Mainflux Labs is a Member of the Construction R&D Project Funded by the EU H2020 Program
Mainflux Labs is a consortium member of the Ashvin: EU H2020 funded, international R&D project which goal is to address long-lasting challenges of the construction industry using digital twins and the power of IoT as assistants for design, construction and maintenance of civil engineering infrastructures.
In the Serbian medieval epic poem, Building of Skadar, three brothers are trying for three years to build a fortress with 300 workers but without any success. What they develop during the day, mountain fairy destroys over the night. Finally, in order to build their fortress, the youngest brother had to wall up his young loving wife alive, within the walls of the fortress as a sacrifice demanded by the fairy.
It seems that a similar destiny pattern from this middle ages legend still follows the construction industry in the 21st century.
According to McKinsey, large projects typically take 20 percent longer to finish than scheduled, and regarding safety, construction is yet by far the most dangerous occupation with a very high rate of fatalities compared to all other industries. Eurostat statistics shows, for example, that only in 2018, 3319 non-fatal accidents were observed per 100 000 persons employed and 5.6 fatal accidents per 100 000 persons employed per year.
Despite the technological developments, the construction industry struggles with improving safety, and besides inaccurate scheduling, also has notoriously low productivity compared to other industries, including large projects usually up to 80 percent over budget (McKinsey).
The value which digital technologies can bring to Architecture, Engineering and Construction, comes from the whole new spectrum of possibilities associated with the real-time continuous data acquisition, management, and provision of as-built conditions.
Digital Twin technology
One of these possibilities is technology referred to in the industry as a Digital Twin, a dynamic digital/virtual representation of real-world or potential physical objects, assets, devices, systems and processes.
Updated in real-time, it can provide detailed information, feedback and insights about its real-world counterparts (on their properties, status and performance ) with the goals to optimize processes, detect current and potential problems, predict results, and build better products.
Relying on the Internet of Things, sensors, realistic simulation and advances in computer vision and machine learning technologies , it provides masses of cumulative measurements of its real-world counterpart and digital reproductions and simulations of the asset. Information is then aggregated in the form of different digital representations, graph models, visualizations and interfaces. Moreover, it enables non-traditional approaches in domains such as productivity monitoring, safety management, 3d reconstruction from raw video input, and remote inspection for predictive maintenance.
As such, Digital Twin is:
- an access point for new machine-learning apps, algorithms, analytics, simulation and visualization technologies, and applications
- foundation for process optimizations, detection of current and potential problems, prediction of results and design of better products
Industry is still slow in adopting new digital technologies, but in recent years different concrete initiatives occur that are trying to address some of these problems.
We at Mainflux Labs are excited to participate in one of these initiatives, namely, ASHVIN — Assistants for Healthy, Safe, and Productive Virtual Construction Design, Operation & Maintenance using a Digital Twin, funded by the EU H2020 Research and Innovation Program on Digital Building Twins (RIA). Led by Technical University Berlin (Technische Universität Berlin) under the Grant Agreement 958161, Ashvin project receives an overall budget of € 5 609 858,75.
Ashvin consortium includes 14 partners from 9 EU member states, with substantial expertise in construction and engineering management, digital twin technology, IoT, and data security/privacy. The project goal is to provide digital solutions with accompanying usage guidelines to improve scheduling forecast, allocation of resources, reduction of costs, and reduce the number of accidents in the construction domain.
Digital Twin within construction industry and Ashvin project context
Within the context of the Ashvin project and construction domain, the digital twins will provide a dynamic digital representation of buildings or infrastructure systems as well as a representation of all important processes around them throughout its product development lifecycle.
Enabled by the IoT technology which provides data from sensors, photogrammetric data, laser and thermal scans, images and video, Digital Twin in this context will include the capacity to accurately simulate buildings’ and infrastructure’ multi-physics behavior and to provide real-time synchronizations between as-designed and as-built models from design /engineering phases, to construction and to maintenance stages.
But furthermore in order to fully leverage on IoT based digital twins, the Ashvin project conceived development of machine learning and AI solutions for data processing and analysis as well as a wide range of design and engineering applications.
Anatomy of Ashvin
Ashvin comprehensive solution will consist of several technology layers:
i) sensing and video/image/monitoring
This technological layer will exploit automatic image-based modelling from sequentially acquired images to evaluate the objects’ geometry, translating 2D images from handheld or drone footage into 3D reflectances.
Moreover, computer vision techniques will be deployed to support automatic drone or robot inspections in construction sites
ii) edge computing
edge computing gateway and edge computing software
iii) IoT platform and game-engine digital twin platform
iv) IoT and data fusion techniques, based on AIdata association techniques to enhance the confidence and the reliability of sensor measurements, improve the spatial and temporal coverage, reduce data ambiguity and contribute in decision Making in demanding tasks such as Productivity monitoring, Safety Management and energy-efficient design.
v) Design and engineering applications — Ashvin Toolkit
Autodesk Dynamo Parametric modeling software, ‘R’ a general-purpose statistical programming environment, AnyLogic simulation platform, Energy Plus Building Energy Simulation, FE-based multi-physics applications, QGIS software, an open-source GIS software application.
Role of the Mainflux Labs
A considerable degree of the ASHVIN project will be dedicated to integrating these technology layers into a fully interoperable and linked platform — the final ASHVIN digital twin platform.
The first integrations are between Mainflux IoT Platform and sensing and monitoring devices, and the Digital Twin Technology game engine-based digital twin platform.
Then comes support and integration with different data fusion techniques, and processing and analysis algorithms and applications. And finally comes design and engineering applications that had to be integrated with the IoT platform.
As indicated so far, the Ashvin Digital Twin platform will be based on the Mainflux open-source IoT Platform. It will serve as the underlying fabric by providing a registry (database) of digitized real-world IoT objects and HTTP REST APIs with CRUD functionality and configuration endpoints necessary for data processing, data acquisition, real-time state transitions, as well as, AI-supported data analysis and simulation capabilities.
Mainflux has the capability to connect to millions of devices supporting a large amount of IoT protocols, such as several secure low-power IoT protocols (MQTT, CoAP, LoRaWAN, LwM2M) or several physical industrial protocols (OPC-UA, Modbus, BLE). The platform is also highly secure: All connections are encrypted via TLS v1.3 and with mutual TLS client-side X.509 certificates — currently representing the highest industry standard.
After all integrations are completed, in order to validate and provide clear evidence of its benefits, Digital Twin Platform and developed ASHVIN apps/toolkit, will be demonstrated on the 9 real construction building, bridge, and industrial projects across Europe.
At the end of the project, the ASHVIN digital twin platform and its applications will enable architects, contractors, consultants, owners, and other stakeholders to utilize data from a wide range of IoT enabled devices, as well as, from different BIM-based design applications all with dashboards for continuous real-time monitoring and control, as well as analysis and decision making.