Bad behaviour or bad luck? Policy approach = define problems and - - PDF document

Bad behaviour or bad luck?

The role of extreme behaviour in the

• ccurrence of severe crashes

Freya Slootmans, Tim De Ceunynck, Stijn Daniels

Introduction ▸Policy approach = define problems and tackle them ▸Belief in ‘mono-causality’ is widespread ▸Crash causation theories: ▸Crashes can be attributed to main causes (triangle Human-Vehicle-Environment) ▸Removing these causes is likely to solve/mitigate the problem ▸Causes are often ‘bad behaviour’

Source: Shinar (2007) , originally from Rumar (1985)

Errors and violations (Reason et al., 1990) ▸Errors: ▸Failure of planned actions to achieve their intended consequences ▸Violations: ▸Deliberate deviations from those practices believed necessary to maintain the safe operation of a

potentially hazardous system

Objective ▸ Quantify the role of extreme behaviour in severe crashes

Data ▸Expert reconstruction reports ▸Severe crashes ▸Antwerp, Belgium ▸165 cases, between 2013 and 2016 ▸64 fatal crashes ▸161 crashes with ≥1 seriously injured ▸Codebook based on the IGLAD-framework (www.iglad.net)

Method ▸Each crash assigned to one of three categories

Categorisation of crashes (≈Wundersitz et al., 2014) ▸Extreme behaviour ▸Infrequently occurring but mostly likely deliberate behaviour, that is known to be a

clear risk factor in traffic

▸Illegal system failure ▸Road user makes error but is not completely compliant ▸System failure ▸Errors made by compliant road users

Extreme behaviour ▸At least one of the following behaviours: ▸“Extreme speeding”, i.e. ≥ 30 km/h above the legal speed limit; ▸“Extreme DUI”, which is driving with a BAC of ≥0.8 g/l (For professional drivers ≥ 0.2 g/l ▸Driving under the influence of drugs; ▸Driving through a red light; ▸Or a combination of ≥2 of the following behaviours: ▸Alcohol BAC between 0.5 g/l and 0.8 g/l; ▸Speeding, i.e. between 10 km/h and 29 km/h above the legal speed limit; ▸Intentional risky behaviour, such as overtaking someone at an intersection, entering an

intersection without looking, driving a vehicle that has no lighting;

▸Driving without having the driver's license required to drive the vehicle in question; ▸Not wearing a seat belt if required (only applicable for people killed in the crash) ▸Not wearing a helmet when mandatory; ▸Walking through a red light (pedestrians).

Illegal system failure ▸≥ 1 of the following behaviours: ▸Alcohol BAC between 0.5 g/l and 0.8 g/l; ▸Speeding, i.e. between 10 km/h and 29 km/h above the legal speed limit; ▸Intentional risky behaviour, such as overtaking someone at an intersection, entering an

intersection without looking, driving a vehicle that has no lighting;

▸Driving without having the driver's license required to drive the vehicle in question; ▸Not wearing a seat belt if required ▸Not wearing a helmet when mandatory; ▸Walking through a red light (pedestrians).

System failure ▸Crashes that cannot clearly be associated with deliberate behaviour such as a lack of rule

compliance, basically the result of errors of compliant road users.

Results (1) Summary of the role of system failures and extreme behaviour

All crashes Fatal crashes Non-fatal crashes # % # % # % System failure 101 61% 34 53% 67 66% Illegal system failure 23 14% 12 19% 11 11% Extreme behaviour 41 25% 18 28% 23 23% Total 165 64 101

Results (2) Types of behaviour found in ‘extreme behaviour’ crashes (n= 41)

Type of behaviour # immediate extreme behavior # combination

• f ≥2 risky

behaviors % Excessive speed (Extreme speeding or speeding) 22 6 68% Driving under the influence 19 1 49% Running a red light 10 4 34% Personal security (seat belt, helmet) 3 1 10% Vehicle requirements (mirrors, lights, tyres) 3 7% Other behavior (a.o. distraction, not looking) 2 7 22% Infractions (a.o. lighting not used, incorrect

• vertaking…)

2 1 7%

Results (3) Behavioural factors shown by the road users classified as ‘illegal system failure’ (n=23)

Results (4) Crash-related characteristics

▸ Extreme behaviour more frequent ▸ Nighttime crashes (14/27) vs. daytime crashes (27/138) (p 0.001) ▸ Weekends (18/46) vs. weekdays (23/119) (p 0.004) ▸ Darkness (22/51) vs. daylight (15/97) (p 0.002) ▸ Extreme behaviour not significnty different according to do: ▸ Number of vehicles involved ▸ Severity of the crash ▸ Weather condition

Limitations

▸ Sample not representative ▸ Urban area ▸ Most severe crashes ▸ The truth is in the eye of the beholder ▸ Experienced coder ▸ Clear protocol ▸ BUT: still room for interpretation: ‘risky behaviour’, ‘distraction’, ‘intention’, ‘moderate offences’… ▸ Categorisation might be too simplistic ▸ ‘Extreme behaviour’ vs. ‘system failure’

Conclusions ▸Even among severe crashes, only a minority is associated with extreme behaviour ▸Most crashes are most likely due to mistakes or errors of generally compliant road users ▸‘Extreme behaviour’ crashes and ‘system failure’ crashes require countermeasures at an entirely

different level

▸Policy efforts to reduce extreme behaviour should be complemented with measures to improve the

entire system