SDDI Architecture

The SDDI has a modular structure and defines an organisational and technical framework consisting of actors, applications, registry, sensors, an urban analytics toolkit, and a virtual district model. Actors are citizens and the city administration, but can also be other stakeholders like public transport services, utility service providers, and real estate firms. IoT and Sensor data comprise local weather and climate stations, regional weather radar, smart meters for energy, gas, and water consumption, video cameras, and traffic sensors. A resource registry is a piece of software which provides persistent storage of registered resources like web services, databases, etc. But, meanwhile provides useful information about each of the registered resources. This helps the user/client to faster and more efficiently look for the required information. Urban analytic tools are software components that, for example, estimate the energy demands or potentials of solar energy production for all buildings, simulate road traffic and pedestrian flows, or perform noise propagation or flooding simulations. Virtual District Model (VDM) which itself covers three main components, Virtual 3D City Model, Building Information Model (BIM), and utility network. The SDDI is based on Open Standards and links systems of different manufacturers in a non-proprietary and extensible way.

What makes the SDDI framework unique compared to others within the field of Smart Cities is the fact that all information, sensors, and applications are located within a semantic 3D city model. The latter is a virtual representation of the physical reality of the district. It consists of the most relevant objects like buildings, streets, vegetation, water bodies, and networks. The 3D model is based on the international standard CityGML and does not only serve for neat visualizations; it is an information hub and essential foundation for most simulations and analytic tools. Within this virtual district model, for example, the energy demands of buildings can be put in relation to their physical conditions and their socio-economic key performance indicators. This way, the impact of planned urban redevelopment projects on the different thematic fields like the environment, mobility, energy, and social affairs can be investigated at the same time.

SDDI is designed considering the requirements of the districts as well as aiming at providing a framework which offers a bottom-up approach for the effective integration of solutions. The key characteristics of the SDDI framework are as follows:

• Redundancy avoidance: In many cases there are datasets describing or related to a specific object. This object is often defined differently in different sources or by various providers. This leads to ambiguity and redundancy of the data which need to be interpreted later. For example, applications such as energy simulation, pedestrian flow simulation, applications involving real-time sensor observations in buildings all require to work with information about the districts’ buildings which might be respresented redundantly within each of the applications. In order to avoid data redundancy, standards play a crucial role. SDDI is designed based on standards from OGC and ISO. For example CityGML can be used to represent buildings just once for all of the applications mentioned above.
• Well-specified data semantics: The challenging point here is that the data are often interpreted differently. This leads to the misuse of the data over time by different users. It is crucial to use data models which present meaningful information understandable by everyone, and therefore a well-specified data semantic is needed. Standards from ISO or OGC are good examples for this characteristic and are considered in the SDDI model.
• Virtual District Model: The two aspects “redundancy avoidance” and “well-specified data semantics” are addressed by introducing a virtual district model (VDM). The VDM contains objects such as buildings, roads, city furniture, water bodies, etc. in addition to networks such as water utility, smart grid or transportation networks. (Percivall et al. 2015) argue that space is a principle method to organize the Smart City. From our point of view space (coordinates, geometry) is not the only method but semantic objects (with spatial properties) – as they are provided by the VDM – should be used as a common denominator for representing and organizing the information pieces from the various application domains of the Smart Districts. Our detailed analyses of the SSD deep dive districts clearly show that nearly all thematic and sensor information are directly related to the objects of the VDM. Some sensors are even measuring properties of the real world objects (e.g. Smart Meters are measuring the power consumption of buildings). Hence, linking the sensors with the respective building objects and properties implicitly specifies the semantics of the sensor observations.
  The VDM is a response to what is mostly missing in the data management of today’s other Smart City initiatives. This is the management of the data through a common digital model of the physical urban environment as the information hub. This can be seen, for example, in IoT and Big Data analytics centered Smart City concepts where obviously the concept of linking the devices to a common data hub is lacking. Based on the experiences gained in the SSD project and through the work with various districts, we can conclude that for almost all cases the districts need to work with or refer to district objects in one way or another. These objects are defined regarding their locations and their physical characteristics in the real world. Hence, it is necessary to have a virtual model of these physical elements of the area – whether it is just for a district or the entire city. Above, the VDM is also key to diverse types of simulations (e.g. energy, traffic, and environmental simulations) and to the estimation of the impacts of planned changes to the district.
• Interoperability: According to ISO 2382-1 (c.f. ISO/IEC 10746-2:2009), the term interoperability is defined as “the capability to communicate, execute programs, or transfer data among various functional units in a manner that requires the user to have little or no knowledge of the unique characteristics of those units”. Interoperability is one of the most important characteristics of this model which covers the semantic and syntactic interoperability. This in fact is the key role of the SDDI which overcomes obstacles such as institutional barriers and avoids vendor lock-in, thus providing openness for extensions, and leading to the sharing of information.
• Extensibility: The realization of the SDDI as a modular, open, and interoperable set of distributed functional components ensures the easy extensibility by new stakeholders, users, sensors, thematic information, and analysis tools. Furthermore, the model should not be stopped at the current development, as technologies are rapidly developing. The structure of SDDI is designed in a way which can be extended in order to meet the future needs and cases.
• Functionality: A standard solution ensures the functionality of the approach and model apart from the use cases. This means that the model is designed such that it can be used for different use cases.
• Transferability: What makes SDDI powerful is that this platform is not developed only to be implemented for one use case or one district but to be implemented in different places in similar ways. This characteristic of SDDI again is due to the extensive use of standards in this infrastructure. For example, there are many cities in the world, which have already developed the 3D model of their cities following the OGC CityGML standard.