Using actor maps and AcciMaps for road safety investigations: Development of taxonomies and meta-analyses

The number of road collisions have plateaued over the past decade, both in the UK and worldwide. One problem is that most road safety investigations focus on the immediate, sharp-end factors rather than delving into the deeper, blunt-end, systems factors. It is argued that the contributions of both of these types of factors need to be understood before further reductions in road collisions can be made. The study reported in this paper developed taxonomies for Actor Maps and AcciMaps from 37 road collision investigation reports undertaken in the UK. The meta-analysis of the Actor Maps showed that relatively few categories of actors are associated with the majority of collisions (i.e., 35/256). Similarly, the meta-analysis of the AcciMaps showed that all of the 1656 actions, events and decisions (or lack of) could be placed into 19 categories. Across the eight AcciMap levels there were 11 categories that appeared most frequently. Both of these taxonomies together with the meta-analysis enabled a summary of the analysis and derivation of interventions at a national level. The study also points toward a common contributory (and protective) network for road collisions. The meta-analysis showed that the ‘sharp-end’ accounted for approximately 40% of the factors (those that are normally investigated in collisions) whilst the blunt-end accounted for approximately 60% of the factors. These blunt-end factors are often not addressed in traditional police-led collision investigations, as these investigations typically focus on establishing individual criminal culpability. Any future road collision investigations which aim to identify no blame safety learning should seek to understand blunt end factors as they create the pre-conditions for incidents to occur.