Pay As You Speed, ISA with Incentives for Not Speeding: a Case of Test Driver Recruitment

Accident Analysis and Prevention 48 (2012) 10–16

Contents lists available at ScienceDirect

Accident Analysis and Prevention

jo urnal homepage: www.elsevier.com/locate/aap

Pay as You Speed, ISA with incentives for not speeding: A case of test driver recruitment

a,∗ a a a a b

Harry Lahrmann , Niels Agerholm , Nerius Tradisauskas , Teresa Næss , Jens Juhl , Lisbeth Harms

Aalborg University, Department of Development and Planning, Fibigerstraede 11, DK-9220 Aalborg, Denmark

Copenhagen University, Department of Psychology, Oester Farimagsgade 22, 5A, DK-1353 Copenhagen K, Denmark

a r t i c l e i n f o

a b s t r a c t

Article history: The Intelligent Speed Adaptation (ISA) project we describe in this article is based on Pay as You Drive

Received 10 September 2009

principles. These principles assume that the ISA equipment informs a driver of the speed limit, warns the

Received in revised form 13 March 2011

driver when speeding and calculates penalty points. Each penalty point entails the reduction of a 30%

Accepted 21 March 2011

discount on the driver’s car insurance premium, which therefore produced the name, Pay as You Speed.

The ISA equipment consists of a GPS-based On Board Unit with a mobile phone connection to a web server.

Keywords:

The project was planned for a three-year test period with 300 young car drivers, but it never succeeded

Intelligent Speed Adaptation

in recruiting that number of drivers. After several design changes, the project eventually went forward

Field trial

Speeding with 153 test drivers of all ages. This number represents approximately one thousandth of all car owners

in the proving ground of North Jutland in Denmark. Furthermore the project was terminated before its

Economic incentives

Trafﬁc safety scheduled closing date. This article describes the project with an emphasis on recruitment efforts and the

project’s progress. We include a discussion of possible explanations for the failure to recruit volunteers

for the project and reﬂect upon the general barriers to using ISA with ordinary drivers.

1. Introduction in the car. ISA can also alert the driver if the car’s speed exceeds

the current speed limit. The speed alert function may assume dif-

1.1. Intelligent Speed Adaptation ferent forms. The most common form is a visual display combined

with an acoustic warning. Researchers have also tested resistance

Recent developments in Intelligent Transport Systems (ITS) in the accelerator that prohibits the driver from increasing speed

have demonstrated that Advanced Driver Assistance Systems (Almquist, 2006).

(ADAS) may play an important role in future improvements in

trafﬁc safety. Over the past two decades, several studies have con-

1.2. A short history of ISA studies

ducted tests of a speciﬁc type of ADAS, Intelligent Speed Adaptation

(ISA). These studies have demonstrated the great safety potential of

The Department of Trafﬁc Technology at the University of Lund

ISA. The estimated net effect of a full ISA implementation is a reduc-

in Sweden conducted the ﬁrst ISA ﬁeld study in 1992. The study

tion of up to 50% in the number of road trafﬁc accidents resulting in

included 75 test drivers, a test car with an active accelerator pedal

fatalities or injuries (Carsten and Tate, 2005; Marchau et al., 2005;

and a ﬁxed test route. It importantly demonstrated that test drivers

Regan et al., 2006; Vlassenroot et al., 2007). Moreover, some schol-

held more positive attitudes towards the equipment after the test

ars also have suggested that ISA may reduce fuel consumption and

drive (Persson et al., 1993). This study was followed by a number

thereby, automobile pollution (Servin et al., 2006).

of ISA projects in several different countries. In Sweden, larger ﬁeld

ISA is not a speciﬁc technology. It belongs to a group of ADAS

trials were conducted between 1999 and 2001. These Swedish ﬁeld

technologies best suited for keeping track of the speed and position

trials tested different types of ISA equipment and included approx-

of a vehicle travelling on a segment of road. By comparing these reg-

imately 5000 drivers (Warner, 2006; Hjälmdahl, 2004; Várhelyi

istrations with the speed limit on various road segments, ISA can

et al., 2004). Studies have also been conducted in the UK, the

inform a driver of the relevant speed limit by means of a display

Netherlands, Belgium, France, Australia and Finland (Jamson et al.,

2006; Regan et al., 2006; Peltola et al., 2004). The ﬁrst Danish ﬁeld

∗ study was completed in 2001. In this “INFATI” project (INFATI is

Corresponding author. Tel.: +45 4056 0375; fax: +45 9815 3537.

an abbreviation for the projects Danish name: Intelligent FArtTIl-

E-mail addresses: [email protected] (H. Lahrmann), [email protected]

pasning), the system continually presented the speed limit on a

(N. Agerholm), [email protected] (N. Tradisauskas), teresa.naess@oﬁr.dk (T. Næss),

[email protected] (J. Juhl), [email protected] (L. Harms). console-mounted display in the car and warned the driver by a

H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16 11

voice message when speeding occurred. The project included 24

test drivers over a six-week period. The results found were con-

vincing. Driving speed, measured at the 85th percentile, decreased

by 5–6 km/h. In contrast to many other ISA experiments, INFATI

also covered speed limits in rural areas, and thereby demonstrat-

ing that speed reductions obtained by using this equipment were

signiﬁcant on rural roads (which in Denmark generally have an

80 km/h speed limit). Furthermore, the use of a voice message for

speed warnings proved efﬁcient and was greatly appreciated by

the test subjects (Boroch, 2002; Lahrmann et al., 2001, 2004). As

of this time, ISA projects have been conducted in several countries

with positive results regarding both the speed-reducing effects of

ISA and the test drivers’ acceptance of ISA equipment. The Swedish

studies reported that test drivers thought themselves better drivers

with ISA, and two out of three test drivers stated they would keep

the system in their car after the test period if the system were free.

Moreover, one out of three test drivers were willing to pay a small

amount for ISA (Biding and Lind, 2002).

Economic incentives have been absent in most previous ISA tri- Fig. 1. The ISA display mounted on the dashboard.

als. In some studies, participants had an ISA-equipped car at their

disposal during the test period, which could be considered a reward

rienced young car owners pay higher car insurance premiums than

for participation (Jamson et al., 2006). Only in the Dutch Belonitor

do older, more experienced drivers. Therefore, young car owners

project and a minor Swedish project were participants rewarded

would proﬁt more from a discount on their car insurance premi-

for not exceeding the speed limit (Belonitor, 2009; Hultkrantz and

ums than would the average motorist. The overall purpose of the

Lindberg, 2009).

PAYS project was to investigate the extent to which a discount on

Many ISA projects make the fundamental assumption that

a relatively high insurance premium combined with an informa-

motorists intend to comply with the legal speed limit. Speeding

tive and advisory ISA system that records behavior would motivate

occurs either when drivers fail to notice all speed signposts or when

young, inexperienced motorists to install ISA equipment in their

they are inattentive to their own speed. Test drivers state that they

cars. In addition to this aim, we also wanted to study how such a

are pleased to have ISA. Despite these ﬁndings, which have been

system would affect the speed behavior and the accident risk of a

accumulated through two decades of ISA research, there has been

young driving population. The aim of the project, therefore, was

no breakthrough of ISA in road trafﬁc. Indeed, the risk of speed-

to test the effect of ISA on both speeding and accidents for 300

ing in a modern, quiet and comfortable car is obvious. The general

young car owners participating in the PAYS project over a three

purpose of ISA, therefore, is to support drivers in maintaining the

year period.

proper speed under the implicit assumption that a driver, if pro-

PAYS was both a large-scale ﬁeld test and a research project that

vided with sufﬁcient information, will not exceed the speed limit.

included four sub-projects.

In fact, this hypothesis has strong support from the results of pre-

vious ISA studies conducted in ordinary road trafﬁc. But why is it

(1) Development of equipment designed for an informative, advi-

that ISA equipment has not been widely introduced? Why have pri-

sory and recording ISA system.

vate investors not rushed into this lucrative market, which includes

(2) Development of digital speed limit maps for the test area cov-

more than 200 million cars in the EU alone? Have researchers

ering the entire 22,000 km road network of the county of North

focused too much on the positive outcomes obtained and too lit-

Jutland, Denmark.

tle on the general acceptance of ISA? Obviously, there has been no

(3) Recruitment of test drivers from the relevant age group.

voluntary introduction of ISA. Perhaps such an introduction suffers

(4) A three-year test period involving 300 young test drivers.

from the well-known obstacle that “those who want it do not need

it, and those who need it do not want it” (Jamson, 2006).

This article focuses on the recruitment of participants for the

ﬁeld trial, which was planned as a simulation of a market intro-

duction of ISA. Other results from the ﬁeld trial have been reported

2. An ISA project based on Pay As You Speed principles

elsewhere (Agerholm et al., 2007; Harms et al., 2008; Tradisauskas

et al., 2009; Lahrmann et al., 2011).

2.1. Project design

On the basis of these two decades of ISA experience, the present 2.2. An informative and recording ISA equipment

ISA project focused on simulating a market introduction of ISA

equipment. The project targeted “those who need it” by address- The project developed an ISA device that performed both

ing car owners aged 18–24 as participants. Car owners in that age informational and recording functions. Technically, the equipment

group were offered a ﬁnancial reward for their participation in the consisted of an On Board Unit (OBU) consisting of three compo-

project, called Pay as You Speed (PAYS). In short, the Pay as You nents. (1) A GPS/GPRS unit with a memory card on which a digital

Speed principle establishes a connection between ﬁnancial reward speed limit map and log data were stored. This unit was placed

for participation and the recorded behavior of the vehicle in terms under the dashboard. (2) A display device, including a loudspeaker,

of speed. mounted in the air nozzle. (3) A GPS antenna located behind the

The target group included a large number of inexperienced rear-view mirror. Fig. 1 shows the display as mounted to the dash-

drivers who are over-represented in trafﬁc accident statistics. In board. Fig. 2 shows a close-up of the display. The OBU in the vehicle

fact, the risk pertaining to that particular age group is up to was connected to a web server through GPRS.

10 times greater than that of their parents’ generation (Brems The system works as follows: the OBU receives a GPS position

and Munch, 2008). Due to their increased accident risk, inexpe- once a second. The OBU calculates the car’s position on the digital

12 H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16

changing during the course of the project. In the context of the

present project and based on the number of complaints received,

our judgment is that the consequences of this situation were lim-

ited. Finally, the OBU had a tracking function. If a car was stolen, the

owner could call a hotline, which would send an SMS to the OBU,

and the OBU would return the position of the car.

The digital speed limit map covers every road in the county of

North Jutland. In the rest of Denmark, only roads with a speed limit

of 90 km/h or higher are covered. Outside these roads, the OBU was

set to react if a driver exceeded the general 80 km/h speed limit on

rural roads. Thus, the speed map covered all the roads in Denmark

except for those in urban areas outside North Jutland. Because

speed limit data are not available for other countries, the system

showed a blank display and did not react to speeding outside of

Denmark’s borders (Sonne, 2005; Juhl et al., 2007).

Fig. 2. The display with the current speed limit and the total number of calculated

penalty points (left) and penalty points for the current trip (right). 3. The ﬁeld trial

3.1. Recruitment of participants for the ﬁeld trial

map (“map matching”) and shows the current speed limit on the

display. It also compares the speed limit with the actual speed of

We established the conditions for the ﬁeld trial in co-operation

the vehicle, and if the vehicle exceeds the speed limit by more than

with a single insurance company, Topdanmark. According to Top-

5 km/h, the OBU emits an initial verbal warning, for example “50”.

danmark’s price list, the premium for an all-risk car insurance for

The warning is repeated every 6 s until the car’s speed drops below

a young car owner with a typical car is approximately D 2400 per

the speed limit +5 km/h. The tolerance of 5 km/h allows the partic-

year. Prior to setting the conditions for participation in the PAYS

ipants to drive at speeds slightly above a given speed limit without

project, Topdanmark conducted a market study among their own

receiving warnings. This tolerance is supported by the ﬁning prac-

customers in the target group to assess interest in the project and

tices of law enforcement ofﬁcials (Copenhagen Police, 2010). After

to identify possible obstacles to participation. The overall results

the second warning, the system will add the sentence “you are driv-

of the study indicated that approximately 16% of the customers in

ing too fast” to the spoken speed limit and will calculate and display

the target group would be very interested and 50% might be inter-

“penalty points” every 6 s. The number of penalty points per warn-

ested in participating in the PAYS project. The study also established

ing is related to the level of excessive speeding, as shown in Table 1.

that both the participant fee and the fact that speeding would be

If the speed is above the speed limit +5 km/h and below the speed

recorded were deciding factors for non-participation (Topdanmark,

limit +20%, the system will impose 1 penalty point every 6 s. In the

n.d.).

interval between the speed limit +20% and the speed limit +30%,

Following the market survey, we decided to offer participants a

the system will impose 2 penalty points every 6 s, and so on. When

discount of up to 30% on their car insurance premiums as a partic-

the car’s speed drops below the speed limit +5 km/h, the warnings

ipation bonus. The bonus would be paid in arrears every 6 months,

cease. This sequence occurs any time the vehicle exceeds the speed

and the participants would have D 0.07 deducted from that bonus

limit.

for each registered penalty point during that period (see the previ-

For every map-matched position, the system also calculates a

ous description of the penalty points system). In response to limited

map-matching quality. If the parameter-controlled estimate indi-

funding for the project and to bring the project closer to simulating

cates that the map-matching quality is too low, the best guess at

market conditions (and despite the results of the market study), we

a speed limit is shown in brackets on the display, and the sys-

decided that the participants would initially pay a participant fee of

tem does not react to speeding. A high-performance map-matching

700, corresponding to the potential monetary gain from the ﬁrst

algorithm has been developed for the project (Tradisauskas et al.,

year of participation, provided the participant did not speed. The

2009).

participants were offered an interest-free loan at the local bank,

When speeding is registered during a trip, the OBU uploads a

and they were guaranteed to receive, at a minimum, the fee as a

log ﬁle to the web server immediately after the conclusion of the

bonus, regardless of their speeding behavior, if they remained with

trip. After a trip, it is possible for the driver to log on to a personal

the project for the full three-year period. The ﬁnal condition for

web page and check where and why penalty points were imposed

participation was to purchase car insurance from the co-operating

(see Fig. 3 for an example). The purpose of the web page is to allow

insurance company, Topdanmark.

drivers to check, and possibly correct, the record. Originally, the

These conditions were applied when recruitment started in

plan was to make it possible to upload map updates to the OBU;

September 2005. A recruitment letter was sent to 6000 young car

however, this feature failed to work satisfactorily. This situation

owners between 18 and 24 years of age in North Jutland. A web

implies that the speed limit map and the actual speed limits were

page was opened where participants could obtain further informa-

tion about the project and express their interest in participating

Table 1

in the test trial. The start of the recruiting drive received substan-

The relationship between speeding and the number of penalty points calculated per

tial media coverage. The result, however, was highly disappointing.

6-s interval of excessive speed.

Only 40 young car owners expressed their interest.

≤

X (H + 5 km/h) awards 0 penalty points H = Speed limit

A more comprehensive market study, including the entire tar-

(H + 5 km/h) < X < (H + 20%) awards 1 penalty point X = Current speed

≤ get group in the county rather than Topdanmark customers alone,

(H + 20%) X < (H + 30%) awards 2 penalty points % = Speeding in percentage

≤ was conducted in November 2005. Again, the target group viewed

(H + 30%) X < (H + 40%) awards 3 penalty points

≤

(H + 40%) X < (H + 50%) awards 4 penalty points the project favorably, as a total of 91% thought well or very well

≤

(H + 50%) X < (H + 60%) awards 5 penalty points

of the project. Again, the main argument against participation was

Etc.

the fee and, to a minor degree, the fact that speeding would be

H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16 13

Fig. 3. Screen dump from the web server showing details of penalty points from a trip. The color of the roads indicates the speed limit and the black arrows indicate the

position and direction of the car for each penalty point. For each of the four arrows, the table indicates the current speed limit, the vehicle speed, the number of assigned

penalty points, the date and time of the speeding incident, the speeding number and the trip number.

recorded. This study revealed, in addition, that members of the pation was also raised to 28 years. Letters were sent to a total of

target group were paying much less than D 2400 for their car insur- 11,400 car owners in North Jutland County; the result was that 180

ance. In fact, approximately 50% of car owners in the group did not persons expressed their interest on the project’s web site.

have an all-risk but only third-party insurance. Moreover, many In May 2006, Topdanmark phoned 1000 of their customers to

customers did not pay the full price for their car insurance because recruit more participants, but this effort yielded only a few more

insurance companies offer several discount options. We estimated participants.

that, with the insurance rate of the target group averaging D 1200, To conﬁrm their involvement, participants were required to sign

it would take not one, but two years of bonuses to acquire a sum a contract with the project. During the summer of 2006, we sent a

equal to the participation fee. Another obstacle to participation was total of 230 contracts to the car owners who had expressed interest

that participants insured by other companies would have to switch via the web site. The result was 50 signed contracts. The remain-

to Topdanmark to participate. The result was that few in either the ing 180 drivers had changed their minds and no longer wished to

broader target group or the target group among the customers of participate in the project. We did not perform a systematic dropout

Topdanmark signed up to participate in the PAYS project. The con- analysis, but phone calls to 19 of the dropouts strongly suggested

clusion was that PAYS would have been a good project if it had that they ultimately did not want to give up their freedom to drive

not been for the participant fee (Markman Customer Management, faster than the speed limit (Agerholm, n.d.). Out of the 11,400 per-

n.d.). sons in the general target group, we were successful in signing

Due to the results of this market study, we were successful in contracts with 50 for a 0.5% participation rate. As for Topdanmark

obtaining additional funding to cover the participant fee, which customers, 3% of the target group decided to participate.

could then be dropped. A new round of recruitment offering free The goal of the PAYS project was to involve 300 young car own-

participation started in February 2006. The age limit for partici- ers. After much time and effort, and with only 50 contracts signed,

14 H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16

we abandoned that goal. Instead, we decided once again to change the remaining ﬁve declared themselves in favor of the idea, but in

the conditions for participation. In the next recruitment round in the end, found various excuses for not participating.

September 2006, the age limit for participation was completely After the test-drive period, we interviewed three of politicians

lifted, making the target group all motorists in the county of North who had the system installed in their cars and the one who declined

Jutland. We marketed the project through newspaper advertise- from the beginning to participate in the test. We asked them ques-

ments and general press coverage. This round yielded another 30 tions about trafﬁc safety, speed, their experiences using the system

interested car drivers. and their opinions on ISA in general and PAYS in particular. If the

During the spring of 2007, 1500 Topdanmark customers were politicians held opinions favorable to ISA, they might act as ambas-

contacted by phone, which yielded a few more signed contracts. sadors for the system to the public and to national politicians. In the

Approximately 125 signed contracts were in hand by the summer next section, we report and discuss statements from these inter-

of 2007. views.

The last recruitment round was conducted during the autumn

of 2007. A recruitment letter was sent to 2000 Topdanmark cus-

tomers and a recruitment e-mail was sent to approximately 2500 3.3.1. Interviews

employees of Aalborg University. We ended the recruitment drive As an initial question, the four politicians were asked how they

in December 2007 with a total of 153 signed contracts, equal to one would characterize themselves as drivers. They all began by claim-

in a thousand car owners in the county of North Jutland. ing to abide by trafﬁc rules. However, the three politicians who

participated in the test drive made several statements proving that

politicians are only human:

3.2. Early termination of the PAYS ﬁeld trial

“With this device in my car I found out that, actually, I do drive

We did not reach the goal of 300 participants, but the ﬁeld trial too fast. When I am in a hurry or when I start my trip 15 minutes

proceeded with the 153 participants. We installed the equipment in too late, I would like to be able to exceed the speed limit. And this

cars starting with a test group in the summer of 2006. The ﬁrst ordi- is exactly the situation where this voice becomes very annoy-

nary participants had the equipment mounted during the autumn ing.” (Politician A)

of 2006, and the last group had the equipment installed at the end

”This is one of the reasons why I ﬁnd this system so useful. I used

of December 2007. Over such a long period of time, some partici-

to be a “quick” driver. Perhaps I was not exactly aggressive, but

pants dropped out for various reasons, and by the end of 2008, only

I was one of those drivers who always leave home a little too

105 systems were in operation. The original rationale for a large

late. I would excuse myself thinking ‘I have to go fast because I

study involving 300 young car owners in a ﬁeld trial for a period

have to be in time for my meeting.’ I do not do that anymore.”

of three years was to obtain sufﬁcient data to enable an analysis of

(Politician D)

the safety-related effects of ISA by examining the number of traf-

ﬁc accidents among the participants. With only 153 participants, ”I use my car a lot. I drive reasonably. But when I had the ISA

. . .

however, the project would probably not generate statistically con- equipment in my car, I was constantly reminded that I was

clusive results on the safety-related effects of ISA. As with the other exceeding the speed limits.” (Politician C)

ISA projects, we would be able to estimate the safety-related effects

These local politicians had various opinions on the different fea-

of the equipment from changes observed in drivers speed behav-

tures of the system. To those who wanted to decide for themselves

ior. But it was not necessary to require participants to drive with

whether to exceed speed limits, the PAYS system was annoying.

the equipment for three years to obtain sufﬁcient speed data to

. . .

describe changes in speed behavior. Therefore, in the autumn of Politician A said: “I do not want this voice in my car at all

2008, we decided to stop the project before the scheduled end. I want to be able to exceed the speed limit outside the cities.

Instead of allowing all participants to drive for three years, we You don’t want to be the one delaying everyone because you

decided to turn off the website and the displays of the 105 remain- abide by all the rules. You have to keep moving. Precisely in

ing participants on November 1, 2008. The next 45 days were then that respect the ‘Pay As You Speed’ system is very annoying. In

used as an after period during which no warnings or penalty points many situations it is better to step on the accelerator and clear

were awarded, but speeding was still logged. We uninstalled the the spot quickly”.

OBU units in January and February of 2009, and on March 1, 2009,

Asked whether PAYS would be a good idea for others, the only

the ﬁeld trial was closed, and data processing began.

useful situation this politician identiﬁed are situations in which

young people are learning to drive or have just obtained their

3.3. Local politicians as ambassadors driver’s licenses.

Politician B – the one who had not tried PAYS – stated that he did

In the midst of the recruiting process described above, we not want the system in his car. However, he conceded that perhaps

decided to involve local politicians in the project. We wanted them “something” should be put in the cars. He was not optimistic that

to be ambassadors in their municipalities and advocate partici- young people could be persuaded to use PAYS voluntarily. If ISA

pation in the project. We also wanted to discover whether local were to be implemented, this politician held the view that every-

politicians could be relied upon to play an important role in imple- one should use it so that no particular group of drivers would feel

menting ISA in general. Are local politicians aware of the new “punished or exposed”.

technical possibilities ISA represents? Do they see ISA as a realistic The third politician (C) was in favor of using a system like PAYS

option, ready for implementation? As with other people, politicians to help drivers, especially young men, reduce their speed. He found

form opinions under the inﬂuence of personal experience. Without the system very helpful and had changed his behavior in trafﬁc. He

having driven with ISA, it is difﬁcult to imagine that experience now uses the warning feature in his own navigation system to lower

and how ISA affects one’s behavior in trafﬁc. We chose 10 local his speed, and he has formed the habit of responding to the warn-

politicians responsible for trafﬁc safety, one from each of the 10 ings. This politician thought the time was right for such a system

municipalities in the county of North Jutland. We asked them to and spoke of the money spent in the healthcare system on victims

have the PAYS equipment installed in their private cars and to use of trafﬁc accidents. “If we put together this money, we could spend

it for at least six months. Four of the ten accepted, one declined and it on something like this. That could work. Prevention. . . Politicians

H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16 15

should regulate and legislate in a way which shows consideration for projected as possible ambassadors for the project and who would

the environment as well as speed and road safety.” have ISA equipment installed in their private cars, only four partic-

The fourth politician (D) was very enthusiastic about PAYS. He ipated. Interviews suggested that even politicians with a positive

had learned to calculate his travel time more precisely. The stress he attitude to ISA did not see a role for themselves in implementing

had felt disappeared, and it became almost unpleasant to exceed ISA. With regard to speeding, one of the politicians made very clear

the speed limit. As for the chances of implementing PAYS on the her resistance to ISA and her desire to retain the ability to speed.

national level, this politician said that the time will never be right; Taken as a whole, the reluctance of young and experienced

politicians will have to make the necessary decisions. In his opin- motorists to participate in this project and the corresponding lack

ion, most people would not voluntarily choose a system like PAYS. of enthusiasm on the part of politicians with local responsibility

Therefore, politicians should decide whether the number of trafﬁc for trafﬁc safety suggests that public opinion in Denmark considers

fatalities is high enough to necessitate a decision. speeding to be permissible. This attitude to speeding is probably the

greatest obstacle to the general implementation of ISA in Denmark.

Unfortunately, the main conclusion to be drawn from this

3.3.2. Political perspectives on ISA

project – the failure to recruit participants – is apparently inconsis-

In conclusion, two of the local politicians were quite positive

tent with the results of most other ISA projects. Positive results from

towards PAYS. The third politician who participated in the test and

using ISA have generally been reported, and no problems related to

the politician who did not try the system were rather negative and

the recruitment of participants to ISA projects have been reported

did not think that local political leaders could play an important

or discussed in any previous studies. The intent of the PAYS project

role in promoting ISA. All of these leaders believed that national

was to simulate a market situation, whereas the previous ISA stud-

politicians would need to be involved to implement such a system.

ies have included one or more of the following conditions: (1) ISA

They did not mention political leaders at the EU level. The intervie-

coverage of only a limited number of roads; (2) the provision of a

wees did mention national legislation as the most efﬁcient means

project car, free of charge; (3) only a few young drivers participat-

of promoting ISA.

ing; and (4) a short test period. All of these factors tend to reduce the

When speaking as politicians, local political leaders grasped the

general validity and reliability of the test results obtained. The fact

problem of excessive speeding and desired reduced speed. When

that 20 years of positive results from ISA research have not stimu-

speaking as drivers, however, they admitted to being inclined to

lated the commercial introduction of ISA-equipped cars or OBUs on

break speed limits. Lack of experience with Intelligent Transport

the ITS aftermarket (except in Sweden, where it is possible to buy

Systems in general, aversion to using ISA themselves and the fact

a ﬂeet management system with an ISA function [www.sepab.se])

that they do not see municipalities as participants in the imple-

leaves the impression that there is no demand for ISA on the mar-

mentation of ITS or ISA means that local politicians are not likely to

ket either in Denmark or elsewhere. Because ISA is a promising and

make the ﬁrst move towards testing, promoting or implementing

technically sound safety system, introducing ISA in road trafﬁc will

ISA in their local communities.

require action beyond waiting for market forces to operate. If we

divide drivers in three groups with respect to their attitude to ISA,

4. Discussion and concluding remarks (1) those who need ISA but do not want it, (2) those who need ISA

and accept it after having tried it, and (3) those who want ISA but

In the PAYS project, we succeeded in developing a reliable and do not need it, a market-based strategy predicated on the volun-

technically advanced ISA system, but we did not reach our goal of tary use of ISA will reach only Group 3. For ISA to reach Groups

300 young drivers participating in the PAYS project for three years; 1 and 2, we require alternatives to purchasing ISA on the market.

instead, we procured 153 participants of all ages. Our ﬁrst plan for The following are alternative methods for implementing ISA. (1)

the project included a fee of 700 combined with a discount on Mandatory ISA for selected groups with a high risk of accidents,

insurance rates, corresponded to a normal, market-based plan, but such as inexperienced young drivers. Inexperienced drivers would

only 40 drivers signed up under those conditions. The recruitment be required to practice proper speed control by driving 20,000 km

of a suitable driver population for research purposes developed into in a car equipped with a recording ISA system prior to receiving a

a strenuous and unsuccessful struggle. This might be a local, Danish, full driver’s license. (2) Mandatory ISA to be included in the cat-

phenomenon, but it may also reﬂect a tendency among drivers in alogue for speeding sanctions to enhance compliance with speed

general. They may not wish to relinquish the option to speed, even limits, in much the same way as the Alco lock is currently deployed

for economic gain. to prevent drinking and driving in such countries as Sweden and

In fact, even an economic incentive in the form of a 30% insur- Finland (Danish Ministry of Justice, 2010). (3) Mandatory ISA in cars

ance discount was not sufﬁcient to persuade many drivers to use performing public services and support for companies installing ISA

this equipment. The strenuous and unsuccessful efforts to recruit in their cars.

participants for the project led us to conclude that large psycholog- Finally, based on the PAYS results, we can conclude that ISA is

ical barriers must be overcome before young motorists will permit technically ready for market introduction, but also that ISA is not

the installation of ISA in their cars. The ﬁnancial incentive offered market driven. To implement ISA on a large scale, society must both

was far from enough for car owners in the county of North Jut- legislate its use and actively support the voluntary adoption of the

land to accept ISA over an extended period of time. In principle, system.

the target group was favorable to ISA and the PAYS project, but

when it came to installing an OBU in their own cars, they invariably

found excuses for refusing to do so. Prospective participants sup- Acknowledgements

plied many excuses, with the predominant one being the drivers’

desire to remain free to speed if they desired. The authors would like to thank the following organizations,

Removing the age limit did not change this pattern. It was not funds, businesses and individuals for their ﬁnancial support of

possible by any reasonable means to reach the goal of 300 par- the project: The Danish Ministry of Transport, Fonden Østifterne,

ticipants in this project. In fact, only 0.14% of all car owners in Det Obelske Familiefond, Sonofon and SMC-biler. The authors also

North Jutland and 0.9% of Topdanmark customers in North Jutland acknowledge the contributions to the project made by Ian Sonne

volunteered to participate in the ﬁeld trial. Moreover, of the 10 Berg, North Jutland County, and Keld Bruun Hansen, Christina K.

local politicians with responsibility for trafﬁc safety that the plan Daub and Christian Tangdal, all of Topdanmark. We also wish to

16 H. Lahrmann et al. / Accident Analysis and Prevention 48 (2012) 10–16

acknowledge the expert monitoring group of the project for their Juhl, J., Lahrmann, H., Sonne, I.B., Heide, P., 2007. Spar på farten. GeoforumPerspektiv

11, 25–38 (in Danish).

suggestions and support, with special thanks to Martin Hellung

Lahrmann, H., Agerholm, N., Tradisauskas, N., Bertelsen, K., Harms, L., 2011. Pay as

Larsen of The Danish Road Safety and Transport Agency, Jesper

you speed, ISA with incentive for not speeding, results and interpretation of

Sølund of The Danish Road Safety Council and Vagn Bech, North speed data. Accident analyses and prevention. Special Issue on Intelligent Speed

Adaptation.

Jutland County (now with the Danish Road Directorate).

Lahrmann, H., Madsen, J.R., Boroch, T., 2001. Intelligent speed adaptation develop-

ment of a GPS-based ISA-system and ﬁeld trial of the system with 24 drivers.

References In: Proceedings from Eighth World Congress on Intelligent Transport Systems ,

Sydney, Australia, September 30–October 4, 2001.

Lahrmann, H., Madsen, J.R., Boroch, T., 2004. Intelligent speed adaptation. Dansk

Agerholm, N., Juhl, J., Sonne, I.B., Lahrmann, H., 2007. Speed map maintenance –

Vejtidsskrift 81 (1), 18–22 (in Danish).

experiences from the intelligent speed adaptation project. Spar paa Farten. In:

Marchau, V.A.W.J., van der Heijden, R.E.C.M., Molin, E.J.E., 2005. Desirability of

6th European Congress on ITS , Aalborg, Denmark, June 18–20, 2007.

advanced driver assistance from road safety perspective: the case of ISA. Safety

Agerholm, N., n.d. Recruitment for the pay as you drive project (unpublished working

Science 43 (1), 11–27.

paper, in Danish).

Markman Customer Management, n.d. Analyzing the Recruitment for the Pay As You

Almquist, S., 2006. Loyal Speed Adaptation, Speed Limitation by Means of an Active

Drive Project (unpublished working paper, in Danish).

Accelerator and its Possible Impacts in Built-up Areas. Bulletin 232. Lund Insti-

Peltola, H., Tapio, J., Rajamäki, R., 2004. Intelligent speed adaptation (ISA) – recording

tute of Technology, Lund, Sweden.

ISA in Finland. In: Proceedings from Via Nordica.

Belonitor, 2009. Rijkswaterstaat. http://wiki.fot-net.eu/index.php?title=Belonitor

Persson, H., Towliat, M., Almquist, S., Riesser, R., Magdeburg, M., 1993. Speed Limiters

(viewed March 1, 2009).

for Cars. A Field Study of Driving Speeds, Driver Behaviour, Trafﬁc Conﬂicts and

Biding, T., Lind, G., 2002. Intelligent Speed Adaptation (ISA), Results of Large-scale

Comments of Test Drivers During Driving in City Trafﬁc. Department of Trafﬁc

Trials in Borlänge, Lidköping, Lund and Umeå During the Period 1999–2002. No.

Planning and Engineering, University of Lund, Sweden (in Swedish).

89, Vägverket, Borlänge, Sweden.

Regan, M., Young, K.L., Triggs, T.J., Tomasevic, N., Mitsopoulos, E., Tierney, P., Healy,

Boroch, T., 2002. Intelligent speed adaptation in Denmark, ﬁeld trial of a GPS-based

D., Tingvall, C., Stephan, K., 2006. Impact on driving performance of intelligent

system for intelligent speed adaptation. In: Proceedings form Conference Esafety

speed adaptation, following distance warning and seatbelt reminder systems:

, Lyon, France, September 16–18, 2002. ITS Europe, ERTICO.

key ﬁndings from the TAC SafeCar project. IEE Intelligent Transport Systems 153

Brems, C., Munch, K., 2008. Risiko i traﬁkken 2000–2007. DTU Transport (in Danish).

(1), 51–62.

Carsten, O.M.J., Tate, F.N., 2005. Intelligent speed adaptation: accident savings and

Servin, O., Boriboonsomsin, K., Barth, M., 2006. An energy and emissions impact

cost–beneﬁt analysis. Accident Analysis and Prevention 37 (3), 407–416.

evaluation of intelligent speed adaptation. In: Proceedings form Intelligent

Copenhagen Police, 2010. FAQ. http://www.politi.dk/Koebenhavn/da/servicemenu/

Transportation Systems Conference, IEEE , Toronto, Canada, September 17–20,

FAQ/ATK/ (viewed November 16, 2010).

2006, pp. 1257–1262.

Danish Ministry of Justice, 2010. Comments on the Bill Amending the Road

Sonne, I.B., 2005. Establishing speed map in the northern part of Denmark. In: Pro-

Trafﬁc Act and Criminal Code (in Danish). https://www.retsinformation.dk/

ceedings from the 5th European Congress on ITS , Hannover, Germany, June 1–3,

Forms/R0710.aspx?id=130978 (viewed November 22, 2010).

2005.

Harms, L., Klarborg, B., Lahrmann, H., Agerholm, N., Jensen, E., Tradisauskas, N., 2008.

Topdanmark, n.d. The Potential among Topdanmark’s Own Customers to Participate

Controlled study of ISA effects: comparing speed attitudes between young vol-

in the Pay As You Speed Project (unpublished working paper, in Danish).

unteers and external controls, and the effect of different ISA treatments on the

Tradisauskas, N., Juhl, J., Lahrmann, H., Jensen, C.S., 2009. Map matching for intelli-

speeding of volunteers. IET Intelligent Transport Systems 2 (2), 154–160.

gent speed adaptation. IET Intelligent Transport Systems 3 (1), 57–66.

Hjälmdahl, M., 2004. In-vehicle Speed Adaptation: on the Effectiveness of a Volun-

Vlassenroot, S., Broekx, S., De Mol, J., Panis, L.I., Brijs, T., Wets, G., 2007. Driving with

tary System. PhD thesis, Bulletin 223, Lund University, Sweden.

intelligent speed adaptation: ﬁnal results of the Belgian ISA-trial. Transportation

Hultkrantz, H., Lindberg, G., 2009. Pay As You Speed: An Economic Field Experiment.

Research, Part A (Policy and Practice) 41 (3), 267–279.

Working Paper. Örebro University, Sweden.

Várhelyi, A., Hjälmdahl, M., Hydén, C., Draskóczy, M., 2004. Effects of an active accel-

Jamson, S., 2006. Would those who need ISA use it? Investigating the relationship

erator pedal on driver behaviour and trafﬁc safety after long-term use in urban

between drivers speed choice and their use of a voluntary ISA system. Trans-

areas. Accident Analysis and Prevention 36, 729–737.

portation Research Part F: Trafﬁc Psychology and Behaviour 9 (3), 195–206.

Warner, H.W., 2006. Factors Inﬂuencing Drivers’ Speeding Behaviour. Dissertation

Jamson, S., et al., 2006. Intelligent Speed Adaptation – Literature Review and Scoping

from the Faculty of Social Sciences Uppsala University, Sweden.

Study. University of Leeds, Leeds, United Kingdom.