Employees Comfort measurement in offices is really important to increase the productivity of the teams. Digital Twins enable Building Managers to interact with building 3D models, monitor environmental data, implement cost savings sustainable models and create what-if simulation scenarios. During this talk a real world use case of Digital Twin implementation will be presented deep diving on the architectural choices and considerations that made the implementation fast and easy only through Open source solutions.

1. Antonino Sirchia
Accelerating the development
of AI-enabled Digital Twins

2. Engineering Presentation Template 2
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

3. Senior Researcher at Engineering Ingegneria Informatica Spa.
I hold a Master’s Degree in Software Engineering from the University of Palermo.
In 2017 I started working as a Technical Manager of the Digital Enabler Ecosystem Platform of
Engineering Ingegneria Informatica.
I’ve experience both in market-oriented and R&D projects in multiple domains like Smart
Mobility, Environmental monitoring, Smart Building, e-Health, Smart Agriculture, Smart Lighting.
Antonino Sirchia
Senior Researcher
Engineering
3

4. 4
At a Glance: a Global Company
READ WHITE PAPER
1.3 Bn€
REVENUE FY2021
OFFICES
PROJECTS
40+
YEARS OF GROWTH
RESEARCH
& INNOVATION
6 Development Labs
100+ Live Research Projects
€40m+ Investments / YR
450+ Data Scientists and Researchers
IT & MANAGEMENT ACADEMY
“ENRICO DELLA VALLE"
25,000 Man Days of Training / YR
1,300+ Professional Certifications
9,900+ Web Classroom Participants
ASSETS
11 Portfolio Areas
20+ Group Companies
10+ Competence Center
3 Data Center Tier IV
22K Managed Servers
250K Workplaces Managed
ADVISORY TECHNOLOGY &
IMPLEMENTATION
PROPRIETARY
SOLUTIONS
MANAGED
SERVICES
11,600+
EMPLOYEES
60+
OFFICES AROUND
THE WORLD
GLOBAL HQ
ROME, ITALY
30+
PROPRIETARY SOLUTIONS
FOR ALL MARKET SECTORS

5. Engineering Presentation Template 5
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

6. 6
What a Digital Twin is
“A digital twin is a real-time
virtual representation of a
real-world physical system or
process that serves as the
indistinguishable digital
counterpart of it for practical
purposes, such as system
simulation, integration,
testing, monitoring, and
maintenance.”
Source: Wikipedia
Digital
Twin

7. 7
Simulation
 Analyze effects of an action without
actually put it in place in the real
world
 Fasten weaknesses and strengths
detection
Monitoring and testing
 Keep the «physical twin» under control
 Be always aware about the current
values for defined key indicators
 Validate decisions that led to some
actions on the real-world
 Collect evidences for new decisions
and actions
Integration
 Digitally integrate systems that are
physically segregated (siloes)
 Create new knowledge coming from
the “mashup” of scattered data sources
(e.g. devices, open data, legacy
systems)
 Apply AI/ML models to predict events
or find correlations between apparently
unrelated data
Maintenance and Reactivity
 Plan maintenance promptly
 Apply predictive maintenance
 React automatically on some events
 Reduce the inefficiencies

8. Engineering Presentation Template 8
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

9. Digital Twins and Smart Buildings 9
Digital Twins and Smart Buildings
ENGINEERING / THE DIGITAL TRANSFORMATION COMPANY
Detect failures
In case some systems, like the air-conditioning or the heating ones, fail the Building Manager
would be able to promptly trigger the maintenance procedures.
Increase energy efficiency
Through the automatic regulation of physical assets (e.g. actuations on windows, curtains,
lights, air-conditioning, heating, etc.…) the building’s energy consumption can be drastically reduced;
causing benefits for the wallets and for the environment.
Increase the employees comfort
Employees comfort monitoring in offices is really important to increase the productivity of the teams.
Through a correct parameterization of the Digital Twin, the comfort level in a building can be kept
always at optimum levels increasing, de-facto, the people productivity.

10. Engineering Presentation Template 10
• The correlation between Comfort and Productivity is well
known since 2008
• Productivity increases and decreases in accordance with
comfort
• Digital Twins are just a new tool in the box to monitor and
react appropriately to ensure an optimum comfort to
employees
Haynes, Barry. (2008). The impact of office comfort on productivity.
Journal of Facilities Management. 6. 37-51. 10.1108/14725960810847459.
Comfort and Productivity
ENGINEERING / THE DIGITAL TRANSFORMATION COMPANY
10
Digital Twins and Smart Buildings

11. Engineering Presentation Template 11
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

12. Real world use case in Italy 12
Focchi premise Digital Twin
ENGINEERING / THE DIGITAL TRANSFORMATION COMPANY
Data from sensors
• Surface temperature
• Environmental temperature
• CO2 concentration
• Humidity
• Luminosity
• Sunlight radiance
• External weather station
Building topology
• Rooms map
• Sensor location
• IFC 3D Model
Calculated data:
• Comfort per room
• Productivity variance per room
Picture source: https://focchi.it

13. 13
Simulation
 Analyze effects of an action without
actually put it in place in the real
world
 Fasten weaknesses and strengths
detection
Monitoring and testing
 Keep the «physical twin» under control
 Be always aware about the current
values for defined key indicators
 Validate decisions that led to some
actions on the real-world
 Collect evidences for new decisions
and actions
Integration
 Digitally integrate systems that are
physically segregated (siloes)
 Create new knowledge coming from
the “mashup” of scattered data sources
(e.g. devices, open data, legacy
systems)
 Apply AI/ML models to predict events
or find correlations between apparently
unrelated data
Maintenance and Reaction
 Plan maintenance promptly
 Apply predictive maintenance
 React automatically on some events
 Reduce the inefficiencies
Focchi premise Digital Twin

16. Engineering Presentation Template 16
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

17. Keypoints:
• Protocol transformation on the Edge
• Dynamic Device Management
• Data Harmonization
• Context Management
• Rule-based automatic reaction
• Artificial Intelligence models training
and usage
• 3D visualization on web
17
A Digital Twin architecture
Control plane
Data plane
Command plane

18. Benefits:
• No need to support too many protocols on the
cloud
• Anticipate mission critical operations «near» to
the source
• Compliance with privacy policies
Pitfalls:
• Need for connection agreements with the
customer
• Resource limitations
18
Protocol transformation on the Edge
Control plane
Data plane
Command plane

19. Benefits:
• Possibility to register new devices on the fly
• Possibility to have a unique control room for
multiple device «hubs» like AWS IoT, Azure IoT,
edge installations and others
• Possibility to update device firmwares remotely
• Possibility to install processing bundles remotely
19
Dynamic Device Management
Control plane
Data plane
Command plane

20. Benefits:
• Easiness of analysis
• Easiness of rule application
• Increased interoperability with third party systems
Pitfalls:
• One step more to manage in the data lifecycle
20
Data Harmonization
Control plane
Data plane
Command plane

21. Adopted standard: Smart Data Models
Ref.: https://smartdatamodels.org/
Data models used:
• Building (LINK)
• WeatherObserved (LINK)
• AirQualityObserved (LINK)
• KeyPerformanceIndicator (LINK)
The program provides many more data models that can be used to create even more complex twins.
e.g. people, building groups, neighborhoods, cities, …
21
Data Harmonization

22. 22
Data Harmonization – example
INPUT DATA (UL 2.0):
v0|2577|v1|2603|…
HARMONIZED DATA (JSON):
{
‘temperature’: 22.7,
‘rain’: 0,
‘windSpeed’: 17,
…
}

23. Benefits:
• You always have a «picture» of the current status
of your twin
• Data Brokerage point across the system
23
Context Management
Control plane
Data plane
Command plane

24. Benefits:
• Automatization of simple and repetitive data
analysis in RT
• Enabler for smart interactions with the field
24
Rule-based automatic reaction
Control plane
Data plane
Command plane

25. Benefits:
• Forecasting on data
• Correlation analysis
• Improvement of employees comfort and
productivity
25
Artificial Intelligence models
training and usage
Control plane
Data plane
Command plane

26. Benefits:
• Immersive experience for the end user
• Easiness of representation of data
26
3D visualization on web
Control plane
Data plane
Command plane

27. Engineering Presentation Template 27
1. What a Digital Twin is
2. Digital Twin and Smart Buildings
3. Real world use case in Italy
4. The architecture
5. The implementation
Agenda

28. The implementation 28
Digital Enabler ™ allows you to
collect, harmonize, integrate,
analyse and render multi-
dimensional and multi-source data.
The Engineering’s Digital Ecosystem Platform

29. • Native IoT Management (HTTP, MQTT, LoRa devices)
• Federation of multi-cloud device management hubs (e.g. AWS IoT, Azure IoT, …)
• (Near-)Edge management according to the Liquid Computing pattern
• Definition of workflows and batch processes
• Data harmonization into standard data models through no-code tools
• IoT actuation rules definition and complex events processing
• AI/ML models training, testing and usage
• Serverless functions implementation and database provisioning
• Identity and Access management
• Dashboards and visualizations (2D and 3D) creation
Everything powered by Open Source tools!
The implementation 29
Digital Enabler services

30. Keypoints:
• Usage of Digital Enabler IoT Device Manager
• Orchestration of IoT devices managed by other
platforms (such as AWS, Azure, etc).
• Usage of a FIWARE Hub
• MQTT protocol
• IoT Agent UL2.0
• Orion Context Broker
30
Device Management

31. Keypoints:
• Native Support for most of the Smart Data
models
• Many native operators for atomic operations
• Multiple triggers (HTTP, MQTT, Kafka, Schedule)
• Multiple sinks (HTTP, MQTT, Kafka)
31
No-code Data Harmonization

32. 32

33. Keypoints:
• Powered by KNative
• Multi language (JS, Java, Python, etc…)
• Horizontal autoscaling
• CLI available for local development
• Functions can be exposed via HTTP
33
Serverless function implementation

34. Keypoints:
• Based on IFC standard
• Powered by xeokit bim viewer
opensource library
34
3D visualization

35. Engineering Ingegneria Informatica:
https://eng.it
Digital Enabler platform:
https://www.eng.it/our-platforms-solutions/digital-enabler
For more information come and talk with us at the desk!
References
#LifeAtEngineering
#ThinkDigital
#hiring #WorkWithUs

36. Antonino Sirchia
Thank you
ENGINEERING / THE DIGITAL TRANSFORMATION COMPANY