WEBVTT
Language: en-US

1
00:00:10.010 --> 00:00:15.520
<v Isabelle Fritzen>Buildings generate an enormous amount of
data every day from HVAC systems and

2
00:00:15.520 --> 00:00:21.850
<v Isabelle Fritzen>lighting to occupancy sensors and energy
meters, but before any analytics or AI can

3
00:00:21.850 --> 00:00:25.950
<v Isabelle Fritzen>create value, there is a less visible step
that must happen first.

4
00:00:26.389 --> 00:00:29.309
<v Isabelle Fritzen>Adding all the data into a usable format.

5
00:00:29.549 --> 00:00:34.150
<v Isabelle Fritzen>In this episode of Buildings of Tomorrow,
we are going to unpack why onboarding

6
00:00:34.150 --> 00:00:39.840
<v Isabelle Fritzen>building data has traditionally been one
of the most complex parts of digitalizing

7
00:00:39.840 --> 00:00:45.590
<v Isabelle Fritzen>buildings and how AI is starting to change
that, and, more importantly, how this is

8
00:00:45.590 --> 00:00:51.049
<v Isabelle Fritzen>paving the way to the next phase of
building operations, self-learning systems.

9
00:00:51.419 --> 00:00:54.819
<v Isabelle Fritzen>Hello, I'm Isabelle Fritzen, and I will be
your host.

10
00:00:55.229 --> 00:00:59.740
<v Isabelle Fritzen>Joining me today is Paul Baumann from
Siemens Smart-Infrastructure Buildings.

11
00:01:00.110 --> 00:01:01.750
<v Isabelle Fritzen>Thank you, Paul, for joining us.

12
00:01:01.950 --> 00:01:03.909
<v Isabelle Fritzen>And let's start with the basics.

13
00:01:04.430 --> 00:01:09.349
<v Isabelle Fritzen>Before we can analyze building data or
generate insights, we first need to onboard

14
00:01:09.349 --> 00:01:09.809
<v Isabelle Fritzen>it.

15
00:01:10.110 --> 00:01:14.330
<v Isabelle Fritzen>In reality, how easy or difficult is that
process?

16
00:01:15.470 --> 00:01:16.330
<v Paul Baumann>Thank you for having me.

17
00:01:16.589 --> 00:01:20.150
<v Paul Baumann>Yes, in reality, it's often the hardest
part to start the journey.

18
00:01:20.970 --> 00:01:26.540
<v Paul Baumann>For instance, most buildings today have
dozens of systems installed like heating,

19
00:01:26.610 --> 00:01:32.139
<v Paul Baumann>ventilation, cooling, lighting, meters,
access control, all generating data.

20
00:01:33.019 --> 00:01:37.610
<v Paul Baumann>But the problem is that every system
speaks a slightly different language.

21
00:01:38.400 --> 00:01:43.589
<v Paul Baumann>The naming conventions are different, the
structure are different, and sometimes the

22
00:01:43.589 --> 00:01:46.599
<v Paul Baumann>commutation is incomplete or even missing.

23
00:01:47.550 --> 00:01:54.150
<v Paul Baumann>So before you can run analytics or AI,
someone needs to make sense out of all this

24
00:01:54.150 --> 00:01:54.589
<v Paul Baumann>data.

25
00:01:55.470 --> 00:02:01.430
<v Paul Baumann>Traditionally, that means engineers
manually going through thousands of points just

26
00:02:01.430 --> 00:02:08.429
<v Paul Baumann>to understand what each M represents to
ensure that later AI can also understand

27
00:02:08.429 --> 00:02:08.949
<v Paul Baumann>the data.

28
00:02:09.740 --> 00:02:14.860
<v Paul Baumann>That's why onboarding can take weeks or
even months in large buildings,

29
00:02:16.039 --> 00:02:21.679
<v Isabelle Fritzen>and part of the reason for that complexity
is that buildings rarely have just one

30
00:02:21.679 --> 00:02:22.229
<v Isabelle Fritzen>system, right?

31
00:02:22.720 --> 00:02:28.229
<v Isabelle Fritzen>You might have HVAC from one vendor,
lighting from another, and different

32
00:02:28.229 --> 00:02:31.699
<v Isabelle Fritzen>generations of technology layered on top
of each other.

33
00:02:32.000 --> 00:02:37.419
<v Isabelle Fritzen>But what typically happens when
organizations try to connect all of that together.

34
00:02:38.270 --> 00:02:42.330
<v Paul Baumann>So yes, what usually happens is attempt to
overcome this fragmentation.

35
00:02:43.250 --> 00:02:48.649
<v Paul Baumann>Each system produces valuable data, but it
exists in its own structure and

36
00:02:48.649 --> 00:02:54.610
<v Paul Baumann>interface, so operators often end up
juggling several dashboards just to understand

37
00:02:54.610 --> 00:03:00.100
<v Paul Baumann>what's happening in the building instead
of undergoing costly integration of all

38
00:03:00.100 --> 00:03:00.649
<v Paul Baumann>systems.

39
00:03:01.690 --> 00:03:06.580
<v Paul Baumann>That remonstration is exactly why data on
boarding and normalization are so

40
00:03:06.580 --> 00:03:07.000
<v Paul Baumann>important.

41
00:03:07.880 --> 00:03:14.039
<v Paul Baumann>And platforms like BillingX create a
common data layer where information from

42
00:03:14.039 --> 00:03:15.240
<v Paul Baumann>different systems.

43
00:03:16.160 --> 00:03:22.229
<v Paul Baumann>Whether they are backnet, mothballs, Knes,
or whatever proprietary protocols can be

44
00:03:22.229 --> 00:03:22.979
<v Paul Baumann>sanitized.

45
00:03:24.059 --> 00:03:30.690
<v Paul Baumann>Once the data is standardized and
normalized, operators can move from managing

46
00:03:30.690 --> 00:03:35.619
<v Paul Baumann>individual systems to understanding the
building as one connected environment.

47
00:03:36.500 --> 00:03:42.960
<v Isabelle Fritzen>OK, Paul, but in other words, before a
building can become smart, someone first has

48
00:03:42.960 --> 00:03:46.759
<v Isabelle Fritzen>to translate all the raw information
coming from the systems.

49
00:03:47.169 --> 00:03:52.759
<v Isabelle Fritzen>When you actually connect a building
system for the first time, what does that raw

50
00:03:52.759 --> 00:03:54.839
<v Isabelle Fritzen>data typically look like?

51
00:03:55.279 --> 00:03:57.070
<v Paul Baumann>It's usually not very intuitive.

52
00:03:57.399 --> 00:04:03.860
<v Paul Baumann>You might see data points named like 1103.

53
00:04:04.360 --> 00:04:10.990
<v Paul Baumann>Or AI 147 for bid management system that

54
00:04:10.990 --> 00:04:13.949
<v Paul Baumann>originally created them, those names make
sense.

55
00:04:15.419 --> 00:04:20.820
<v Paul Baumann>Furthermore, often these names are also
written in native language of the one who

56
00:04:20.820 --> 00:04:21.500
<v Paul Baumann>created them.

57
00:04:22.250 --> 00:04:27.549
<v Paul Baumann>But for a platform trying to analyze
billing performance, they don't intermediately

58
00:04:27.549 --> 00:04:29.519
<v Paul Baumann>tell you what they represent.

59
00:04:30.470 --> 00:04:37.029
<v Paul Baumann>Is it a chilled water sensor, a supply air
temperature, which piece of equipment?

60
00:04:37.269 --> 00:04:38.369
<v Paul Baumann>Where is it located?

61
00:04:39.109 --> 00:04:45.230
<v Paul Baumann>So the first task is translating
annotating those technical names into something

62
00:04:45.230 --> 00:04:51.079
<v Paul Baumann>meaningful like chilled water supply
temperature sensor or it's a room temperature

63
00:04:51.079 --> 00:04:52.269
<v Paul Baumann>of zone 3.

64
00:04:52.709 --> 00:04:54.609
<v Paul Baumann>Is what needs to be done.

65
00:04:55.269 --> 00:04:59.320
<v Paul Baumann>That translation and annotation step is
really what onboarding is about.

66
00:04:59.670 --> 00:05:01.269
<v Isabelle Fritzen>So that was very helpful.

67
00:05:01.549 --> 00:05:06.839
<v Isabelle Fritzen>And if you had to explain onboarding to
someone outside the technical world like

68
00:05:06.839 --> 00:05:09.130
<v Isabelle Fritzen>myself, how would you describe it?

69
00:05:09.709 --> 00:05:16.230
<v Paul Baumann>Well, I usually compare it to organizing,
let's say, a books library before anyone

70
00:05:16.230 --> 00:05:18.410
<v Paul Baumann>can actually read or search for a book.

71
00:05:18.820 --> 00:05:23.489
<v Paul Baumann>You first need to label them, categorize
them, and place them in the right section.

72
00:05:24.720 --> 00:05:26.709
<v Paul Baumann>Building data is very similar.

73
00:05:27.250 --> 00:05:33.690
<v Paul Baumann>Every sensor, a meter, a piece of
equipment produces signals, but those signals need

74
00:05:33.690 --> 00:05:39.149
<v Paul Baumann>to be labeled and organized so that
software applications can understand them.

75
00:05:39.329 --> 00:05:45.549
<v Paul Baumann>What platforms like Billing A do is help
automate that process with the help of AI.

76
00:05:45.929 --> 00:05:51.170
<v Paul Baumann>We collect and analyze the information
gathered from various connected systems.

77
00:05:52.269 --> 00:05:57.149
<v Paul Baumann>The information is gathered from technical
floor plan drawings, from underlying

78
00:05:57.149 --> 00:06:02.529
<v Paul Baumann>automation systems and engineer
configuration files of controllers, billing

79
00:06:02.529 --> 00:06:09.070
<v Paul Baumann>management stations, or even from
documentation of files available with a

80
00:06:09.070 --> 00:06:09.730
<v Paul Baumann>customer.

81
00:06:09.950 --> 00:06:16.350
<v Paul Baumann>This allows us to build a data structure
of the bill needed for further analytics

82
00:06:16.350 --> 00:06:17.149
<v Paul Baumann>and reporting.

83
00:06:18.700 --> 00:06:23.160
<v Paul Baumann>Using the building X data model, we create
a common understanding of the

84
00:06:23.160 --> 00:06:29.480
<v Paul Baumann>complexities in the underlying billing
data by adding meaning to it,

85
00:06:30.059 --> 00:06:35.399
<v Paul Baumann>defining the relationship within the
structure, and identifying the dependencies

86
00:06:36.329 --> 00:06:37.730
<v Paul Baumann>across disciplines and domains.

87
00:06:38.739 --> 00:06:40.179
<v Isabelle Fritzen>That's a very good example.

88
00:06:40.369 --> 00:06:44.559
<v Isabelle Fritzen>So where does most of the time typically
go during onboarding?

89
00:06:45.190 --> 00:06:49.140
<v Paul Baumann>Well, uh, the biggest time thing is
annotating validation.

90
00:06:49.500 --> 00:06:56.070
<v Paul Baumann>So, uh, engineers need to identify every
sensor, meter, valve, piece of equipment

91
00:06:56.070 --> 00:06:56.910
<v Paul Baumann>and the data stream.

92
00:06:57.760 --> 00:07:03.880
<v Paul Baumann>Then they verify that each data point is
mapped correctly to a piece of equipment to

93
00:07:03.880 --> 00:07:04.700
<v Paul Baumann>which it belongs.

94
00:07:05.399 --> 00:07:10.500
<v Paul Baumann>In a large commercial building, you might
be dealing with data points ranging from

95
00:07:10.619 --> 00:07:14.220
<v Paul Baumann>1000 to 100,000 data points.

96
00:07:15.059 --> 00:07:20.390
<v Paul Baumann>Of course, not all data points in the
underlying system running on site are

97
00:07:20.390 --> 00:07:22.369
<v Paul Baumann>necessarily for onboarding into the cloud.

98
00:07:22.989 --> 00:07:28.470
<v Paul Baumann>It becomes more important for the
onboarding engineer to exactly understand the data

99
00:07:28.470 --> 00:07:34.899
<v Paul Baumann>available on site and then select the data
points and assets which are useful for

100
00:07:35.350 --> 00:07:41.390
<v Paul Baumann>for analytics, insight generation, and
autonomous operation of a building via a

101
00:07:41.390 --> 00:07:43.809
<v Paul Baumann>cloud platform like Building X.

102
00:07:44.720 --> 00:07:50.700
<v Paul Baumann>So yeah, that's why historically
onboarding required a lot of mental effort.

103
00:07:51.200 --> 00:07:57.739
<v Paul Baumann>It was not unheard of that an onboarding
technician may take 2 to 3 months

104
00:07:57.959 --> 00:08:03.959
<v Paul Baumann>to complete an onboarding process for such
a building, with a lot of time spent in

105
00:08:03.959 --> 00:08:08.880
<v Paul Baumann>going back and forth between understanding
what is available in the building and

106
00:08:08.880 --> 00:08:12.119
<v Paul Baumann>what is needed to be on boarded to the
cloud.

107
00:08:13.279 --> 00:08:16.739
<v Isabelle Fritzen>Yes, and that brings us to AI, the
buzzword.

108
00:08:17.079 --> 00:08:22.799
<v Isabelle Fritzen>If onboarding has traditionally been such
a manual process, where does AI start to

109
00:08:22.799 --> 00:08:23.660
<v Isabelle Fritzen>change this?

110
00:08:24.109 --> 00:08:27.540
<v Paul Baumann>So AI helps by recognizing patterns in
data.

111
00:08:28.559 --> 00:08:35.140
<v Paul Baumann>Instead of a human manually identifying
every data point, AI models can look

112
00:08:35.140 --> 00:08:41.880
<v Paul Baumann>at how signals behave, compare them to
known equipment patterns, and suggest

113
00:08:41.880 --> 00:08:45.020
<v Paul Baumann>what those points likely represent.

114
00:08:46.409 --> 00:08:51.349
<v Paul Baumann>For instance, the system might detect a
group of signals that behave like a typical

115
00:08:51.809 --> 00:08:58.409
<v Paul Baumann>air handling unit, supply air temperature,
return air, fan speed, valve

116
00:08:58.409 --> 00:09:02.900
<v Paul Baumann>position, and automatically suggest that
equipment structure.

117
00:09:03.450 --> 00:09:09.429
<v Paul Baumann>That is where platforms like Billing X are
increasingly using AI system on boarding

118
00:09:10.260 --> 00:09:11.169
<v Paul Baumann>to access this process.

119
00:09:12.270 --> 00:09:17.690
<v Paul Baumann>We use a combination of techniques to
ensure accurate and reliable results.

120
00:09:18.479 --> 00:09:24.700
<v Paul Baumann>Natural language processing like NLP helps
us understand the meaning of each data

121
00:09:24.700 --> 00:09:25.840
<v Paul Baumann>point and description.

122
00:09:26.630 --> 00:09:32.469
<v Paul Baumann>While machine learning classification
models then map these data points to the right

123
00:09:32.469 --> 00:09:39.369
<v Paul Baumann>equipment systems and assign confidence
scores, rules based on billing X data model

124
00:09:39.989 --> 00:09:42.450
<v Paul Baumann>further improve classification accuracy.

125
00:09:43.309 --> 00:09:50.150
<v Paul Baumann>In more complex cases, again, large
language models, LLM helps resolve

126
00:09:50.150 --> 00:09:53.989
<v Paul Baumann>tiebreaks and increase coverage while
maintaining accuracy.

127
00:09:55.099 --> 00:10:00.770
<v Isabelle Fritzen>And we are also starting to hear about
things like Ask Building X, which introduce

128
00:10:01.020 --> 00:10:03.690
<v Isabelle Fritzen>generative AI into the building space.

129
00:10:04.059 --> 00:10:06.789
<v Isabelle Fritzen>How does that fit into this evolution?

130
00:10:07.619 --> 00:10:09.570
<v Paul Baumann>Well, it fits perfectly.

131
00:10:10.099 --> 00:10:14.780
<v Paul Baumann>Ask BuildingX is a good example of how to
generate AI makes building information

132
00:10:14.780 --> 00:10:16.539
<v Paul Baumann>much easier to access and use.

133
00:10:17.369 --> 00:10:23.049
<v Paul Baumann>In building operations, engineers and
facility teams often need to look up details

134
00:10:23.049 --> 00:10:28.059
<v Paul Baumann>about equipment, things like
specifications, installation requirements, or

135
00:10:28.059 --> 00:10:28.770
<v Paul Baumann>documentation.

136
00:10:29.989 --> 00:10:36.049
<v Paul Baumann>And instead of searching through manual or
product catalogs, they can simply ask

137
00:10:36.869 --> 00:10:38.489
<v Paul Baumann>questions in natural language.

138
00:10:38.909 --> 00:10:45.330
<v Paul Baumann>For instance, someone might ask, Can you
show me the dimension of my humidity sensor

139
00:10:45.549 --> 00:10:47.229
<v Paul Baumann>installed in a giving building?

140
00:10:48.030 --> 00:10:54.590
<v Paul Baumann>AI can then retrieve that information
instantly from relevant documents or system

141
00:10:54.590 --> 00:10:59.950
<v Paul Baumann>knowledge, so it acts like a digital
system for billing systems, helping operators

142
00:10:59.950 --> 00:11:04.590
<v Paul Baumann>quickly access the information they need
without digging through multiple sources.

143
00:11:06.070 --> 00:11:10.590
<v Paul Baumann>What is also important is how we support
the democratization of skills.

144
00:11:11.349 --> 00:11:17.669
<v Paul Baumann>A junior technician, for instance, can
onboard data or connect systems with more

145
00:11:17.669 --> 00:11:19.570
<v Paul Baumann>confidence using semantic enrichment.

146
00:11:20.750 --> 00:11:26.299
<v Paul Baumann>When paired with ask building acts, they
have the right knowledge at the fingertip

147
00:11:26.919 --> 00:11:28.080
<v Paul Baumann>based on the context.

148
00:11:29.130 --> 00:11:33.609
<v Paul Baumann>This helps close the knowledge gap we
often see in facility management today.

149
00:11:34.950 --> 00:11:40.950
<v Paul Baumann>And of course this becomes even more
powerful when the underlying data and devices

150
00:11:40.950 --> 00:11:44.900
<v Paul Baumann>are properly structured through platforms
such as Building X.

151
00:11:45.549 --> 00:11:50.549
<v Isabelle Fritzen>So once the data is structured and all
those systems are connected, that's when the

152
00:11:50.549 --> 00:11:52.479
<v Isabelle Fritzen>real value starts to appear.

153
00:11:52.909 --> 00:11:55.349
<v Isabelle Fritzen>What changes for operators at that point?

154
00:11:55.659 --> 00:11:58.969
<v Isabelle Fritzen>What becomes possible that wasn't possible
before?

155
00:11:59.500 --> 00:12:01.299
<v Paul Baumann>Well, a lot changes.

156
00:12:01.380 --> 00:12:02.679
<v Paul Baumann>A lot becomes possible.

157
00:12:03.140 --> 00:12:06.359
<v Paul Baumann>First, operators gain true operational
visibility.

158
00:12:06.859 --> 00:12:12.460
<v Paul Baumann>Instead of looking at isolated systems,
they can see how equipment across the

159
00:12:12.460 --> 00:12:15.119
<v Paul Baumann>building is performing in one place.

160
00:12:16.030 --> 00:12:20.390
<v Paul Baumann>Second, platforms can run advanced
analytics such as fault detection and

161
00:12:20.390 --> 00:12:21.229
<v Paul Baumann>diagnostics.

162
00:12:21.950 --> 00:12:28.109
<v Paul Baumann>This means the system can automatically
identify inefficiencies or equipment issues

163
00:12:28.109 --> 00:12:30.890
<v Paul Baumann>before they even become a major problem.

164
00:12:31.440 --> 00:12:37.119
<v Paul Baumann>And third, You can start optimizing
performance across a building or even across the

165
00:12:37.119 --> 00:12:37.900
<v Paul Baumann>entire portfolio.

166
00:12:38.760 --> 00:12:44.340
<v Paul Baumann>For instance, in building acts,
applications like Operations Manager can identify

167
00:12:44.880 --> 00:12:50.400
<v Paul Baumann>operational anomalies, while Energy
Manager can benchmark energy performance across

168
00:12:50.400 --> 00:12:51.320
<v Paul Baumann>multiple buildings.

169
00:12:52.039 --> 00:12:57.719
<v Paul Baumann>This is when buildings start moving from
reactive operations towards data-driven

170
00:12:57.719 --> 00:12:58.109
<v Paul Baumann>operations.

171
00:12:58.520 --> 00:13:03.359
<v Isabelle Fritzen>Yes, and this is also where we start
moving from simply smart buildings towards

172
00:13:03.359 --> 00:13:07.640
<v Isabelle Fritzen>something more advanced like self-learning
models that are foundational to

173
00:13:07.640 --> 00:13:09.070
<v Isabelle Fritzen>autonomous buildings.

174
00:13:09.520 --> 00:13:13.219
<v Isabelle Fritzen>How does the system actually get smarter
over time?

175
00:13:14.049 --> 00:13:18.000
<v Paul Baumann>Well, it happens gradually as the system
learns from operational data.

176
00:13:18.500 --> 00:13:24.520
<v Paul Baumann>Once building data is structured and
continuously collected, AI models can start

177
00:13:24.520 --> 00:13:30.059
<v Paul Baumann>recognizing patterns in how the building
behaves, things like occupancy pattern,

178
00:13:30.299 --> 00:13:34.700
<v Paul Baumann>where the impact, equipment performance,
or even energy consumption.

179
00:13:35.659 --> 00:13:40.359
<v Paul Baumann>Over time, the system can identify what
normal looks like.

180
00:13:40.710 --> 00:13:46.320
<v Paul Baumann>That means it can detect deviation
earlier, predict potential issues, and even

181
00:13:46.320 --> 00:13:47.909
<v Paul Baumann>recommend operational improvements.

182
00:13:48.960 --> 00:13:54.000
<v Paul Baumann>So instead of reacting to alarms,
operators can start managing buildings

183
00:13:54.000 --> 00:13:54.880
<v Paul Baumann>proactively.

184
00:13:56.539 --> 00:14:01.760
<v Paul Baumann>That's really the transition from a smart
building to an autonomous building.

185
00:14:02.500 --> 00:14:08.700
<v Paul Baumann>For instance, a software engineer office
hosts a quarterly planning meeting with a

186
00:14:08.700 --> 00:14:12.539
<v Paul Baumann>lot of stakeholders visiting the office
for 3 days every quarter.

187
00:14:13.450 --> 00:14:16.690
<v Paul Baumann>When the occupancy of a building is more
than doubled.

188
00:14:17.440 --> 00:14:23.400
<v Paul Baumann>The system will learn this behavior of a
building and ensure that the various

189
00:14:23.400 --> 00:14:30.200
<v Paul Baumann>systems of a building, such as comfort,
lighting, fire safety, security, energy,

190
00:14:30.760 --> 00:14:37.659
<v Paul Baumann>are prepared and ready for the spike in
occupancy and address the demands for the

191
00:14:37.900 --> 00:14:39.400
<v Paul Baumann>occupant accordingly.

192
00:14:40.340 --> 00:14:44.849
<v Paul Baumann>As another example, seasonality in the
hospitality industry is an important

193
00:14:44.849 --> 00:14:46.119
<v Paul Baumann>consideration for a hotel.

194
00:14:46.809 --> 00:14:52.729
<v Paul Baumann>The service technicians need to be made
aware by our work orders to perform

195
00:14:52.729 --> 00:14:59.010
<v Paul Baumann>preventive maintenance checks in time for
peak season when a hotel building needs to

196
00:14:59.010 --> 00:15:00.890
<v Paul Baumann>be in prime working condition.

197
00:15:02.059 --> 00:15:07.840
<v Paul Baumann>The system learns this based on the
billing behavior over time and can schedule

198
00:15:07.840 --> 00:15:12.520
<v Paul Baumann>these work orders for the service
technicians autonomously or via facility

199
00:15:12.520 --> 00:15:13.299
<v Paul Baumann>management teams.

200
00:15:14.380 --> 00:15:20.859
<v Isabelle Fritzen>Paul, I believe for the operators to rely
on that data they need to trust it, right?

201
00:15:21.340 --> 00:15:24.250
<v Isabelle Fritzen>So how is that trust established?

202
00:15:25.239 --> 00:15:28.169
<v Paul Baumann>Trust obviously is built in several ways.

203
00:15:28.250 --> 00:15:33.859
<v Paul Baumann>So first, through transparent data models
where operators can clearly see how data

204
00:15:33.859 --> 00:15:36.619
<v Paul Baumann>points are structured and mapped to
equipment.

205
00:15:37.770 --> 00:15:43.570
<v Paul Baumann>Second, through industry standards and
interoperability, which ensures systems can

206
00:15:43.570 --> 00:15:50.010
<v Paul Baumann>communicate reliably, and 3rd, through
security and independent verification.

207
00:15:50.409 --> 00:15:56.979
<v Paul Baumann>For instance, digital platforms like
BillingX and like have achieved Smart System

208
00:15:58.150 --> 00:15:59.969
<v Paul Baumann>verified platinum certification.

209
00:16:00.590 --> 00:16:06.059
<v Paul Baumann>Which validates both the building
management layer and the cloud platform against

210
00:16:06.059 --> 00:16:08.729
<v Paul Baumann>strict cybersecurity and performance
criteria.

211
00:16:09.440 --> 00:16:13.520
<v Paul Baumann>That kind of independent verification
gives building owners confidence that their

212
00:16:13.520 --> 00:16:15.700
<v Paul Baumann>data and operations are protected.

213
00:16:16.539 --> 00:16:21.619
<v Paul Baumann>In the spirit of openness and flexibility,
the billingX data model has been designed

214
00:16:21.619 --> 00:16:27.489
<v Paul Baumann>to be flexible and extensible while also
maintaining a mapping of all major industry

215
00:16:27.489 --> 00:16:28.840
<v Paul Baumann>standards anthologies.

216
00:16:29.750 --> 00:16:36.380
<v Paul Baumann>And data models such as Project Haystack,
BRI schema, Ashtray 223P, Real

217
00:16:36.380 --> 00:16:38.270
<v Paul Baumann>Estate core, and so on and so forth.

218
00:16:39.719 --> 00:16:45.049
<v Isabelle Fritzen>Thank you so much for already sharing a
lot of insights today, but before we wrap

219
00:16:45.049 --> 00:16:47.349
<v Isabelle Fritzen>up, let's make this really practical.

220
00:16:47.530 --> 00:16:52.489
<v Isabelle Fritzen>If a building owner or facility manager
listening to this conversation wants to

221
00:16:52.489 --> 00:16:57.130
<v Isabelle Fritzen>start their digital journey tomorrow, what
are the three things they should think

222
00:16:57.130 --> 00:16:58.039
<v Isabelle Fritzen>about first?

223
00:16:58.409 --> 00:17:00.669
<v Paul Baumann>Well, I would suggest the following 3
things.

224
00:17:01.210 --> 00:17:06.930
<v Paul Baumann>First, treat billing data as a strategic
asset, not just a byproduct of a billing

225
00:17:06.930 --> 00:17:07.469
<v Paul Baumann>system.

226
00:17:08.089 --> 00:17:15.040
<v Paul Baumann>Second, Choose platforms that support open
standards and interoperability so systems

227
00:17:15.040 --> 00:17:17.439
<v Paul Baumann>can integrate easily, and 3.

228
00:17:18.479 --> 00:17:24.079
<v Paul Baumann>Invest early into a strong data
foundation, because once your billing data is

229
00:17:24.079 --> 00:17:30.540
<v Paul Baumann>structured, it becomes much easier to
unlock analytics, automation, and eventually

230
00:17:30.729 --> 00:17:32.260
<v Paul Baumann>self-learning capabilities.

231
00:17:33.079 --> 00:17:39.469
<v Isabelle Fritzen>So the messy truth about building data is
this most buildings already have the

232
00:17:39.469 --> 00:17:40.449
<v Isabelle Fritzen>information they need.

233
00:17:40.760 --> 00:17:44.670
<v Isabelle Fritzen>The real challenge is turning that
information into something usable.

234
00:17:45.040 --> 00:17:50.760
<v Isabelle Fritzen>Once that happens, AI can start connecting
the dots, helping operators see patterns,

235
00:17:51.040 --> 00:17:55.880
<v Isabelle Fritzen>make better decisions, and gradually move
toward buildings that improve themselves

236
00:17:55.880 --> 00:17:56.680
<v Isabelle Fritzen>over time.

237
00:17:57.020 --> 00:18:02.140
<v Isabelle Fritzen>And that's where the future of
self-learning autonomous buildings really begins.

238
00:18:02.500 --> 00:18:05.180
<v Isabelle Fritzen>Paul, thanks so much for joining us today.

239
00:18:05.530 --> 00:18:06.000
<v Paul Baumann>Thank you.
