Speaker 2>Um, so an asset according to Sensai is the
smallest replaceable unit of a machine

323
00:19:13.290 --> 00:19:14.170
<v Speaker 2>that has data.

324
00:19:14.290 --> 00:19:15.869
<v Speaker 2>It's the thing that you're maintaining.

325
00:19:16.849 --> 00:19:21.250
<v Speaker 2>Um, you don't want to have like some big,
let's say, let's say you have something

326
00:19:21.250 --> 00:19:24.969
<v Speaker 2>that you call a fan, and you say, all
right, I want to get this fan into the app.

327
00:19:25.089 --> 00:19:29.550
<v Speaker 2>But actually, what you're calling a fan is
a motor, a gearbox, and an impeller.

328
00:19:30.130 --> 00:19:31.760
<v Speaker 2>To us, that's 3 assets.

329
00:19:32.170 --> 00:19:36.800
<v Speaker 2>If you have data specifically on the
motor, and specifically on the gearbox and the

330
00:19:36.800 --> 00:19:39.859
<v Speaker 2>impeller, that is 3 assets, and we should
monitor them separately.

331
00:19:40.170 --> 00:19:43.329
<v Speaker 2>So you're getting cases that point you to
exactly where the problem is.

332
00:19:44.290 --> 00:19:47.770
<v Speaker 2>Also, when you split them up, driving
condition indicators is can be a little bit

333
00:19:47.770 --> 00:19:52.010
<v Speaker 2>simpler than for a wider, like a larger
machine that might be more heavily

334
00:19:52.010 --> 00:19:54.390
<v Speaker 2>influenced by process variables.

335
00:19:55.310 --> 00:19:56.349
<v Speaker 2>Not always the case.

336
00:19:57.790 --> 00:20:00.369
<v Speaker 2>Um, so that's what we would consider to be
an asset.

337
00:20:01.290 --> 00:20:04.650
<v Speaker 2>Um, I would also say from that poll.

338
00:20:05.510 --> 00:20:12.160
<v Speaker 2>Um, While we do tend to start with simpler
assets of the same type, Sensai works

339
00:20:12.160 --> 00:20:12.920
<v Speaker 2>great in bulk.

340
00:20:13.050 --> 00:20:17.640
<v Speaker 2>So if you want to onboard one motor, or
100 motors, to me it's the same amount of

341
00:20:17.640 --> 00:20:18.089
<v Speaker 2>work.

342
00:20:18.869 --> 00:20:21.319
<v Speaker 2>Um, so I would just say get 100 of those
in.

343
00:20:21.510 --> 00:20:24.439
<v Speaker 2>It's the same amount of configuration if
they're basically the same thing.

344
00:20:25.380 --> 00:20:27.469
<v Speaker 2>Get 100 of those in, get them generating
value.

345
00:20:27.579 --> 00:20:30.000
<v Speaker 2>Let's go back and look at the more complex
ones once that's done.

346
00:20:30.959 --> 00:20:36.439
<v Speaker 2>Um, but the other thing that's important
not to overlook is to consider assets that

347
00:20:36.439 --> 00:20:39.560
<v Speaker 2>are actually interesting to the end users,
regardless of type.

348
00:20:39.800 --> 00:20:40.619
<v Speaker 2>We want to balance.

349
00:20:41.109 --> 00:20:43.880
<v Speaker 2>Whoever's using Sensii should care about
what's in there.

350
00:20:44.020 --> 00:20:46.500
<v Speaker 2>So, we try to balance the two.

351
00:20:48.959 --> 00:20:53.930
<v Speaker 2>Um, so I would say that's one of the main
roadblocks is just what assets to pick.

352
00:20:54.050 --> 00:20:58.469
<v Speaker 2>If we go the wrong route, it can really
influence the success of the project.

353
00:20:59.459 --> 00:21:03.300
<v Speaker 2>Uh, and we've done this so many times that
we kind of, we kind of know, let's start

354
00:21:03.300 --> 00:21:06.619
<v Speaker 2>with the simple ones that you have many
of, and a few that are interesting to the

355
00:21:06.619 --> 00:21:08.060
<v Speaker 2>end users if they're not the same.

356
00:21:08.790 --> 00:21:09.930
<v Speaker 2>And let's go that route.

357
00:21:10.729 --> 00:21:17.209
<v Speaker 2>Uh, the other roadblock is that the
integration design is always the blocker.

358
00:21:17.250 --> 00:21:19.050
<v Speaker 2>If there's a blocker, it's the
integration.

359
00:21:19.290 --> 00:21:24.329
<v Speaker 2>So, um, it's important to choose, you
know, a partner to do that with, that has a

360
00:21:24.329 --> 00:21:25.430
<v Speaker 2>lot of experience.

361
00:21:25.760 --> 00:21:28.369
<v Speaker 2>Um, so we, we often help with the
integration design.

362
00:21:28.530 --> 00:21:33.119
<v Speaker 2>We have a full, um, developer
documentation website that has all the information

363
00:21:33.119 --> 00:21:33.750
<v Speaker 2>that's needed.

364
00:21:35.219 --> 00:21:37.670
<v Speaker 2>And if we have more questions on that,
just pop them in the chat.

365
00:21:37.859 --> 00:21:42.699
<v Speaker 2>And that's kind of a complicated one, but
we might be able to follow up with you um

366
00:21:42.699 --> 00:21:43.500
<v Speaker 2>outside of the webinar.

367
00:21:46.650 --> 00:21:46.660
<v Liz McGinn>OK.

368
00:21:48.010 --> 00:21:48.530
<v Liz McGinn>Awesome.

369
00:21:48.780 --> 00:21:50.579
<v Liz McGinn>So we are going to poll number 2.

370
00:21:51.689 --> 00:21:54.739
<v Liz McGinn>what is the biggest cause of data quality
issues?

371
00:21:55.170 --> 00:22:00.670
<v Liz McGinn>Samppling frequency, sensor placement,
lack of contextual data or noise in the data.

372
00:22:01.449 --> 00:22:05.719
<v Liz McGinn>Um, before we answer that question, we'll
go into our next phase.

373
00:22:08.869 --> 00:22:13.469
<v Speaker 2>Alright, so the second phase, once we
start discussing our assets and getting data

374
00:22:13.469 --> 00:22:15.739
<v Speaker 2>into the app, that kicks off deploy.

375
00:22:16.650 --> 00:22:20.310
<v Speaker 2>So in this phase, you know, if we need to
install sensors, we do.

376
00:22:20.569 --> 00:22:21.969
<v Speaker 2>I'm sorry, you're not seeing my mouse.

377
00:22:22.170 --> 00:22:22.689
<v Speaker 2>Wrong screen.

378
00:22:22.849 --> 00:22:23.380
<v Speaker 2>There we are.

379
00:22:24.290 --> 00:22:26.239
<v Speaker 2>Um, so we can install sensors.

380
00:22:27.170 --> 00:22:32.099
<v Speaker 2>Um, if needed, we can help with that,
point you in the right direction.

381
00:22:33.339 --> 00:22:35.300
<v Speaker 2>The next step is to pre-process the data.

382
00:22:35.420 --> 00:22:37.540
<v Speaker 2>Now, that's sort of, it's almost a
misnomer.

383
00:22:37.900 --> 00:22:40.500
<v Speaker 2>You can do it on your side before the data
gets to us.

384
00:22:40.579 --> 00:22:44.609
<v Speaker 2>You can do it in an edge device, you can
do it in 100 different ways, but you can

385
00:22:44.609 --> 00:22:47.410
<v Speaker 2>also do it in the app itself after the
data comes in.

386
00:22:47.819 --> 00:22:51.000
<v Speaker 2>So we work with you to figure out what's
the best way to do that.

387
00:22:51.380 --> 00:22:56.069
<v Speaker 2>Um, and again, that pre-processed word,
that just means remove the off states, um,

388
00:22:56.300 --> 00:23:00.040
<v Speaker 2>try to isolate a single state in the data,
basically create a condition indicator.

389
00:23:01.479 --> 00:23:07.040
<v Speaker 2>Get it coming in on board, create the
hierarchy, create your assets, configure them,

390
00:23:07.160 --> 00:23:10.359
<v Speaker 2>change the criticality, you know, we do
all this stuff and we discuss while we do

391
00:23:10.359 --> 00:23:10.560
<v Speaker 2>it.

392
00:23:11.910 --> 00:23:16.459
<v Speaker 2>Um, and then finally we pretty much kick
off with training.

393
00:23:17.670 --> 00:23:20.750
<v Speaker 2>So we, we train the end users, we get them
in, ready to go.

394
00:23:20.989 --> 00:23:23.089
<v Speaker 2>Usually we wait a couple of weeks in
between.

395
00:23:23.859 --> 00:23:30.380
<v Speaker 2>You know, these two, because we don't want
users coming in, while in the beginning,

396
00:23:30.459 --> 00:23:34.060
<v Speaker 2>there's going to be some false positives
as, as the app is learning.

397
00:23:34.219 --> 00:23:36.119
<v Speaker 2>It's absolutely normal.

398
00:23:36.500 --> 00:23:40.140
<v Speaker 2>Um, sometimes the users will have never
even seen their data before if it's new

399
00:23:40.140 --> 00:23:40.939
<v Speaker 2>sensors.

400
00:23:41.260 --> 00:23:45.890
<v Speaker 2>So we want to give the app some time to
learn and start giving good cases, um,

401
00:23:46.099 --> 00:23:48.859
<v Speaker 2>before we get users trained and in so they
don't become disillusioned.

402
00:23:51.000 --> 00:23:52.920
<v Speaker 2>OK, so that is what is deploy.

403
00:23:53.040 --> 00:23:53.979
<v Speaker 2>Who's involved?

404
00:23:55.020 --> 00:24:01.540
<v Speaker 2>IT, uh, integration team and the Sunside
delivery team, as well as possibly machine

405
00:24:01.540 --> 00:24:02.290
<v Speaker 2>experts.

406
00:24:02.410 --> 00:24:06.400
<v Speaker 2>Um, we usually have somebody nominated on
the customer side that we call the PDM

407
00:24:06.400 --> 00:24:10.060
<v Speaker 2>champion, and that's the person kind of
connecting us to everyone who needs to be

408
00:24:10.060 --> 00:24:10.500
<v Speaker 2>connected.

409
00:24:10.599 --> 00:24:14.699
<v Speaker 2>They're involved in everything, and they
might be the first end user, the sort of

410
00:24:14.699 --> 00:24:15.380
<v Speaker 2>power user.

411
00:24:16.369 --> 00:24:19.359
<v Speaker 2>Um, not always the case, but that's often
what we see.

412
00:24:20.890 --> 00:24:24.709
<v Speaker 2>So, Liz, if you want to read the poll
results.

413
00:24:28.229 --> 00:24:29.800
<v Speaker 2>If there are poll results.

414
00:24:40.130 --> 00:24:41.010
<v Speaker 2>Liz, are you there?

415
00:24:43.780 --> 00:24:48.719
<v Liz McGinn>Sorry, um, the audience thinks that the
lack of contextual data, uh, uh, is the

416
00:24:48.719 --> 00:24:50.469
<v Liz McGinn>biggest cause of data quality issues.

417
00:24:50.589 --> 00:24:51.180
<v Liz McGinn>Is that correct?

418
00:24:51.260 --> 00:24:51.670
<v Liz McGinn>Oh,

419
00:24:52.270 --> 00:24:55.989
<v Speaker 2>that this is a trick question because
whatever you say is correct, that that's

420
00:24:55.989 --> 00:24:56.849
<v Speaker 2>correct for you.

421
00:24:57.930 --> 00:24:59.810
<v Speaker 2>Um, they can all be problematic.

422
00:25:00.060 --> 00:25:02.160
<v Speaker 2>I'm just gonna go back just to see that
list again.

423
00:25:02.959 --> 00:25:08.510
<v Speaker 2>Um, sampling frequency, I don't know if
you're taking a data point once a day, that

424
00:25:08.510 --> 00:25:09.939
<v Speaker 2>might not be enough, right?

425
00:25:10.239 --> 00:25:15.369
<v Speaker 2>Um, Kind of need continuous monitoring in
place to really be able to fully utilize

426
00:25:15.369 --> 00:25:16.969
<v Speaker 2>sensi sensor placement.

427
00:25:17.010 --> 00:25:21.290
<v Speaker 2>If they're not placed well or if they're
placed inconsistently, that could be a

428
00:25:21.290 --> 00:25:21.989
<v Speaker 2>problem.

429
00:25:22.449 --> 00:25:26.630
<v Speaker 2>Lack of contextual data doesn't allow us
often to isolate a state in the data.

430
00:25:27.000 --> 00:25:33.329
<v Speaker 2>Um, if your data is, is influenced by
outside factors, you want to be able to

431
00:25:33.329 --> 00:25:37.910
<v Speaker 2>account for that in, in, you know, in the
processing so that you're not being told

432
00:25:38.329 --> 00:25:40.170
<v Speaker 2>for temperature and a bearing, for
instance.

433
00:25:40.619 --> 00:25:43.500
<v Speaker 2>You know, that your bearing is in trouble,
but really just it's getting hotter

434
00:25:43.500 --> 00:25:43.680
<v Speaker 2>outside.

435
00:25:44.800 --> 00:25:48.670
<v Speaker 2>So if you're monitoring temperatures, as
an example, you probably wanna have ambient

436
00:25:48.670 --> 00:25:49.739
<v Speaker 2>temperature as well.

437
00:25:50.819 --> 00:25:55.939
<v Speaker 2>Um, noise in the data, you know, that can
be, that can actually be a product of even

438
00:25:55.939 --> 00:25:58.500
<v Speaker 2>sensor placement and lack of contextual
data.

439
00:25:59.339 --> 00:26:03.290
<v Speaker 2>But again, Sensei is an incredible tool,
but if you feed it.

440
00:26:04.790 --> 00:26:08.270
<v Speaker 2>Terrible quality data, it's, it's not
going to give you the best results.

441
00:26:08.469 --> 00:26:10.489
<v Speaker 2>So that's definitely something to keep in
mind.

442
00:26:16.329 --> 00:26:17.410
<v Speaker 2>OK, operate.

443
00:26:17.890 --> 00:26:21.930
<v Speaker 2>So operate really just becomes sort of the
steady-state, regular use of the app.

444
00:26:22.010 --> 00:26:26.250
<v Speaker 2>And again, we might be an operate for some
assets and design or deploy in other

445
00:26:26.250 --> 00:26:26.969
<v Speaker 2>assets.

446
00:26:28.000 --> 00:26:33.599
<v Speaker 2>Um, so the important part in operate is
we're monitoring cases.

447
00:26:33.890 --> 00:26:39.479
<v Speaker 2>And what someone like me would be trying
to glean from the user is, what is

448
00:26:39.479 --> 00:26:39.849
<v Speaker 2>interesting to you?

449
00:26:39.930 --> 00:26:41.260
<v Speaker 2>What do you want to see?

450
00:26:41.650 --> 00:26:45.160
<v Speaker 2>If this case is interesting to you, tell
us why.

451
00:26:45.250 --> 00:26:49.160
<v Speaker 2>We can help to make sure, you know, we can
tune the algorithms, we can change

452
00:26:49.160 --> 00:26:52.510
<v Speaker 2>configuration to make sure you keep seeing
more of these and less of other things.

453
00:26:53.250 --> 00:26:57.459
<v Speaker 2>Um, if the, if the results you're getting
are not interesting to you, we need to

454
00:26:57.459 --> 00:26:58.459
<v Speaker 2>know exactly why.

455
00:26:58.540 --> 00:27:01.329
<v Speaker 2>Are you not interested when vibration is
decreasing?

456
00:27:01.780 --> 00:27:02.540
<v Speaker 2>Probably not.

457
00:27:02.979 --> 00:27:07.180
<v Speaker 2>Um, but tell us that, and we can, we can
help give feedback into the app to make

458
00:27:07.180 --> 00:27:09.939
<v Speaker 2>sure that it doesn't tell you about things
that you're not interested in.

459
00:27:10.839 --> 00:27:14.719
<v Speaker 2>That sounds really straightforward, but
sometimes if we don't have the right people

460
00:27:14.719 --> 00:27:20.680
<v Speaker 2>in the room, which is one of our common
roadblocks and operate phase, They don't

461
00:27:20.680 --> 00:27:22.400
<v Speaker 2>know what's interesting to them.

462
00:27:22.959 --> 00:27:26.160
<v Speaker 2>They might look at a case and say, well,
yeah, that data changed, but I don't know

463
00:27:26.160 --> 00:27:26.770
<v Speaker 2>what it means.

464
00:27:26.880 --> 00:27:28.699
<v Speaker 2>So I'm just going to close it with, you
know.

465
00:27:29.969 --> 00:27:32.189
<v Speaker 2>Really not like, not helpful feedback.

466
00:27:32.770 --> 00:27:39.420
<v Speaker 2>Um, that, I would say is Jeez, these are
all three kind of lumped into the same

467
00:27:39.420 --> 00:27:39.849
<v Speaker 2>thing.

468
00:27:40.180 --> 00:27:46.069
<v Speaker 2>This is one of the biggest factors to a
project being successful or not.

469
00:27:46.900 --> 00:27:49.209
<v Speaker 2>Um, the app can't use itself.

470
00:27:49.300 --> 00:27:52.810
<v Speaker 2>Like I said in the beginning, predictive
maintenance is a strategy.

471
00:27:53.140 --> 00:27:54.290
<v Speaker 2>SenseEye is a tool.

472
00:27:54.579 --> 00:27:55.839
<v Speaker 2>Somebody has to use it.

473
00:27:56.540 --> 00:28:01.339
<v Speaker 2>So if, if we don't have the right people
in the room to kind of go into these cases

474
00:28:01.339 --> 00:28:04.819
<v Speaker 2>and figure out what's happening, and what
does the data mean for the state of the

475
00:28:04.819 --> 00:28:05.160
<v Speaker 2>machine.

476
00:28:05.420 --> 00:28:09.770
<v Speaker 2>And if they don't even have the poll to
say, to tell somebody, hey, you go take a

477
00:28:09.770 --> 00:28:12.680
<v Speaker 2>reading on that machine, hey, you go
replace this on that other machine.

478
00:28:13.060 --> 00:28:18.829
<v Speaker 2>If they can't, if they don't have that
sort of hierarchy in place, There's not so

479
00:28:18.829 --> 00:28:20.400
<v Speaker 2>much value that they can get.

480
00:28:21.420 --> 00:28:26.260
<v Speaker 2>So we want to make sure we have the right
people in the room who understand what is

481
00:28:26.260 --> 00:28:29.339
<v Speaker 2>interesting or not to them in the app,
because if they don't know, then they're not

482
00:28:29.339 --> 00:28:30.089
<v Speaker 2>going to get value.

483
00:28:30.140 --> 00:28:34.180
<v Speaker 2>And if they don't get value, their, their
buy-in and their involvement is just going

484
00:28:34.180 --> 00:28:35.819
<v Speaker 2>to fizzle right out.

485
00:28:36.339 --> 00:28:40.979
<v Speaker 2>And it's not reflective on the tool, but
It, it, it can seem that way.

486
00:28:41.260 --> 00:28:44.979
<v Speaker 2>So that's something that we try to
establish early on is make sure we get the right

487
00:28:44.979 --> 00:28:48.500
<v Speaker 2>people, make sure they're, you know, we
get them up and, up and educated with Sensi,

488
00:28:48.540 --> 00:28:49.839
<v Speaker 2>how it works, what they want to see.

489
00:28:50.260 --> 00:28:53.739
<v Speaker 2>And again, starting with simple machines
affects this.

490
00:28:54.569 --> 00:28:57.930
<v Speaker 2>Uh, if you start with the most complex
machine, they might say, yeah, it looks

491
00:28:57.930 --> 00:28:59.589
<v Speaker 2>interesting, but I don't know what the
heck it means.

492
00:28:59.930 --> 00:29:01.189
<v Speaker 2>We've seen that before.

493
00:29:01.959 --> 00:29:04.579
<v Speaker 2>Um, so those are definitely things to keep
in mind.

494
00:29:09.140 --> 00:29:10.400
<v Speaker 2>OK, refined phase.

495
00:29:10.770 --> 00:29:15.180
<v Speaker 2>At this point in a deployment, this is
usually when a customer success manager would

496
00:29:15.180 --> 00:29:19.560
<v Speaker 2>come in and kind of replace my role or
someone like me in the delivery team.

497
00:29:20.540 --> 00:29:23.239
<v Speaker 2>This is one we might go into each asset
type.

498
00:29:24.020 --> 00:29:25.780
<v Speaker 2>Not asset, but asset type.

499
00:29:25.859 --> 00:29:26.930
<v Speaker 2>So look at all the motors.

500
00:29:27.060 --> 00:29:27.969
<v Speaker 2>How are they performing?

501
00:29:28.219 --> 00:29:29.859
<v Speaker 2>How is the case feedback for motors?

502
00:29:29.930 --> 00:29:32.599
<v Speaker 2>Is it, is it good or is it kind of on the
lower side?

503
00:29:33.060 --> 00:29:37.140
<v Speaker 2>If it's not very good, we're going to go
back and revisit those in those kind of

504
00:29:37.140 --> 00:29:39.589
<v Speaker 2>design phase steps and figure out how can
we improve it.

505
00:29:39.689 --> 00:29:41.119
<v Speaker 2>Do they need more sensing?

506
00:29:41.500 --> 00:29:45.060
<v Speaker 2>If they need more sensing, then we'll step
into to deploy for that asset and get

507
00:29:45.060 --> 00:29:50.229
<v Speaker 2>more sensors involved and The whole thing
we're trying to do in refine is just

508
00:29:50.229 --> 00:29:51.089
<v Speaker 2>improve results.

509
00:29:51.630 --> 00:29:52.869
<v Speaker 2>Improvements can always be made.

510
00:29:53.109 --> 00:29:57.229
<v Speaker 2>Um, it might involve scaling the
deployment, going to new sites, getting new assets

511
00:29:57.229 --> 00:29:57.689
<v Speaker 2>in.

512
00:29:58.640 --> 00:30:02.260
<v Speaker 2>Um, we also capture success cases and
document them.

513
00:30:03.310 --> 00:30:08.479
<v Speaker 2>Uh, we may develop additional integrations
such as integrating with, uh, uh, you

514
00:30:08.479 --> 00:30:12.260
<v Speaker 2>know, a CMMS system, you know, a
maintenance management system.

515
00:30:13.479 --> 00:30:16.810
<v Speaker 2>Um, something like that, that would also
help the results you see in Sensai.

516
00:30:18.130 --> 00:30:22.969
<v Speaker 2>Involved here, everybody, depending on
what you want to do, right?

517
00:30:23.130 --> 00:30:26.089
<v Speaker 2>So it's kind of hard to say who exactly
needs to be involved, but there should be a

518
00:30:26.089 --> 00:30:29.650
<v Speaker 2>steering committee here to determine what
do we do next.

519
00:30:30.359 --> 00:30:34.479
<v Speaker 2>Um, and as needed, you can get your, your
machine experts, your end users and your

520
00:30:34.479 --> 00:30:35.500
<v Speaker 2>champion involved.

521
00:30:37.180 --> 00:30:41.599
<v Speaker 2>The common roadblocks that we see, this
isn't just for refine.

522
00:30:41.859 --> 00:30:47.660
<v Speaker 2>Um, it can be throughout, you know, the
full deployment, is a lack of high-level

523
00:30:47.660 --> 00:30:48.119
<v Speaker 2>support.

524
00:30:49.229 --> 00:30:53.150
<v Speaker 2>Really, for the most successful projects
we've seen, there is a desire from the very

525
00:30:53.150 --> 00:30:57.189
<v Speaker 2>bottom, from the maintenance, from the end
users, they want Sensei because they want

526
00:30:57.189 --> 00:31:01.430
<v Speaker 2>to become more efficient, because they're
struggling to, to make all, you know, meet

527
00:31:01.430 --> 00:31:03.089
<v Speaker 2>all the demands that they have in their
day.

528
00:31:03.869 --> 00:31:08.550
<v Speaker 2>Um, but we also need to see high-level
support from the top down, making sure Sensai

529
00:31:08.550 --> 00:31:12.989
<v Speaker 2>is an integrated part of the predictive
maintenance strategy, and not just another

530
00:31:12.989 --> 00:31:17.099
<v Speaker 2>tool, just like, oh, hey, go, go use this,
um, but we're not going to tell you how

531
00:31:17.099 --> 00:31:18.650
<v Speaker 2>your workflow should change, right?

532
00:31:19.760 --> 00:31:24.329
<v Speaker 2>Uh, so we, we need it from the top and the
bottom, and those typically lead to our

533
00:31:24.329 --> 00:31:25.670
<v Speaker 2>most successful deployments.

534
00:31:28.290 --> 00:31:28.650
<v Liz McGinn>Perfect.

535
00:31:28.770 --> 00:31:32.890
<v Liz McGinn>As you can see, we, uh, do offer a lot of
support throughout the whole process,

536
00:31:32.969 --> 00:31:38.130
<v Liz McGinn>which is one of the, uh, great things
about choosing us as a partner, um.

537
00:31:38.920 --> 00:31:44.770
<v Liz McGinn>As you can see here, uh, you can keep
learning um by scanning this QR code or um

538
00:31:44.770 --> 00:31:50.910
<v Liz McGinn>contact us directly, um, or figure out
your ROI potential with our ROI calculator,

539
00:31:51.489 --> 00:31:52.150
<v Liz McGinn>um.

540
00:31:53.329 --> 00:31:58.290
<v Liz McGinn>And yeah, so now we're gonna go into, uh,
I think we have a little bit of time left

541
00:31:58.290 --> 00:31:59.579
<v Liz McGinn>to go into some questions.

542
00:31:59.969 --> 00:32:02.260
<v Liz McGinn>I know that we had um some from the
audience here.

543
00:32:03.670 --> 00:32:10.239
<v Liz McGinn>Um So, one of the questions that came
through is,

544
00:32:10.599 --> 00:32:14.989
<v Liz McGinn>um, they would like to know how effective
is Sensi's diagnosis.

545
00:32:15.400 --> 00:32:17.939
<v Liz McGinn>Do we have any individual metrics for each
asset?

546
00:32:18.400 --> 00:32:20.449
<v Liz McGinn>And how can we define a level of
assertiveness?

547
00:32:21.660 --> 00:32:24.250
<v Liz McGinn>Um, in the relation to the assets.

548
00:32:25.030 --> 00:32:25.390
<v Speaker 2>OK.

549
00:32:25.469 --> 00:32:26.469
<v Speaker 2>That's a great question.

550
00:32:26.589 --> 00:32:29.109
<v Speaker 2>Um, my answer is gonna be really
ambiguous.

551
00:32:29.229 --> 00:32:31.290
<v Speaker 2>It, it absolutely depends.

552
00:32:31.869 --> 00:32:36.390
<v Speaker 2>Um, the better quality of condition
indicators that you send to Sensy, the better

553
00:32:36.390 --> 00:32:38.300
<v Speaker 2>quality of the diagnosis you're gonna get.

554
00:32:38.750 --> 00:32:43.060
<v Speaker 2>We do have, which I didn't really cover in
here, we have this awesome Sensy copilot,

555
00:32:43.150 --> 00:32:48.709
<v Speaker 2>which, um, as, as patterns are being
developed and, and captured on different data

556
00:32:48.709 --> 00:32:53.189
<v Speaker 2>on your assets, it's going to look at not
just that asset, but all other assets of

557
00:32:53.189 --> 00:32:54.069
<v Speaker 2>the same type.

558
00:32:54.709 --> 00:32:57.270
<v Speaker 2>And see, hey, have we seen similar cases
before?

559
00:32:57.339 --> 00:32:58.469
<v Speaker 2>Did they have work events on them?

560
00:32:58.550 --> 00:33:00.550
<v Speaker 2>Did somebody put a note in and explain
what happened?

561
00:33:00.630 --> 00:33:02.109
<v Speaker 2>What case feedback was given for those?

562
00:33:02.150 --> 00:33:05.189
<v Speaker 2>It's going to look at all that and give
you kind of a summary with possible

563
00:33:05.189 --> 00:33:07.250
<v Speaker 2>explanations of what it thinks might be
going on.

564
00:33:08.140 --> 00:33:13.219
<v Speaker 2>So, the quality of that possible
explanation is completely dependent on the quality

565
00:33:13.219 --> 00:33:15.479
<v Speaker 2>of feedback you're giving to the app.

566
00:33:15.939 --> 00:33:22.099
<v Speaker 2>Um, I will say we have customers who at
this point, don't do any maintenance on the

567
00:33:22.099 --> 00:33:25.130
<v Speaker 2>machines that Sensai is monitoring unless
Sensai tells them to.

568
00:33:25.459 --> 00:33:27.640
<v Speaker 2>So that's the sort of level of success
that you can achieve.

569
00:33:28.290 --> 00:33:30.579
<v Speaker 2>Um, there's also a creativity aspect
involved.

570
00:33:30.619 --> 00:33:34.939
<v Speaker 2>I have some customers that are so clever
in thinking outside the box because they

571
00:33:34.939 --> 00:33:36.099
<v Speaker 2>don't have a ton of data.

572
00:33:36.699 --> 00:33:40.339
<v Speaker 2>So they're thinking, how can we leverage
this process data that we have, and how can

573
00:33:40.339 --> 00:33:41.489
<v Speaker 2>we leverage this and that?

574
00:33:41.530 --> 00:33:43.729
<v Speaker 2>Um, they've had huge successes.

575
00:33:44.020 --> 00:33:47.380
<v Speaker 2>I mean, in the millions of dollars in
downtime saved.

576
00:33:48.040 --> 00:33:52.400
<v Speaker 2>Um, just by monitoring like the level in
a, I think they had an oil reservoir and

577
00:33:52.400 --> 00:33:53.719
<v Speaker 2>they're monitoring the level of it.

578
00:33:53.839 --> 00:33:57.839
<v Speaker 2>They had a threshold with the forecast and
the forecast said, hey, you have 2 weeks

579
00:33:57.839 --> 00:33:59.619
<v Speaker 2>until your oil drops below this level.

580
00:34:00.160 --> 00:34:04.979
<v Speaker 2>So they saved somebody around, you know,
they don't have to go by and collect

581
00:34:04.979 --> 00:34:07.640
<v Speaker 2>information anymore on that piece of
equipment.

582
00:34:08.199 --> 00:34:12.898
<v Speaker 2>So it It's also dependent on how you
define success.

583
00:34:13.089 --> 00:34:17.749
<v Speaker 2>Um, how do you, how do you define success
when it's in terms of efficiency?

584
00:34:18.089 --> 00:34:22.819
<v Speaker 2>Uh, what about safety if we're preventing
emergency kind of downtime events, how do

585
00:34:22.819 --> 00:34:24.569
<v Speaker 2>you, how do you quantify that?

586
00:34:24.888 --> 00:34:29.509
<v Speaker 2>Um, that's something that's tricky to do,
but we do work with our customers to

587
00:34:29.849 --> 00:34:34.269
<v Speaker 2>define those success criteria so that they
can see how well they're doing.

588
00:34:35.110 --> 00:34:36.830
<v Speaker 2>Hopefully, that kind of answers the
question.

589
00:34:38.120 --> 00:34:39.280
<v Liz McGinn>Awesome, we got another one.

590
00:34:39.439 --> 00:34:45.429
<v Liz McGinn>Can you explain how Sensai becomes part of
existing workflows with CMMS CMMS

591
00:34:45.429 --> 00:34:46.780
<v Liz McGinn>integrations and so forth?

592
00:34:47.989 --> 00:34:48.350
<v Speaker 2>Sure.

593
00:34:48.550 --> 00:34:52.729
<v Speaker 2>So it's, it's, it's really up to the
customer, but we, we can help.

594
00:34:52.889 --> 00:34:58.899
<v Speaker 2>So Suni does have capabilities, um, to
send information out.

595
00:34:59.149 --> 00:35:03.030
<v Speaker 2>So if a case is opened up and somebody
looks at it and says, yeah, I want a

596
00:35:03.030 --> 00:35:07.669
<v Speaker 2>maintenance event done on this machine, an
integration can be developed so that they

597
00:35:07.669 --> 00:35:12.179
<v Speaker 2>just, they tick a little button in the,
the case itself in the app, and it sends a

598
00:35:12.179 --> 00:35:16.030
<v Speaker 2>notification that will be picked up in the
CMMS to create a work event.

599
00:35:16.459 --> 00:35:20.189
<v Speaker 2>And then you can also go the other way,
such that when that work event is finished

600
00:35:20.189 --> 00:35:23.469
<v Speaker 2>and closed out, it sends it back into
Sunai as a little work event.

601
00:35:24.409 --> 00:35:27.120
<v Speaker 2>Um, which can then be read by co-pilot.

602
00:35:27.260 --> 00:35:27.699
<v Speaker 2>So.

603
00:35:28.510 --> 00:35:30.330
<v Speaker 2>That's one possible integration.

604
00:35:30.469 --> 00:35:35.830
<v Speaker 2>Um, the way that we usually see people
using the app day to day is they'll have like

605
00:35:35.830 --> 00:35:39.270
<v Speaker 2>a core group of people, maybe they're
reliability engineers, maybe they're

606
00:35:39.270 --> 00:35:41.810
<v Speaker 2>maintainers, and they start each morning.

607
00:35:42.770 --> 00:35:47.330
<v Speaker 2>Uh, I don't know, at a roundtable sort of
situation, 2030 minutes, just to make a

608
00:35:47.330 --> 00:35:48.310
<v Speaker 2>plan for the day.

609
00:35:48.810 --> 00:35:52.729
<v Speaker 2>They'll check Senseii, see what Sensai is
saying they should be paying attention to.

610
00:35:52.850 --> 00:35:56.689
<v Speaker 2>That's the whole point of Senseii is that
you don't have to go looking through the

611
00:35:56.689 --> 00:35:59.280
<v Speaker 2>data of all your thousands of machines all
the time.

612
00:35:59.719 --> 00:36:02.610
<v Speaker 2>Sensei is going to tell you what's
changing and what you should really be paying

613
00:36:02.610 --> 00:36:04.889
<v Speaker 2>attention to, so you can prioritize your
actions.

614
00:36:05.550 --> 00:36:09.750
<v Speaker 2>Um, so they'll use Sensi in that way to
kind of prioritize their work for the day or

615
00:36:09.750 --> 00:36:10.370
<v Speaker 2>the week.

616
00:36:10.600 --> 00:36:15.379
<v Speaker 2>Um, they might use it to even Influence
what operations are doing.

617
00:36:16.409 --> 00:36:20.300
<v Speaker 2>If they notice Sensai is telling them that
they're getting really low flow through a

618
00:36:20.300 --> 00:36:24.219
<v Speaker 2>pump and it's gonna cause the pump to
cavitate, they might go back and say, hey,

619
00:36:24.659 --> 00:36:29.310
<v Speaker 2>operations, increase the flow through that
pump so our pump doesn't, you know, die.

620
00:36:29.689 --> 00:36:32.679
<v Speaker 2>Uh, that's another way that they might use
it.

621
00:36:32.909 --> 00:36:37.540
<v Speaker 2>Um, the other thing too is depending on
how crafty you are with your integration or

622
00:36:37.540 --> 00:36:39.399
<v Speaker 2>who you partner with to, to build these.

623
00:36:40.300 --> 00:36:45.620
<v Speaker 2>You can choose to pretty much anything you
see in the app, as far as like we have

624
00:36:45.620 --> 00:36:48.020
<v Speaker 2>KPIs in there that you can document and
track.

625
00:36:48.100 --> 00:36:50.850
<v Speaker 2>Um, it can be pulled out via graphQL.

626
00:36:50.979 --> 00:36:54.239
<v Speaker 2>So if you want to build another dashboard
somewhere that's taking stuff from Scenti,

627
00:36:54.300 --> 00:36:58.020
<v Speaker 2>that's taking stuff from some, I don't
know, oil analysis program you might have.

628
00:36:59.110 --> 00:37:01.850
<v Speaker 2>You can take all that and collate it
together into one little dashboard.

629
00:37:01.919 --> 00:37:03.459
<v Speaker 2>We have folks that do that as well.

630
00:37:05.590 --> 00:37:06.340
<v Liz McGinn>Awesome.

631
00:37:06.800 --> 00:37:07.909
<v Liz McGinn>We've got another question.

632
00:37:08.040 --> 00:37:11.239
<v Liz McGinn>Uh, Sensei seems like a useful tool for
in-house machinery.

633
00:37:11.520 --> 00:37:14.879
<v Liz McGinn>Would it work well for an OEM that wants
to monitor machines that have been

634
00:37:14.879 --> 00:37:19.189
<v Liz McGinn>installed at customer sites, or would the
customer of the OEM need to deploy Sensai

635
00:37:19.189 --> 00:37:20.780
<v Liz McGinn>directly to do their own monitoring?

636
00:37:22.060 --> 00:37:23.100
<v Speaker 2>That's a really good question.

637
00:37:23.219 --> 00:37:25.340
<v Speaker 2>That might be a better question for our
sales team.

638
00:37:25.459 --> 00:37:27.739
<v Speaker 2>I know we've explored that option before.

639
00:37:28.139 --> 00:37:32.860
<v Speaker 2>Um, but as, as part of delivery, I
personally haven't done that type of project, not

640
00:37:32.860 --> 00:37:33.879
<v Speaker 2>saying it's not possible.

641
00:37:34.620 --> 00:37:37.659
<v Speaker 2>Um, I think depending on.

642
00:37:38.780 --> 00:37:42.899
<v Speaker 2>Um, the sensing on the machine, you might
just need to install like a, a gateway or

643
00:37:42.899 --> 00:37:44.189
<v Speaker 2>something to send it straight to Sensi.

644
00:37:44.820 --> 00:37:50.689
<v Speaker 2>Uh, usually we pull data or data is pushed
to us from a customer's historian, but

645
00:37:50.689 --> 00:37:52.020
<v Speaker 2>that's not the only way to get data.

646
00:37:52.139 --> 00:37:56.889
<v Speaker 2>So I think that would be the only
constraint to figure out, and I'm certain that

647
00:37:56.889 --> 00:37:58.189
<v Speaker 2>there's a solution for that.

648
00:37:58.409 --> 00:38:02.699
<v Speaker 2>Um, I would say, uh, see if we can get in
touch with the sales team to figure out

649
00:38:02.699 --> 00:38:03.620
<v Speaker 2>the details of that one.

650
00:38:05.370 --> 00:38:05.689
<v Liz McGinn>Perfect.

651
00:38:05.810 --> 00:38:08.489
<v Liz McGinn>We got time for, I think, maybe one more
question.

652
00:38:08.929 --> 00:38:13.689
<v Liz McGinn>Um, new technologies are often introduced
with a top-down approach where management

653
00:38:13.689 --> 00:38:15.439
<v Liz McGinn>tells employees to use them.

654
00:38:15.729 --> 00:38:18.610
<v Liz McGinn>This can lead to employees setting up new
tech and then forgetting about it.

655
00:38:19.050 --> 00:38:23.570
<v Liz McGinn>What are some best practices to get end
users buy-in and actively use the new

656
00:38:23.570 --> 00:38:23.919
<v Liz McGinn>technology?

657
00:38:24.409 --> 00:38:26.520
<v Speaker 2>Yeah, we see that problem a lot.

658
00:38:26.600 --> 00:38:28.530
<v Speaker 2>Um, part of design phase.

659
00:38:29.300 --> 00:38:33.570
<v Speaker 2>Includes working with the people at the
top, um, might be a steering committee,

660
00:38:33.699 --> 00:38:38.580
<v Speaker 2>might be a champion, whoever is, whoever
is implementing Sensei, we have this

661
00:38:38.580 --> 00:38:44.250
<v Speaker 2>presentation that we fill out that says,
really, it's meant to facilitate a

662
00:38:44.250 --> 00:38:48.020
<v Speaker 2>conversation between those higher-ups and
the people that are having to now use the

663
00:38:48.020 --> 00:38:52.899
<v Speaker 2>software, um, so that they can be on the
same page and we just want to make sure

664
00:38:52.899 --> 00:38:54.639
<v Speaker 2>that the people, the end users are heard.

665
00:38:55.419 --> 00:38:58.649
<v Speaker 2>Because we know that there's a shortage in
personnel.

666
00:38:58.820 --> 00:39:02.560
<v Speaker 2>We know that they're being made to do like
more than they have time to do.

667
00:39:03.060 --> 00:39:06.620
<v Speaker 2>Um, and learning another tool is not on
their list of things that they want to do

668
00:39:06.620 --> 00:39:07.060
<v Speaker 2>right now.

669
00:39:07.209 --> 00:39:11.620
<v Speaker 2>But the whole point of Sensei is to help
them with that workload, help make them

670
00:39:11.620 --> 00:39:12.260
<v Speaker 2>more efficient.

671
00:39:12.379 --> 00:39:15.399
<v Speaker 2>So we have a whole presentation that we do
to try and facilitate that.

672
00:39:16.260 --> 00:39:21.899
<v Speaker 2>Um, our case review calls that are weekly
in the beginning, at least, also, I think

673
00:39:21.899 --> 00:39:25.159
<v Speaker 2>help because it is the end users working
directly with us.

674
00:39:25.419 --> 00:39:30.620
<v Speaker 2>And we, we do try to help them see the
value in the app and we do listen to them so

675
00:39:30.620 --> 00:39:33.459
<v Speaker 2>that we can understand from them what
would be valuable.

676
00:39:33.620 --> 00:39:39.050
<v Speaker 2>So, We do try, but I think in order for it
to be the most successful.

677
00:39:40.010 --> 00:39:43.209
<v Speaker 2>People just need to talk to each other and
listen to each other.

678
00:39:43.530 --> 00:39:46.570
<v Speaker 2>Um, when things come just from high up,
and you're going to use this now, and then

679
00:39:46.570 --> 00:39:48.709
<v Speaker 2>they don't help them actually integrate
it.

680
00:39:49.590 --> 00:39:50.550
<v Speaker 2>That can be problematic.

681
00:39:50.620 --> 00:39:51.669
<v Speaker 2>We try to help with that.

682
00:39:53.330 --> 00:39:54.790
<v Speaker 2>It is, that is a tricky part.

683
00:39:56.520 --> 00:39:58.239
<v Liz McGinn>I agree, um.

684
00:39:59.260 --> 00:40:02.699
<v Liz McGinn>So, anyway, thank you all for your time
today for joining us.

685
00:40:02.860 --> 00:40:08.040
<v Liz McGinn>Um, I hope you enjoyed learning about our
process and how we help you through, um,

686
00:40:08.060 --> 00:40:10.320
<v Liz McGinn>rather than attending new technology and
running away.

687
00:40:10.570 --> 00:40:15.040
<v Liz McGinn>Um, if you want, uh, someone to contact
you, please provide your contact

688
00:40:15.040 --> 00:40:15.139
<v Liz McGinn>information.

689
00:40:15.300 --> 00:40:17.379
<v Liz McGinn>Someone will reach out, um, as soon as
possible.

690
00:40:17.449 --> 00:40:22.889
<v Liz McGinn>And if there's any other questions, um,
we'll be able to, uh, Answer afterwards in a

691
00:40:22.889 --> 00:40:23.629
<v Liz McGinn>follow-up email.

692
00:40:26.989 --> 00:40:28.229
<v Speaker 2>All right, thanks everyone.

693
00:40:28.429 --> 00:40:28.870
<v Speaker 2>Thanks.

694
00:40:29.949 --> 00:40:30.429
<v Liz McGinn>Thank you.

695
00:40:30.590 --> 00:40:32.770
<v Liz McGinn>Look out for uh upcoming webinars as well.
