In order to support automated decision-making, landscape and urban planning require the evaluation of alternative scenarios within the context of geodesign. The evaluation is frequently based on indicators, where the decisive ones are typically called key performance indicators (KPIs). These KPIs often are related to physical objects (e.g. buildings or land parcels) which are stored in geoinformation systems (GIS). Several approaches have been introduced for representing KPIs; however, they pose a problem, as stakeholders and domain specialists often are not capable, lack sufficient knowledge, or are not willing to implement these indicators in the language of the underlying GIS. In this paper we propose a framework that assists domain specialists in expressing indicators, indexes, and their dependencies using a model-driven approach. We define an object-oriented data model for an abstract General Indicator Model (GIM) formally specifying concepts like indicators, numeric indicators, and their compositions. Specific indicators/KPIs from different decision contexts, for e.g. energetic, environmental, or financial assessments, are then defined as concrete subclasses of the GIM. The concrete KPI classes are linked to spatial feature classes from digital city and landscape models (like CityGML) using model weaving. This effectively sets the object context or reference frame of the individual indicators and provides means to automatically derive the values of those indicators from characteristics and attributes of the linked spatial objects. We apply this framework to a case study for a real scenario in energy demand estimation as a proof of concept for our framework. However, the concept can also be used for indicators from other indicator domains to cover e.g. environmental and financial aspects of planned scenarios in decision-making as well.     «

In order to support automated decision-making, landscape and urban planning require the evaluation of alternative scenarios within the context of geodesign. The evaluation is frequently based on indicators, where the decisive ones are typically called key performance indicators (KPIs). These KPIs often are related to physical objects (e.g. buildings or land parcels) which are stored in geoinformation systems (GIS). Several approaches have been introduced for representing KPIs; however, they pos...     »