TwinCity3000: a Tale of Twin Cities

The usage and application of urban digital twins have risen in the last few years, while no clear definition of what constitutes an urban digital twin exists. Therefore, their adoption is usually confined to use cases with heavily predefined goals, leading to purpose-built applications with no or very little room for adaptation for other usages. This leads to the especially timely and costly creation of one-of applications that are difficult to integrate with new or existing workflows, hindering the interoperability of different domains. To move beyond the scope of the original collaboration, an extensible, reusable and open-source framework is required. [1]

Building models in the CityGML format can be used to lay a foundation for urban digital twins. Many European cities distribute them as a representation of the building footprint extruded to match the roof structure detected by different surveying methods, like airborne lidar [2]. However, these models are usually the product of a bulk geometry creation, leading to outdated urban models at the time of their publishing. In most cases, they also contain wrongly modelled geometries, creating the need for building modification and creation. While many urban (related) databases exist, they are often provided as standalone products with no easy integration options, leaving their potential often unused.

We present TwinCity3000, a web-based framework that utilizes the CityJSON CityGML encoding and includes different data sources using open standards that allow for native integration of diverse data sources. With TwinCity3000, we aim to provide urban planners, architects, engineers, and decision-makers with a unified, interactive, modular framework that allows them to bridge the gap between different domains more efficiently. Visual first Create-Replace-Update-Delete (CRUD) operations combined with the built-in versioning tool allow for intuitive handling of existing and new building models.

The integration of standardized APIs makes it easy to extend and implement many open and private data sources directly into one central tool. The framework distinguishes itself from others by focusing solely on the district and building block level, where most of the planning and decision-making in the urban setting takes place. Higher-order abstraction enables users from different backgrounds to participate easily, from early planning stages in the design process to well within the monitoring stage of existing districts. This will enable better participation in urban planning, leading to more informed decisions and better-designed urban environments.