The objective of this task was to select appropriate in-depth Powered Two Wheeler (PTW) accident data and to derive a common methodology for the review of these in-depth cases. The information gathered from this process is an input to later activities reported in D13&D14 to understand the issues of accident causation in order to identify functional requirements of intervention methods that would be effective for accident avoidance or mitigation. Existing in-depth databases, comprising a forensic accident database and the COST 327 database held by LMU, the UK On The Spot (OTS) database and the UK Fatals accident database held by TRL, and the UK OTS database held by the VSRC were used. Accident scenarios previously defined by the APROSYS project formed the basis for the selection of relevant accidents. National data from Germany, Italy, Spain and the Netherlands was shown to fit well these APROSYS scenarios. However, since the in-depth datasets available to the PISa project were largely from the UK OTS project, it was necessary to consider whether the accident scenarios identified in Task 2.1.1 are also relevant in the accident statistics for Great Britain and to confirm if, and to what extent, they differ. A review of the national statistics for Great Britain (STATS19) was therefore undertaken. Although this analysis showed that the priority of the scenarios is different, largely due to the different kinds of PTW in the UK fleet compared to that in Europe, it was concluded that a vast majority of the STATS19 accident cases could be described using the APROSYS scenarios. Analysis was also carried out to ensure that the two regional OTS databases held by TRL and the VSRC were representative of the national GB statistics. Thus, as a result of the general applicability of the APROSYS scenarios and the representativeness of the OTS data, it was considered that the in-depth cases from Great Britain could be used and that the findings form these cases were applicable to European accidents in general. Each team reviewed their existing in-depth accident cases according to an outline of accident characteristics and selected, according to a developed set of selection criteria, a number of cases (+/- 20) which fell within the scenarios of relevance as identified by 2.1.1 and which contained sufficient in-depth information. The selected cases were analysed in detail to determine their characteristics (reciprocal vehicle positions, vehicle speeds, etc), which in turn could allow the detection of a dangerous situation (e.g. stability hazard) or a pre-crash condition. A series of inter-team workshops was held to establish a common understanding to the analysis of the in-depth accident cases. This analysis addressed the pre-crash, crash and post crash accident phases. Case summaries were produced to aid the inter-team case review and validation process. Drive through video data for each accident case (where available) was also used to obtain aspects of pre-crash information. Deliverable D11&D15 - Report summarising the in-depth accident case and video analysis - provides further information. The detailed case reviews considered accident characteristics including accident causation, vehicle (PTW and opponent) characteristics, environmental factors, human factors, PTW rider and opponent vehicle occupant characteristics. All the selected cases were reviewed by all three teams to clarify and confirm the issues and employ a common understanding and approach to the case analysis process. The result of this analysis was the finalisation of a list containing 43 intervention functions taken from the in-depth and fatal accident case analyses. This list was taken forward into Tasks 2.3.1 and 2.3.2 in order to assign safety functions to each of the functions and then prioritise them as described in Deliverable D13&D14 - Report summarising the selection and prioritisation of solutions. (Author/publisher)
Abstract