human factors at level crossings

> Cale / Dinhobl / Grippenkoven – UN-ECE LC-expert group, human factors subgroup (23.10.2014, Geneva)

Human Factors at Level Crossings Results of questionaire

UN-ECE Group of Experts on Safety at Level Crossings Subgroup on Human factors:

M. Cale - CogniTo G. Dinhobl - ÖBB-Infrastruktur AG

J. Grippenkoven - German Aerospace Center e.V. (DLR)

Geneva, 23.10.2014


Content • Results of human factors - questionnaire

• Conclusion

• Next steps - establish a LC-model

• Next steps – LCAD (Level Crossing Appreciation Device)

• Outlook: future steps


Schedule & Milestones

1. Review of (existing) studies:

– Human Factors at level crossings

– Psychological models

2. Model(s) for level crossing application, including a list about derived assumptions and hypotheses about human (mis-)behaviour at level crossings

3. Human Factors solutions List of Tools existing tools and ‚theoretical‘ ones

4. Suggestion(s) for further work/research: towards proving that our tools work

Draft

paper

available

9/2014

1/2015

5/2015

5/2015 3


Results of human factors - questionnaire

• 24 feedbacks received

• from 22 countries

• not included here: issues on ‘survey of technology and technological solutions’

(see Section III-e of report of last meeting) – questions

• Question # 38: LC layout

• Question #39: railside protection equipment

• Question #44: ITS-solutions

• results by question:


0

5

10

15

20

25

Fa

ilu

re &

la

ck

of

ris

k a

wa

ren

es

s o

f

dri

ve

r/p

ed

es

tria

ns

La

ck

of

ca

re/d

istr

ac

tio

n

Ed

uc

ati

on

Le

ga

l

Su

n g

lare

/we

ath

er

co

nd

itio

n

Re

sp

on

sa

bil

ity

Fin

an

ce

Fa

ilu

re o

f

tec

hn

olo

gy

Sp

ee

d r

oa

d

Dru

nk

dri

vin

g

Mis

ca

lcu

lati

on

Fa

inti

ng

Q1: What are the three main causes behind level crossing accidents in your country?


Q2: Does your country have any solutions and/or creative and innovative countermeasures to solve these problems?

0

1

2

3

4

5

6

7

8

No

Awar

enes

s ca

mpai

gn

Rem

oval /

ove

r-&under

pass

Obst

acle

det

ectio

n dev

ices

Illusi

on at L

C

Pre

sence

of P

olice

Technic

al m

easu

res

300m

m S

TOP L

ine

2 m

bef

ore tr

ack

‚snak

e‘ p

ath fo

r ped

estria

ns

Com

plete

cove

ring b

y lo

ng bar

rier

s

Del

ay o

f clo

sure

of b

arrier

Des

ign r

ules

LED li

ghts

Act

ion p

lan

Rum

ble s

trip

s

Red

uce w

idth

and n

umber

of l

anes


Q3: Do you have any research studies or papers on human factors relating to the behaviour of road users around level crossings which you could share?

0

2

4

6

8

10

12

14

16

18

No

Som

e (b

ut not n

amed

)

ILLU

SIO

N O

N R

AIL

WAY T

RACK, 2

012

Pai

nting s

leep

ers

by ye

llow

Train

whis

tling

Hoar

dings

at L

C

Eric

Sch

oene,

TU D

resd

en

Rum

ble s

trip

s

E.C

.Will

ewort

h, 197

8

RSSB: I

mpro

ving b

ehav

iour

at L

C


Q4: Are there any educational programmes in your country that focus on the awareness of the road users concerning level crossings safety?

0

2

4

6

8

10

12

14

16

No

Gen

eral

awar

enes

s ca

mpai

gn

Sch

ools/c

hildre

n

"Sie

Bab

We"

(Gey

seV)

Strat

egy


Q5: Referring to the list of human factors at the start of this section E, has your country taken any action to improve safety at level crossings on the basis of these causative factors?

0

1

2

3

4

5

6

7

8

9

10

No

Gen

eral

awar

enes

s ca

mpai

gns

Vis

abili

ty im

prove

men

t

Wid

e an

gle L

ED

Contr

ol by

police

Ver

tical

sig

ning fo

r bet

ter at

tentio

n

Ela

stic

pla

tes

inst

ead o

f asp

halt

Rai

l sid

e pro

tect

ion s

yste

m (I

E)

Bar

rier

Soci

al a

dvertis

ing o

f the

rule

s


Q6: Do you have any other comments?

0

5

10

15

20

25

No

Educa

tion fo

r ro

ad d

rive

rs a

nd ped

est..

.

Exp

ert g

roup to

iden

tify

solu

tions

More

Res

earc

h, lik

e ill

usions

on LC

Cam

paign ‚S

top a

nd live

‘


Results of questionnaire - conclusion (1): What did we get?

• causes (Q1): – lack of risk awareness

– care(less)

– distraction

• solutions (Q2 & Q5): – awareness

– removal of LC

– enhance visability! (site-specific & wide-angle LED lights)

• research (Q3): only few existing

• LC-educational programmes (Q4): often included in general awareness campaigns

• general (Q6): education


Observations from the survey results

– Human factors as priority area in level crossing accidents identified by all

responses

– Tools and solutions often have a technological focus and do not focus on human factors

– Tools and solutions are often based on impressions of singular incidents and

• …Not theory driven

• …Not structured

• … but by „trial & error“-method

Theories and models from the field of human factors can be used in order to describe the process of crossing a LX


Next Steps of LC-human factors subgroup

• establish a LC-model (of human information procassing): the basics / foundation of a ‘toolbox’

• LC assessment device: the ‚toolbox‘


establish a LC-model - useful models

• Fast and Slow: Systems 1 and 2 (Kahnemann, 2011)

• 3-Levels of performance (Rasmussen, 1983)

• Model of information processing (Wickens & Hollands, 1999)


„Fast and Slow“: Systems 1 and 2 (Kahnemann, 2011)

System 1 – „Fast“

• Defining Characteristivs: – Automatic

– Effortless

– Unconscious

• No sense of voluntary control

• „Seeing and acting“

System 2 – „Slow“

• Defining Characteristics – Controlled mental process

– Effortful

– Deliberate and conscious

• With Control or self-awareness

• „Reasoning and acting“


The „Toolbox“ 1st draft

System 1 - Fast System 2 - Slow


„3-Levels of performance“

(Rasmussen, 1983)


System 1 – Fast

System 2 - Slow

Skill – based behavior

Rule – based behavior

Knowledge – based behavior



Information processing (Wickens & Hollands, 1999)


Attention

System 1 – Fast

System 2 - Slow

Skill – based behavior Rule – based behavior

Knowledge – based behavior

Sensory processing

Perception Cognition Motivation

Performance



LCAD

Level Crossing Appreciation Device

The birth of a human factors tool box

28

> Cale / Dinhobl / Grippenkoven – UN-ECE LC-expert group, human factors subgroup (23.10.2014, Geneva) 29

1. Attention

2. Perception

3. Cognition

4. Motivation

5. Performance

Five groups of functions

LCAD


1. Attention

Monitoring the road

Refraining from distracting activities (cell phones,

GPS, radio, food, drink etc.)

Influence of problematic factors (fatigue, alcohol,

emotional state etc.)

Factors competing for attention ( kids in car,

advertisements, other vehicles, radio program etc. )

LCAD


2. Perception

Perception of signs, lights, markings and barriers

Perception of distance, speed, changes in road etc.

Perception of the behavior of other road users

LCAD


3. Cognition

Realizing that one is approaching a LX and what is

expected

Recognizing dangers and limitations

Being able to predict occurrences and the behavior of

other road users

Availability of knowledge relating to level crossings

LCAD


4. Motivation

Understanding dangers entailed

Social pressure, current and/or local norms

Self esteem

Target risk

Exaggerated feeling of control

LCAD


5. Performance

Ability to control the vehicle over the rails

Choice of speed and distance from the curb

Being able to choose when to stop

Adaptation to road surface, visual clarity, opposing

traffic etc.

LCAD


Function ac1

ac2

ac3

ac4

ac5

Attention X X

Perception X X X

Cognition X

Motivation X X

Performance X

Accident Types

LCAD


LCAD

Accident Driver History and Environment LX in accident direction /

opposite direction

Date of accident Fatigue indicator How many additional accidents are known from the past

five years?

What is the average speed 500 m

before the lx ?

Exact position File of driver Fatalities / injured / damage / damage to LX What is the average speed 250 m

before?Season Does driver live

close to lx?

Is the setting urban or rural ? What is the average speed 150 m

before?

Time 00:00-06:00 07:00-

18:00 18:00-22:00

22:00-24:00

Age How many other accident were recorded during the past

five years in a 2km radius ?

What is the average speed 50 m

before?

Reason for trip Gender Estimate the number of vehicles passing lx daily How many meters before lx do you

recognize it clearly ?

Number of people in

vehicle

Years of

Education

What is the average speed in a 2km radius Are the required signs available ?

Estimated speed of

vehicle

Driver’s

explanation (if

available)

Estimate the number of accidents per vehicle in a 2km

radius

Are there additional or not standard

signs ?

Vehicle condition Estimate number of DUI per vehicle in a 2km radius How many signs are there in the 150 m

before the lx ?

Third party

involvement

What is the recommended speed at the LX ? How many signs are there in the 50 m

before the lx ?

Other Estimate enforcement/driven km in a 2km radius


Outlook: Future Steps of LC-human factors subgroup

Up to next meeting in January 2015

• model: to finish a model to identify and evaluate solutions for different LC-situations (accident investigation and LC planning) - as presented at spring meeting 2014

• LCAD: first steps with ‚Austrian accident investigation body‘ to get experience and first examples


Thank you for your attention & your (future) support!

human factors at level crossings

Documents