Data-Driven Metrics Applied to Traffic Crashes to Improve Observability in Smart Cities

Abstract

Data is a crucial factor for monitoring and understanding events related to Smart Cities. Data can be discovered and integrated from different sources and has the potential to be interpreted in multiple ways. Traffic crashes, for example, are common events that occur in cities. A significant amount of historical data related to traffic crashes is publicly available for analysis and can be used by a wide range of stakeholders. Measuring the impact of Smart Cities solutions usually relies on data collection, analysis, and metrics before and after such solutions are implemented. This paper presents an observable data-driven bottom-up methodology to create the Critical Composite Index (CCI), a Key Performance Indicator developed to measure traffic crash severity as a singular value. The CCI can be used by both domain experts and non-domain experts to be informed about traffic crashes on the roadways. This paper the development of the CCI using historical, government agency reported, and publicly accessible traffic crash data. The CCI can be modified or extended to align with specific reporting traffic crash criteria by modifying the weights of traffic crash features. The observable data-driven bottom-up methodology development enables the transformation of raw data into a metric that can contribute to the observability of Smart Cities.