Safety

Touch screens even more dangerous than regular phones

A study carried out in The Netherlands has compared the behavior of cyclists when distracted in different ways. The study looked at using both conventional and touchscreen signs, texting, listening to music, talking on the phone, talking to another cyclist, and even playing a game on their phone. Each of these distractions was compared against

All forms of distraction caused the cyclists

This was a relatively small study with just 24 participants, recruited by word of mouth. Each person rode the section of public cycle path on their own bicycle, while being distracted in a number of ways.

The importance of this study is set in the context of previous research. This has shown that 17% of cyclists usually make phone calls and listen to music while riding. Other studies have shown cyclists are less aware of hazards. Importantly,

Abstract:

Although it has been shown that making phone calls or sending text messages while riding a bicycle can have a negative impact on bicyclist’s behaviour, in countries such as the Netherlands the operation of a mobile phone while cycling on a bicycle is not illegal and is actually quite common. In recent years conventional mobile phones with a physical keypad are increasingly being replaced by smartphones with a touch screen. The operation of a touch screen phone ironically cannot be done purely ‘by touch’ due to the lack of tactile feedback, and instead requires fixations on a relatively small screen. The question therefore can be asked whether the operation of touch screen telephones deteriorates cycling behaviour more than operation of a conventional mobile phone.

Twenty-four participants completed a track on their own bicycle while sending a text message from a conventional and a touch screen mobile phone. In addition the effects of other common activities that can accompany bicycling were studied, including texting at the same time as listening to music, talking on a mobile phone or cycling next to someone and speaking with this companion, and playing a game on a touch screen phone while bicycling. The impacts of all the above conditions on cycling performance and visual detection performance were compared with control conditions in which participants cycled with either one or two hands on the handlebars and were not required to perform any secondary tasks.

Bicycle speed was reduced in all telephone conditions and in the condition when cycling next to someone. Lateral position variation increased in all telephone conditions. Use of the touch screen led to a more central position in the cycle lane and resulted in worse visual detection performance compared with the operation of a conventional mobile phone. The main effect of listening to music was that an auditory signal to stop cycling was missed by 83% of the participants. In conclusion, while all investigated types of phone deteriorated cycling performance, the use of a touch phone has a larger negative effect on cycling performance than a conventional mobile phone. With touch screen smartphones taking the place of conventional mobile phones and being used for other purposes than verbal communication, these effects on cycling performance pose a threat to traffic safety.

Reference:

The effects of operating a touch screen smartphone and other common activities performed while bicycling on cycling behaviour

De Waard, Dick; Lewis-Evans, Ben; Jelijs, Bart; Tucha, Oliver; Brookhuis, Karel 2014

In: Transportation Research Part F: Traffic Psychology and Behaviour v 22

Read the full paper…

Cycling at night is dangerous because bikes are often not seen by motor vehicles. This study measured the distance at which approaching drivers noticed cyclists, and compared this with when the cyclists thought they would become visible. It showed that cyclists thought they had been seen well before the motorists actually saw them. The cyclists also had false ideas about which clothing would enhance their visibility. For example, they underestimated how much reflectors on their ankles and knees improved their visibility. They also falsely believed that a fluorescent vest without any retroreflective material would improve their nighttime visibility, in this case overestimating their visibility by a factor of seven.

Twenty five cyclists took part, half were regular cyclists and half occasional. They performed the tests at night, wearing black clothing and with a light on the handlebars. The tests were then repeated with the addition of a fluorescent vest, then a fluorescent vest with retroreflective tape and finally with the fluorescent retroreflective vest plus ankle and knee reflectors. Tests were also carried out with the light static, flashing and turned off. For each test, the cyclist cycled in place on a training stand at the end of a 300 meter section of straight and relatively flat road. A car approached the cyclist from the other end of the road. Stationary headlights were positioned close the cyclist and directed towards the oncoming car, to replicate the glare of an oncoming vehicle. As the car approached the cyclists pressed a button when they believed the driver would be able to see that they were a person. The distance was then recorded. This was compared with actual visibility distances obtained from a previous study.

It was found that when cyclist’s visibility was at its worst the cyclists greatly overestimated their visibility. However, when the most effective clothing was worn (a retroreflective vest with retroreflective bands on the limbs) cyclists underestimated their visibility. This may explain why cyclists rarely wear the most effective high visibility clothing. When cyclists wore a fluorescent vest with a static light they overestimated the distance at which oncoming drivers would first recognize them by a factor of seven.

Abstract:

Bicycling at night is more dangerous than in the daytime and poor conspicuity is likely to be a contributing factor. The use of reflective markings on a pedestrian’s major joints to facilitate the perception of biological motion has been shown to greatly enhance pedestrian conspicuity at night, but few corresponding data exist for bicyclists. Twelve younger and twelve older participants drove around a closed-road circuit at night and indicated when they first recognized a bicyclist who wore black clothing either alone, or together with a reflective bicycling vest, or a vest plus ankle and knee reflectors. The bicyclist pedalled in place on a bicycle that had either a static or flashing light, or no light on the handlebars. Bicyclist clothing significantly affected conspicuity; drivers responded to bicyclists wearing the vest plus ankle and knee reflectors at significantly longer distances than when the bicyclist wore the vest alone or black clothing without a vest. Older drivers responded to bicyclists less often and at shorter distances than younger drivers. The presence of a bicycle light, whether static or flashing, did not enhance the conspicuity of the bicyclist; this may result in bicyclists who use a bicycle light being overconfident of their own conspicuity at night. The implications of our findings are that ankle and knee markings are a simple and very effective approach for enhancing bicyclist conspicuity at night.

Reference:

“Bicyclists overestimate their own night-time conspicuity and underestimate the benefits of retroreflective markers on the moveable joints”
Wood, Joanne M. (School of Optometry and Vision Science, Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, QLD, Australia); Tyrrell, Richard A.; Marszalek, Ralph; Lacherez, Philippe; Carberry, Trent Source: Accident Analysis and Prevention, v 55, p 48-53, 2013
<a href=””>Read the full paper…</a>

 

A study has shown that motorcycles with a T-shaped arrangement of lights are seen and identified as a motorcycle more quickly. This may also be effective for bicycles. However, the improvement shown by this study was very small.

Method

The method used was to show people photographs, some of which contained motorbikes and see how quickly and accurately the people could identify the bikes. Some of the photo’s which contained motorbikes were photo edited so that different light configurations were simulated. In total three configurations of lights were used, a control with a normal headlight and two experimental configurations. The first experimental condition added lights at the fork and handlebars to create a T-shaped arrangement. The second experimental condition added helmet lights to the T-configuration. The photo’s which did not contain motorbikes showed other traffic scenes such as other vehicles and intersections. To gain a meaningful measure of how well drivers would identify a bike in actual driving, they were not asked to identify motorbikes. They were asked to spot the vehicles they would need to pay attention to if they were entering the traffic from a side road. The participants were not told that the focus of the study was motorbikes. The time they took to identify the bikes and their gaze behavior were both recorded.

Results

In over 99.5 percent of the photo’s shown to the participants, all of the vehicles were correctly identified. An analyses of variance (ANOVA) showed that there was a significant reduction in the time to identify motorbikes when the T-shaped lights were used. However, the benefits are very slight, especially when compared to the dramatic improvement in visibility when using retroreflective markers at the ankles and knees

 

Abstract:

One often discussed reason for motorcyclists’ high risk of accidents is that they are less conspicuous to other road users than are other motorised vehicles. The present paper aims to identify a front signal pattern created by additional light sources that would make motorcycles clearly and quickly distinguishable from other vehicles, and that would therefore facilitate localisation and identification of motorcycles. Results of a laboratory experiment have shown that motorcycles with a T-shaped light configuration are more quickly identified, particularly when the motorcycles are in visual competition with other motorised road users. Furthermore, analysis of gaze behavior showed that they were faster fixated by the subjects in the experiment, and the mean duration of fixations was shorter. The practical implications of this experiment and the need for further research are discussed.

Reference:

“Recognisability of different configurations of front lights on motorcycles”
Lars Rößger Katharina Hagen Jens Krzywinski Bernhard Schlag
Accident analysis and prevention. , 2012, Vol.44(1), p.82-87

Read the full paper…

Research carried out by Queensland University of Technology shows that reflective clothing including ankle and knee markers greatly increases the distance from which drivers can see cyclists. Surprisingly, the use of lights can actually reduce this distance, possibly because the lights glare the drivers.

Riding a bike at night is dangerous. Cyclists are often hit by a vehicle because the driver didn’t see them. This study set out to determine what effect the use of different types of reflective clothing would have on the visibility of cyclists to drivers. This was achieved using an experiment in which drivers traveled along a road towards a cyclist and indicated when they first saw them. The cyclists wore black clothing. The effectiveness of adding a reflective vest, and a reflective vest with the addition of ankle and knee reflectors was then tested. Variations of lighting were also included in the experiment. These included no light, a static light, and a flashing light.

24 drivers took part in the study. These were made up of a young and an old group. The young group was 6 men and 6 women between the ages of 18 and 35 years. The old group was 8 men and 4 women between the ages of 66 and 80 years.

The most effective high visibility clothing was a vest with the addition of ankle and knee reflectors. This greatly increased the distance at which cyclists were seen. The use of a light, however, did not improve the cyclists’ visibility to drivers. At one extreme, when only black clothing was worn and no lights were used, the older drivers didn’t see the cyclists at all in 73 percent of laps and the mean distance at which they were seen as only 5 meters. At the other extreme, when a vest, ankle and knee reflectors were worn, younger drivers never failed to see the cyclists, regardless of whether or not they used lights. They actually saw them from the greatest distance without lights, a mean distance of 223 meters.

Surprisingly, by far the greatest improvement in visibility was seen by adding the ankle and knee reflectors. Averaged over the driver age groups and lighting conditions, the cyclists dressed only in black were seen at 20 meters. With a vest this increased to 38 meters, but when ankle and knee reflectors were added the distance was 118 meters. It is thought that placing markers on the extremities makes people more visible due to our visual sensitivity to patterns of human motion, refered to as biomotion. This effect has been documented extensively for pedestrians..

Perhaps even more surprisingly, the use of lights did little to improve the cyclists’ visibility. When the angle and knee reflectors were worn the lights actually reduced visibility, possibly because the lights glare the drivers. These counterintuitive results may mean that cyclists believe they can be seen when they can not. This could lead to dangerous situations.

Abstract:

Bicycling at night is more dangerous than in the daytime and poor conspicuity is likely to be a contributing factor. The use of reflective markings on a pedestrian’s major joints to facilitate the perception of biological motion has been shown to greatly enhance pedestrian conspicuity at night, but few corresponding data exist for bicyclists. Twelve younger and twelve older participants drove around a closed-road circuit at night and indicated when they first recognized a bicyclist who wore black clothing either alone, or together with a reflective bicycling vest, or a vest plus ankle and knee reflectors. The bicyclist pedalled in place on a bicycle that had either a static or flashing light, or no light on the handlebars. Bicyclist clothing significantly affected conspicuity; drivers responded to bicyclists wearing the vest plus ankle and knee reflectors at significantly longer distances than when the bicyclist wore the vest alone or black clothing without a vest. Older drivers responded to bicyclists less often and at shorter distances than younger drivers. The presence of a bicycle light, whether static or flashing, did not enhance the conspicuity of the bicyclist; this may result in bicyclists who use a bicycle light being overconfident of their own conspicuity at night. The implications of our findings are that ankle and knee markings are a simple and very effective approach for enhancing bicyclist conspicuity at night.

Reference:

“Using reflective clothing to enhance the conspicuity of bicyclists at night“
Joanne M Wood, Richard A Tyrrell, Ralph Marszalek, Philippe Lacherez, Trent Carberry, Byoung Sun Chu
Accident analysis and prevention, 2012, Vol.45, p.726-730

Read the full paper…

The ability to stop quickly in response to a hazard is critical to safety. Modern bicycle disk brakes are generally capable of locking the wheels under most conditions. Stopping distance is therefore limited by the point at which wheels skid and by the deceleration at which the rider will pitch-over (go over the handlebars). Skidding occurs when the braking force exceeds the friction between the tires and the road. Better rubber compounds and wider tires with lower pressure will increase this friction. Road conditions such as water and oil on the road will also have a very significant effect on tire friction. The deceleration at which pitch-over occurs depends on the position of the combined center of gravity (CofG) for the cyclist and bike. If the CofG is close to the ground and far behind the front wheel then the stopping distance will be reduced. Under conditions where the stopping distance is limited by friction rather than pitch-over, the optimal distribution of force between the front and rear brake will produce the shortest stopping distance. Anti-lock brakes are a practical way to prevent skidding but electronic braking-pressure distribution system can achieve this optimal distribution of braking force.

In this open access paper, dynamic equations are presented which give the optimal distribution of braking force between the front and rear wheel. When braking momentum causes weight to shift from the rear wheel to the front wheel. The friction force for the rear wheel, therefore, reduces as the total braking force increases. At the point just before pitch-over occurs all of the braking force is due to the front wheel.

A simple way to increase the force on the rear wheel in relation to the front wheel is to use a smaller disk rotor on the rear wheel. To achieve optimal braking the equations presented in the paper can be used to control an electronic braking-pressure distribution system. This is most likely to be practically implemented in an e-bike which also has electronically controlled regenerative braking.

Abstract:

This paper presents an investigation on the braking performance and safety for a bicycle riding on a straight and an inclined paths. The equations of motion for a wheel model as well as the model for ideal synchronous braking are derived to acquire the shortest braking distance and improve the riding stability. The optimal design of the bicycle braking is obtained based on the simulation results with various bicycle geometries, ratios of brake force, and road friction.

Reference:

“Synchronous brake analysis for a bicycle”
David Lie Cheng-Kuo Sung
Mechanism and machine theory, 2010, Vol.45(4), p.543-554

Read the open access paper…

A study carried out in Hungary has shown the importance of road quality for childrens’ safety when cycling. The research was carried out retrospectively by reviewing hospital records and then sending a questionnaire to the parents of injured children.

The researchers focus on the importance of roads and helmets. This makes sense for those involved in directing government initiatives. For those of us interested in bicycle design, it’s also interesting that about 10 percent of injuries were caused by spokes. It is not specified which of these were falls in which impingement of the spokes was the cause and which were injuries purely due to a body part being entrapped by the spokes. In both cases, however, a design which prevents body parts and foreign objects from entering between the spokes would have avoided the injury.

Similarly, 3.6 percent of injuries were caused by the handlebar. It is implied that these were the result of the child landing on the handlebar which then caused a penetrating wound. Similar impacts with the saddle and seat post are also mentioned with all other parts of the bicycle accounting for 2.2% of injuries. Good design could prevent these types of injury.

Abstract:

Introduction:

The purpose of this study was to investigate the characteristics and the outcome of bicycle injuries in paediatric patients according to the living environment, and to create guidelines for injury prevention.

Patients:

The evaluation was performed in part based on hospital database of 1803 in- and out-patient children treated at the Paediatric Surgical Department of Pe´cs/Hungary between 2000 and 2006, and at the Department of Paediatric Surgery at the Heim Pal Hospital Budapest between 2004 and 2006. Additionally questionnaires were mailed to the patients’ families to gain follow-up information. We analysed three groups according to demographic density (village, midsize town and large town).

Results:

We found, that poor road quality played an important role as a contributing factor of injuries in villages. The number of bicycle spoke-injuries was higher in villages (13%), than in midsize towns (4.6%) and the large town (9.9%). In villages, 5% of children injured wore a helmet; this rate was 9% in midsize towns and 9.1% in the large town. Head injury was more common in villages, while in midsize towns and the large town arm injuries proved to be predominant.

Discussion:

Prevention strategies targeting bicycle injuries in children should take into account the population density. This analysis revealed a substantial difference in the use of safety devices, and in the characteristics of injuries occurring in villages, indicating that there is a need for special attention regarding this higher risk population.

Reference:

Bicycle injuries in children: An analysis based on demographic density
Katalin Kiss Zsuzsanna Pótó András Pintér Sándor Sárközy
Accident analysis and prevention. , 2010, Vol.42(6), p.1566-1569

Read the full paper…

Going ‘over the handlebars’ is a common type of crash for cyclists. The high center of gravity and short wheelbase makes bicycles susceptible to this type of pitch-over if the rate deceleration is too high. Common causes are braking too hard with the front brake, hitting an obstacle or something getting stuck in the front wheel. This study observed all each of these types of events under controlled conditions in order to better understand the dynamics. A trained stunt person was instructed to ride a bicycle at approximately 20 kph (12.5 mph) towards a pitch-over initiation area. He would then either brake hard, ride into a fixed obstacle or trigger a dowel pin to be fired through the front wheel. The resulting pitch-over crash was filmed using high-speed photography.

.Targets placed on the cyclist and test track allowed the motion of the cyclist to be tracked and analyzed. It was found that crashes resulted in the cyclist landing on his head with a position which loaded the cervical spine. This is dangerous crash configuration. Previous research has shown that 6 percent of all bicycle accidents are caused by something being caught in the spokes. Preventing this cause as well as improving the longitudinal stability of the bicycles are therefore important safety factors.

Abstract:

In a controlled environment, bicycle crash tests were performed to analyze the dynamics of bicycle pitch-over, and the effect different test conditions would have on the motion of the bicyclist and the potential for serious head and neck injuries. The tests being evaluated within this paper will focus on a single size rider and bicycle with speed controlled between 19.3 and 22.5 kph (12.0 and 14.0 mph). The variables of the test conditions were the method of pitch-over initiation and configuration of the bicycle. Pitch-over was initiated by applying the front brake, impacting a rigid barrier, and inserting a dowel into the spokes of the front wheel. The bicycle configuration was varied by changing the frame, front fork, front wheel and stem. For all testing performed the event was captured on high speed video. As the testing progressed a force plate was added to measure the vertical loads and an impact plate was added to the barrier impacts to measure the horizontal impact loads. The data acquired from this series of tests provides insight into the dynamics of bicycle pitch-over, injury mechanisms, and the effect of bicycle component failure.

Reference:

Analysis of bicycle pitch-over in a controlled environment
Bretting, Gerald P. (Collision and Injury Dynamics Inc, United States); Jansen, Henricus P.; Callahan, Michael; Bogler, John; Prunckle, John Source: SAE International Journal of Passenger Cars – Mechanical Systems, v 3, n 1, p 57-71, August 2010

Read the full paper…

A study carried out by researchers in the USA has identified a number of factors which are strongly associated with cyclists receiving severe injuries in collisions. The most dangerous factor is motor vehicle speed. When cyclists collide with vehicles traveling at over 50 mph (80 kph) they are more than 16 times more likely to die from their injuries, when the speed is over 30 mph (48 kph) the risk is doubled. When the vehicle’s speed is less than 20 mph (32 kph) there is a greatly reduced risk of harm to the cyclist. This clearly shows the importance of avoiding high-speed traffic when cycling. Other factors which more than double the risk of death include poor weather, dark streets, morning rush hour, head-on collisions, a truck being involved, alcohol consumption by either the driver or the cyclist and the cyclist being over 55 years old.

It is also interesting to note that when the cyclist is at fault they are more likely to be severely injured than when the driver is at fault.

Abstract:

This research explores the factors contributing to the injury severity of bicyclists in bicycle-motor vehicle accidents using a multinomial logit model. The model predicts the probability of four injury severity outcomes: fatal, incapacitating, non-incapacitating, and possible or no injury. The analysis is based on police-reported accident data between 1997 and 2002 from North Carolina, USA. The results show several factors which more than double the probability of a bicyclist suffering a fatal injury in an accident, all other things being kept constant. Notably, inclement weather, darkness with no streetlights, a.m. peak (06:00 a.m. to 09:59 a.m.), head-on collision, speeding-involved, vehicle speeds above 48.3 km/h (30 mph), truck involved, intoxicated driver, bicyclist age 55 or over, and intoxicated bicyclist. The largest effect is caused when estimated vehicle speed prior to impact is greater than 80.5 km/h (50 mph), where the probability of fatal injury increases more than 16-fold. Speed also shows a threshold effect at 32.2 km/h (20 mph), which supports the commonly used 30 km/h speed limit in residential neighborhoods. The results also imply that bicyclist fault is more closely correlated with greater bicyclist injury severity than driver fault.

Reference:

Bicyclist injury severities in bicycle-motor vehicle accidents
Kim, Joon-Ki (Washington University in St. Louis, Department of Civil Engineering, One Brookings Drive, St. Louis, MO 63130-4899, United States); Kim, Sungyop; Ulfarsson, Gudmundur F.; Porrello, Luis A. Source: Accident Analysis and Prevention, v 39, n 2, p 238-251, March 2007

Read the full paper…