Abstract:

For most DOTs, crash data are presented very traditionally through evaluating single attributes and presenting the information as either a bar chart, line chart, or in a table. The data are typically presented as a trend, to show the number of crashes over time, or table when multiple values are present. In Iowa, over 130+ coded or derived fields are included in the crash data. However, only a handful of these variables are ever presented in the crash facts to avoid inundating user with hundreds of tables and charts. In contrast, a crash dashboard can provide users with an interactive experience to understand the relationships between crash variables and to graphically filter through the data to further understand relationships that are key to identifying strategies to effectively address safety. As part of an effort with the Iowa DOT’s Motor Vehicle Enforcement (MVE) and the Iowa State Patrol (ISP), an interactive dashboard was developed to focus on heavy truck crashes and to allow officers to target enforcement actions. The dashboard provides officers with a data driven tool which is routinely used to develop enforcement strategies based on updated crash data as opposed to data from multiple years ago. The charts and tabs were created through input on desired functionality by MVE and ISP users. For example, users wanted a map to graphically show the location of the crashes being selected and this is retained across all modules. The web-based dashboard (https://reactor.ctre.iastate.edu/heavy-truck-crash-tool/) is separated into eight modules which organizes similar crash attributes together. For example the ‘Time’ module includes crash statistics such as time of day, day of week, lighting conditions, and month. Users begin on an overview page (left image) which provides an introduction of the tool and data along with several initial context filters. The dashboard is entirely interactive which allows users to navigate between the module then hover over the charts to see more information or to use any of the charts as filters to dive deeper into a specific crash attribute. Hovering over any of the charts will provide a trendline for the corresponding variable by federal fiscal year. Figure 1, right side, shows the increasing trend while hovering over driver distraction. Selecting a variable (or multiple) will filter all of the charts across all modules to only include data for that corresponding attribute. This allows users to explore the relationship between variables by understanding the impacts specific attributes have on other crash attributes. As an example, the Iowa State Patrol began prioritizing vehicles that were following too close after they identified that the major cause of “Followed too close” increased when they filtered the dashboard by the routes they covered and the types of crashes they were focused on preventing. The functionality of the heavy truck dashboard has allowed MVE and ISP to continually improve and update their enforcement strategies to most effectively address heavy truck safety in Iowa.