[Title will be auto-generated]

No 2 · 2000

News from Denmark, Finland, Iceland, Nor way and Sweden

Taxation of Road Freight Transport in Europe page 4 What roundabout design provides the highest possible safety page 17

Editorial notes Nordic Road & Transport Research is a joint publication of six public road and transport research organisations in the Nordic countries, Denmark, Finland, Iceland, Norway, and Sweden. The main objective of the publication is to disseminate research results and news from the institutions, especially to researchers and decision makers. Each institution is responsible for the selection and presentation of the material from its own scope of activities.

Nordic Road & Transport Research is published three times a year. It is regularly sent out, free of charge, to recipients selected by the five joint publishers. Free sample copies are also sent out on special request.

Reproduction and quotation of the text are allowed if reference is made to the author and source. However, legislation regulates and restricts the right to reproduce the illustrations. Please contact the respective publishing institution for information.

Advertising is not accepted.

Correspondence about the contents of the publication: Please write to the author or to the respective publishing organisation.

Requests for back issues, and notification of address changes: Readers outside the Nordic countries: please write to the Editor-in-chief at the VTI in Sweden.

Readers in the Nordic countries: please contact the publishing institution of your country.

Addresses: see back cover.

The Editorial Board consists of the following representatives of the publishing institutions.

Editor-in-Chief Sigvard Tim, Swedish National Road and Transport Research Institute Denmark Helen Hasz-Singh, Danish Road Institute Finland Kari Mäkelä, Technical Research Centre of Finland, Communities and Infrastructure Iceland Hreinn Haraldsson, Public Roads Administration Norway Helge Holte, Norwegian Public Roads Administration Harald Aas, Institute of Transport Economics

Production: VTI Information Place of publication: Linköping, Sweden Issue: 3,500 ISSN: 1101-5179

2

Volume 12 Æ No.2 ÆAugust 2000

Danish Road Directorate (DRD) The Road Directorate, which is a part of The Ministry of Transport, Denmark, is responsible for development and management of the national highways and for servicing and facilitating traffic on the network. As part of this responsibility, the Directorate conducts R&D, the aim of which is to contribute to efficient road management and to the safe use of the network. The materials research component is carried out by the Danish Road Institute while other R&D activities — primarily safety and environmental research — are carried out by the Directorates operational departments.

Technical Research Centre of Finland (VTT), Communities and Infrastructure Communities and Infrastructure, employing a staff of 175, is one of the nine Operating Units of the Technical Research Centre of Finland (VTT), which has a total staff of 3,000. Research at this Operating Unit covers all aspects of transport, road engineering, geotechnology and urban planning. The unit has active international relations and a prominent role in these research areas on the national level.

Public Roads Administration (PRA), Iceland The duty of PRA is to provide society with a road system according to its needs and to offer service aiming at safe, unobstructed traffic. The number of employees is about 340.

Applied research concerning road construction, maintenance and traffic and safety is to some extent performed or directed by the PRA. The authority with its Research and Development division is responsible for road research in Iceland.

Norwegian Public Roads Administration (NPRA) The Norwegian Public Roads Administration is one of the administrative agencies under the Ministry of Transport and Communications in Norway. The NPRA is responsible for the development and management of public roads and road traffic, as well as the Vehicle Department. This responsibility includes research and development of all areas related to road transport, and the application of R&D products.

Institute of Transport Economics (TØI) The Institute of Transport Economics is the national institution for transport research and development in Norway. The main objectives of the Institute are to carry out applied research and promote the application and use of results through consultative assistance to public authorities, the transport industry and others.

The Institute is an independent research foundation employing about one hundred persons.

The Swedish National Road and Transport Research Institute (VTI) is responsible for research and development in road construction, maintenance, road traffic and transport, railroads, rail transport, vehicles, road user behaviour, traffic safety and the environment.

The Institute is state-owned and has a total of 215 employees.

Cover Photo: Naturfotograferna NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

C O N T E N T S Taxation of Road Freight Transport in Europe ............................................................... 4 Technical Research Centre of Finland (VTT), Communities and Infrastructure

Analysis of the telework potential is needed for estimating impacts on transport ......................................................................................................................................................... 6 Technical Research Centre of Finland (VTT), Communities and Infrastructure

Thin pavements with synthetic binder used in Denmark .................................... 9 Danish Road Directorate (DRD)

New friction measuring device -for safer driving during winter .............. 11 Norwegian Public Roads Administration (NPRA)

Developments of performance contracts in Norway ........................................... 13 Institute of Transport Economics (TØI)

What roundabout design provides the highest possible safety ................. 17 Swedish National Road and Transport Research Institute (VTI)

Optimisation of clothoid lengths ............................................................. 22 Swedish National Road and Transport Research Institute (VTI)

Ride comfort and motion sickness in tilting trains ................................ 23 Swedish National Road and Transport Research Institute (VTI)

A N N O T A T E D

R E P O R T S

Danish Road Directorate (DRD) ............................................................................................................. 24 Technical Research Centre of Finland (VTT), Communities and Infrastructure) ............... 25 Institute of Transport Economics (TØI) ................................................................................................ 26 Norwegian Public Roads Administration (NPRA) ........................................................................ 28 Swedish National Road and Transport Research Institute (VTI) ............................................. 29

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

3

TECHNICAL RESEARCH CENTRE OF FINLAND (VTT), COMMUNITIES AND INFRASTRUCTURE

Taxation of Road Freight Transport in Europe The taxation of road freight transport is based on diverse tax regimes, and the levels of taxes and charges on heavy goods vehicles are extremely varied. New regulations by the European Union aim at harmonizing the systems of taxation, while basing levels of taxes and charges on increasingly territorial standards.

Freight transport taxation is a topical issue in Europe for a number of reasons. The European Parliament and Council recently

issued a new Directive (Charging of heavy goods vehicles for the use of certain infrastructures, 1999/62/EC) on user charges that will affect tax systems in the member countries of the EU. High fuel prices have led to demands in many countries to reduce taxes on diesel fuel. Moreover, the general trends in transport policy have had a crucial impact on transport pricing and taxation in Europe.

A study has been conducted at the Technical Research Center of Finland, commissioned by the Ministry of Transport and Communications, to explore the systems

of road freight taxation in Europe. The study offers a useful complement to the ongoing work of the ECMTs (European Conference of Ministers of Transport) Ad hoc Group on Fiscal and Financial Aspects of Transport Markets, in developing methodologies for international comparison of the level and structures of transport taxation.

The taxes and charges examined in the study include vehicle taxes, infrastructure user charges, road tolls, and fuel-related duties. The study is limited to heavy goods vehicles (HGVs) in road freight transport.

The countries that responded to the survey and which are included in the study are the United Kingdom, Germany, France, Belgium, Luxembourg, Spain, Portugal, Austria, Switzerland, Lithuania, Latvia, Denmark, Sweden, Norway, and Finland.

The data were collected from the various ministries and expert organizations in Europe during November 1999-March 2000. The figures represent the situation at the beginning of 2000. However, changes are being planned in many countries and some of the figures will soon be outdated.

Vehicle taxes and user charges

Hauliers are subjected to varied taxation regimes in Europe.

4

Vehicle taxes are periodic (normally annual) taxes levied on vehicles in their country of registration. The minimum rates of taxation for vehicles and vehicle combinations with a total weight of 12 tonnes and over are specified in the Directive. According to the Directive, the taxes should be differentiated by the vehicle s total weight, axle number and suspension system.

User charges are payments of a specified amount conferring the right for a vehicle to use over a given period the infrastructures concerned (generally motorways). The Directive specifies the maximum user charges for various vehicle categories, differentiated by the vehicle s axle number and emission class. Six countries (Germany, Belgium, the Netherlands, Luxembourg, Denmark and Swe-

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

TECHNICAL RESEARCH CENTRE OF FINLAND (VTT), COMMUNITIES AND INFRASTRUCTURE

den) are currently participating in a common user charge system, the Eurovignette.

Road tolls are payments of a specified amount for a vehicle travelling between two points in the infrastructures concerned. In Europe, various forms of road tolls are collected, at least in Spain, France, Italy, Portugal, Hungary, Austria, Norway and

vehicle tax

Eurovignette

impose other fuel-related taxes and duties, such as environmental taxes, stockpiling fees, etc.

Based on Euroshell s statistics on diesel fuel prices and the information obtained from the respondents of this study, the prices and duties on diesel fuel are shown in Figure 2. Again, the highest level of taxation

Other user charges

AT BE CH* DE DK ES FI FR GR LT LU LV NO PT SE UK 0

2000

4000

6000

8000

10000

Title: Taxes and User Charges on Heavy Goods Vehicles in Europe.

Author: Jari Juga Series: Ministry of Transport and Communications, Report 23/2000 Language: English

euro/year

*) Note: The maximum total weight in Switzerland is 28 tonnes.

Figure 1. Annual Vehicle taxes and user charges ( /year), excluding r oad tolls, on a semitrailer (2+3 axles, total weight 40 tonnes) in the various European countries in 2000.

Greece. Unfortunately, the statistics on annual road tolls paid by owners of heavy goods vehicles are very deficient.

Figure 1 shows the annual vehicle taxes and user charges on an articulated vehicle (tractive unit and a semi-trailer, 2+3 axles, 40 tonnes total weight) in the various European countries. The heaviest taxation currently exists in the United Kingdom. It should be noted, however, that no correction has been made in these figures for Switzerland, where the total vehicle weight is limited to 28 tonnes. In addition, no estimates of annual road tolls are included in these figures.

Fuel-related duties

Fuel-related duties are governed by Directives 92/81/EEC (Harmonization of the structures of excise duties on mineral oils) and 92/82/EEC (Approximation of the rates of excise duties on mineral oils). Besides the excise duty on diesel fuel, several countries

port taxation, levels of taxes and charges will be determined locally within the framework outlined in a wider European context.

The question of transport taxation is intricately linked to the principles of transport pricing. Efficient taxation regimes should be based on marginal external costs (cost of pollution, accidents, road wear, etc.), as concluded by CEMT in its report on fiscal and financial aspects of transport markets. In addition, tax systems should be transparent and non-discriminatory between national and foreign transport operators.

The study shows that systems and levels of road freight taxation in Europe are very diverse. However, steps are clearly being taken in many countries to introduce taxation regimes that strengthen the territorial nature of taxes and charges on road freight transport. For instance, there is a current debate on whether the Eurovignette system should be replaced by distance-based user charges in the coming years. New member candidates for the EU have also initiated measures to harmonize their taxation with EU standards.

can be found in the U.K., with Norway, another oil producer, taking second place. In all the countries and vehicle categories studied, fuel-related duties clearly represent the largest tax item for road freight transport, given the standard annual driving performances of HGVs in these countries.

GR

Future trends

DE

duties

price

PT LT

ES LU LV BE AT FI

As EU policy heads towards an increasingly territorial system of freight transport taxation in Europe, distance-based user charges and road tolls for road freight transport will certainly appear in the next few years.

Although the EU is also trying to harmonize systems of trans-

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

SE DK FR CH NO UK 0

0,2

0,4

0,6

0,8

1

1,2

euro/litre

Figure 2. Diesel fuel prices and fuel-related duties in European countries ( /litr e, VAT excluded), Januari 2000.

5

TECHNICAL RESEARCH CENTRE OF FINLAND (VTT), COMMUNITIES AND INFRASTRUCTURE

Analysis of the telework potential is needed for estimating impacts on transport One of the main reasons for interest in telework is its potentially beneficial impact on mobility patterns.

Telework raises numerous expectations concerning transport, ranging from energy and time savings, as well as alleviation of congestion, to the reduction of carbon dioxide and other greenhouse gas emissions.

Not much can be said, however, of the future impacts of telework on transport unless a clearer view of the telework potential is at hand.

Introduction

Telework should be seen as an element in a broader context of the information society. It is essentially a new way of organising work and services. Actually, teleworking is part of a whole bunch of activities performed utilising new information and communication technologies (ICT), regardless of time and place. These are called teleactivities, and they represent telepresence. In other words, it is possible to be virtually present through ICT at the office, school, university, banks, shops and hospitals creating a potential for substituting physical transport with virtual mobility. Teleactivities comprise telework, tele-education, teleshopping, telebanking and other teleservices, telemedicine etc.

Telework and its penetration into society is also associated with changes in the nature and concept of work. Work is becoming increasingly flexible and mobile, and can be done in various places, anytime, anywhere, even on the move.

A Project on the Finnish Telework Potential

VTT Communities and Infrastructure has completed a two-year project estimating the Finnish telework potential, financed by

the European Social Fund (ESF) and the Ministry of Labour (Heinonen 1999a). The size of the working population was calculated as regards the possibilities of workers to transfer to teleworking under certain conditions.

The project had a strong futures studies approach, meaning that the changing nature and significance of work for human beings was also dealt with. The project utilised a GIS system (Geographical Information System) to analyse the telework potential. The suitability of the Finnish working population to telework was first assessed on the basis of the occupation structure. It is clear that teleworking is not suited to all jobs. Moreover, within an occupation that is suitable for telework there are certain tasks that cannot be carried out in telepresence. The statistics on the occupations were analysed and the number of

people in telefeasible occupations was calculated.

Benefits and Obstacles of Telework

Various prerequisites and benefits of telework, as well as looming obstacles and risks, were analysed as key elements affecting the realisation of the telework potential. For employees themselves, the greatest risks identified are feelings of isolation and burn-out. Apart from benefits, telework may have contradictory repercussions for the environment and society, such as a scattered community structure and increased vehicle kilometres during leisure (Heinonen & Weber 1998). Therefore, a balanced introduction of telework into society aims at maximising possible benefits from telework and minimising obstacles and disadvantages.

TELEWORKING BENEFITS FOR DIFFERENT ACTORS For the employee • Savings in commuting time and expenses • Strengthening working motivation • Flexibility of the working mode • Fitting work into his/her own rhythm and situation of life • Peace to do work, followed by a higher work efficiency • Emphasis on the contents of work • Increase in employability For the employer • Lower overhead costs • Increase in productivity • Keeping skilled employees and recruiting new ones • A better company image concerning environmental issues For the environment and society • Less traffic • Decrease in emissions caused by fuel consumption and traffic • Less consumption of resources • Savings in infrastructure • Improved local economy Figure 1. Possible benefits from telework.

6

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

TECHNICAL RESEARCH CENTRE OF FINLAND (VTT), COMMUNITIES AND INFRASTRUCTURE

Long commuting distances can be considered a major motivating factor for telework. In Finland, people who have a long commuting distance from home to work (more than 50 km) represent only some 4% of the working population. However, they account for a third of vehicle kilometres.

work force of a country. Towards this basic set are directed driving forces promoting realisation of the telework potential on the one hand, and obstructing forces on the other. Knowledge and experience of the desirable benefits of telework strengthen these prerequisites, while the suspected disadvantages and risks function as obstacles.

Calculation model for the telework potential Basic population

Telework potential Teleworking population Effects

Prerequisites of the Realisation Obstacles

Time

Figure 2. Model for estimating the telework potential and its different effects, when realised for a teleworking population.

Calculation Model of the Telework Potential

Numerous estimates exist of current teleworking populations in various countries, but are often compiled from very different criteria and lack comparability. An alternative approach, adopted in our project, is to calculate the telework potential in a country.

The forecasting method for estimating the telework potential is based on certain available parameters. The size of the Finnish telework potential was calculated to include 450 000 employees according to tighter state-of-the-art criteria and 820 000 employees according to looser proactive criteria (respectively 20% and 40% of the working population). The calculation model is a process schedule, applicable on a national/provincial/municipal or single company level. Data on e.g. the occupations of the inhabitants/employees, locations of habitation, and work and commuting distances are needed for the model.

The research target is, for example, the

The estimation model is divided into four main stages: ¥ defining the basic population targeted for the study; ¥ transforming the chosen population into a teleworking potential (analysing the professional structure of the basic population), ¥ anticipating the realisation of the telework potential, ¥ calculating the impacts of the realised potential.

Savings from Telework

Estimation of the implications of the potential is divided into several partial functions, depending on which effects are under consideration. The first target for savings is the decreased commuting traffic due to teleworking. This is also the calculation base for other effects and savings accumulated by telework. A certain decrease in the commuting mileage saves a certain number of hours, which means fewer tons of emissions. A decrease in traffic accidents can also be estimated.

When an employee has used public transport, two-wheel transportation, a car pool, or has walked, decreases in emissions are naturally not achieved. In these cases the effects on transport manifest as savings in commuting time, and as diminished risks of traffic accidents.

The possibilities to reduce commuting trips through telework can thus be calculated. If a person transfers to telework for one day a week, the number of his/her commuting trips and the total mileage will drop by 20% (naturally, this requires one full day and not two part-time days a week). If telework was to be done by 200 000 Finns usually travelling by car (e.g. an average commuting distance of 10 km each way in the greater Metropolitan area) one day a week (44 days a year), the annual drop in commuting traffic would be 8.8 million commuting roundtrips and a mileage of 170 million kilometres (see table 1). The socio-economic savings on vehicles would amount to FIM 140 million a year, including FIM 35 million as fuel expenses. The yearly savings for private households would be about FIM 300 million.

Number of commuting roundtrips

8.8 million

Time used for commuting

6 million hours

Commuting mileage

170 million kilometres

Driving expenses/socio-economy

FIM 140 million

Driving expenses/private households

FIM 300 million

Accident expenses

FIM 20 million

Emission costs

FIM 10 million

Infrastructure expenses

Postponement of investment decisions

Table 1. Yearly savings brought about by telework (200 000 teleworkers one day/week).

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

7

Promoting the Realisation of the Telework Potential

The Finnish telework potential is thus promising, but the key question is how to utilise it efficiently. Organisations should be made aware of the new dimensions of work, above all in relation to the environmental and ethical viewpoints, both as an employability factor of human telework resources and as an image factor for the companies themselves. In telework the core competence of the teleworker can be utilised efficiently, free from the chains of place and time, while at the same time improving the teleworker s quality of life.

Greatest benefits for transport could be reaped if those with long commuting distances who normally drive to work could be transferred to telework. It is becoming more and more natural in the global information society for computers and the Internet partially to take over the functions

8

of a car as a symbol of mobility (Heinonen 1999b). Telework is but a facet of the new e-mobility, while a big issue still remains whether telework can be realised as net savings of transport.

Contact: Sirkka Heinonen, Ph.D., senior research scientist VTT Communities and Infrastructure POB 1901, 02044 VTT, Finland tel + 358 9 456 6288, fax + 358 9 464 174 email: sirkka.heinonen@vtt.fi

Bibliography Heinonen (1999a): Heinonen, Sirkka, Analysis of the Finnish Telework Potential. Calculation Model. VTT Communities and Infrastructure, Helsinki 1999, 85 p. (electronic version forthcoming at the website of the Ministry of Labour in Finland, http://www.mol.fi/esf/ennakointi/ raportit/telework.pdf Heinonen (1999b): Heinonen, Sirkka, Car or Computer? Telepresence as a New Paradigm for Future Mobility. In: Mauranen. Tapani (ed.); Traffic, Needs, Roads. Perspectives on the Past, Present and Future of Roads in Finland and the Baltic Area, pp. 241-247.

Heinonen, Sirkka & Weber, Matthias, Recent Experience with Teleworking: Effects on Transport. IPTS Report, February/98, p. 27-33.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

DANISH ROAD DIRECTORATE (DRD)

Thin pavements with synthetic binder used in Denmark

One of the inspected thin pavements with synthetic binder in good condition.

Thin pavements with synthetic binder have been used in Denmark for more than 20 years as a combined waterproofing and surfacing layer on steel and concrete bridges.

The Danish Road Institute has just completed an evaluation of older as well as newer thin pavements with synthetic binder in Denmark. It has been found that these types of pavements are a good alternative to ordinary bitumen-based waterproofing/bridge pavements.

Since the 1930 s, it has been common practice in Denmark for large steel and concrete bridge decks to be waterproofed. The waterproofing systems have been under constant development and the most commonly used waterproofing system on concrete bridges today is a system based on polymer modified bituminous sheets.

Other waterproofing systems, such as thin pavements with synthetic binder, are also used. Approximately 5% of the Danish bridges are paved with thin pavements containing synthetic binder.

In the Danish Road Standards, the waterproofing systems are classified on the basis of the Annual Daily Traffic (ADT).

According to the Standards, thin pavements with synthetic binder can be used on roadway bridges with an ADT less than 2,000 vehicles. When further positive re-

sults become available with this type of pavement, it can be expected that thin pavements with synthetic binder can also be allowed on roadway bridges with an ADT of 2,000-4,000 vehicles. Thin pavements with synthetic binder have mainly been used on steel and concrete bridge decks with an ADT of less than 2,000 vehicles, as well as footpaths, bicycle tracks and pedestrian bridges. On an experimental basis, however, these pavements have been used on bridge decks with an ADT of up to 46,000 vehicles.

Thin pavements with synthetic binder have been used in Denmark for more than 20 years. Since the lifetime of thin pavements with synthetic binder has been estimated to be 10 - 15 years, it is now possible to evaluate the durability of this type of pavement and thereby its use as a waterproofing/bridge pavement. For this purpose, the Danish Road Institute has examined some 36 thin pavements with synthetic binder by evaluating their common condition. The survey has created a significant database with information on the most common types of damage and their causes.

Thin pavements with synthetic binder

The most commonly used types of thin pavement with synthetic binder are 2-layer pavements and 1-layer pavements. 2-layer

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

pavements consist of a waterproofing layer and a surface layer, each layer based on pure synthetic binder, normally acrylic or epoxy compositions. 1-layer pavements consist of a combined waterproofing and surfacing layer based on a synthetic mortar, normally epoxy compositions and filler/sand.

In recent years, more attention has been paid to environmental conditions, and pavements based on polyurethane modified epoxy are now being used, instead of tar epoxy. For both designs, it is necessary to apply a suitable friction material on the pavement surface. For these types of pavements, the general pavement thickness is between 5 and 10 mm, depending on the actual traffic load.

Inspection in situ of thin pavements with synthetic binder

From 200 existing thin pavements with synthetic binder on steel and concrete bridges, 25 large and small bridges (including bascules and ferry berths) with thin pavements with synthetic binder were selected for inspection as a pavement on the roadway, footpath/bicycle track, bascule bridge, pier or edge beam.

Since there may be several types of material, pavement constructions, base, traffic load types and age differences on the

9

DANISH ROAD DIRECTORATE (DRD)

same bridge, there may be several types of pavement with synthetic binder on an individual bridge, which have been evaluated separately. Therefore 36 pavements with synthetic binder have been evaluated on the 25 concrete and steel bridges.

The age of the examined pavements ranges from 1 to 17 years, with an average age of 8 years and an ADT from 200 to 46,000 vehicles, with an average ADT of 6,500 vehicles on 18 roadway bridge pavements.

Evaluation of the inspected pavements

On the basis of the inspections, it can generally be stated that the pavements are in relatively good condition. This also applies to those pavements which have been exposed to a considerably greater traffic load than specified in the Danish Road Standards.

This indicates that thin pavements with synthetic binder applied under optimal conditions can provide a satisfactory alternative to ordinary bitumen-based waterproofing/bridge pavements - also for bridges with an ADT greater than 2,000 vehicles.

However, a number of different types of damage were observed. The damage used to evaluate the pavements may be due to the materials, workmanship and/or influence from the surroundings, such as traffic load/amount, weather conditions and/ or conditions during the construction of the concrete/steel base.

The different types of damage observed during the inspection are listed below according to their frequency: 1. Failures due to poor workmanship 2. Scaling 3. Polishing of chippings 4. Crack formation 5. Loss of chippings 6. Blistering

Furthermore, the following observations were made: ¥ Damage such as polishing and loss of chippings appears exclusively on roadway bridge pavements. ¥ Thin pavements based on synthetic mortars seem to have a greater tendency to crack than thin pavements based on purely synthetic materials strewn with chippings. ¥ Calcined bauxite used as skid resistance material seems to be more resistant to polishing than electric arc steel slag.

Report

The evaluation of the inspection of the thin pavements with synthetic binder is docu-

mented in a final report in February 2000.

The report has been published electronically and can be seen on web site http:// www.vd.dk. In addition to the final report, a data report has been issued (only in hard copy).

This article is a summary of the paper Thin pavements with synthetic binder used in Denmark pr esented at the 24th International Baltic Road Conference, Riga, 21-23 August 2000.

Jeanne Rosenberg, Asphalt Department, Danish Road Institute P.O. Box 235, Elisagaardsvej 5, DK-4000 Roskilde

Chippings Wearing course

Membrane sanded with chippings Primer Base Chippings Wearing course Membrane Primer Base

Chippings Wearing course (mortar) Membrane Primer Base

Illustrations of a 2-layer construction of a thin pavement with synthetic binder.

Chippings

Synthetic mortar Primer Base

Illustration of a 1-layer construction of an alternative thin pavement with synthetic binder.

10

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

NORWEGIAN PUBLIC ROADS ADMINISTRATION (NPRA)

New friction measuring device - for safer driving during winter More stringent requirements for greater traffic safety are increasing the need for improved friction on the roads. The work aimed at achieving higher friction through the use of better maintenance methods and more advanced techniques has been given a high priority within the Norwegian Public Roads Administration.

SALTAR is a newly developed friction measuring device, where measurements can be carried out without the need to apply the vehicle s brakes.

Many accidents can be ascribed to poor road grip. Even though speeding is the primary cause of accidents, it is obvious that greater friction is of importance when it comes to preventing accidents and damage. This work is mainly targeted at winter conditions, but also at new pavements and the repair of pavements.

What is friction?

To create standards relating to friction conditions on roads, it is important to be aware of all the factors that influence friction.

Until 10-15 years ago, the term friction referred to the car s braking distance when the brake pedal was fully depressed and the wheels were totally locked. In recent years, research and testing have revealed that friction during a braking phase is very complex, and that it is influenced by many different conditions. Friction depends on wheel slip in relation to the surface. The term slip refers to the difference between free rolling wheel speed and the wheel speed when the brakes are applied as a percentage of free rolling wheel speed.

This means that when the brakes are not applied to car wheels at all, there is 0% slip, but when the wheels are totally locked through braking, the slip is 100%.

When the wheels start to brake, the friction increases rapidly until the slip is about 20%. After that, the friction will decrease to its lowest value when the wheels are totally locked at 100% slip. This can be

seen very clearly in the case of clean and wet surfaces.

The friction sequence described can vary considerably depending on the surface conditions, the speed of the car, air temperature, tread and rubber quality of the car tyres, as well as contaminants on the road such as ice, snow and water. Typical graphs describing the friction conditions for different road surfaces depending on the wheel slip are shown in figure 1, page 12.

To achieve the best possible braking effect, new cars today are equipped with ABS brakes allowing a maximum of 1520% slip. All these conditions need to be considered when setting requirements based on friction on the roads.

Friction requirements

Friction requirements must reflect the standard laid down for traffic safety in relation to traffic density and the road s speed level. They must be based on the braking systems and tyres modern cars are equipped with, as well as the road surface quality the market can offer. Furthermore, it is necessary to consider older cars with conventional braking systems and the fact that the car tyres change as they are worn down. The requirements must be based on well-organised, technically advanced and systematic road maintenance.

Documentation

When requirements have been set, it is only natural and necessary to demand documentation from all constructors of new pavements or operators clearing snow, sanding and salting the roads. This documentation should relate to quality assurance and control, showing that the work has been carried out in accordance with the standards and requirements. In a situation where there is increasing competition between private contractors and the Production Department of the Public Roads Administration, distinct requirements and documentation for correct work are necessary in order to make an objective evaluation and

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

the right choice among those competing for the work.

Measuring equipment

To prepare documentation regarding the friction conditions on roads, it is necessary for friction measuring equipment to be available that is directly related to the standards, and for this equipment to give distinct and comparable measurement results. The requirements should also as far as possible reflect the road grip the cars will experience.

In the period 1988-90, the Directorate of Public Roads, in co-operation with the Norwegian company Norsemeter, developed a highly advanced friction measuring unit called OSCAR. This unit could measure friction at all speeds without braking the vehicle. The unit can measure either the total friction sequence, or with the required percentage slip. The Public Roads Administration has demonstrated this advanced

11

NORWEGIAN PUBLIC ROADS ADMINISTRATION (NPRA)

unit at many international conferences and exhibitions, and has participated in tests on different friction measuring devices from all over the world as part of its work aimed at trying to develop international standards for measuring methods and requirements.

In 1993, a smaller and less expensive unit was developed called ROAR. This unit has been sold to ten different countries and is used to measure friction both on roads and at airports. The Public Roads Administration has four of these units. Depending on the equipment and software for processing data, each costs between NOK 0.5-0.7 million ex. VAT. In addition to these units, the Public Roads Administration has several hundred decelerometers installed in regular cars which, during braking, give a picture of the friction on the road. Braking in traffic is not good and something the Norwegian Public Roads Administration wants to stop doing.

New equipment

In order to develop a piece of equipment that could be used without braking the vehicle, and at a price which would make it possible to acquire a number of these for control and documentation, a working

party was set up in 1997 within the Public Roads Administration comprising representatives from various road offices. The group devised a product specification for a newer and more basic measuring unit, at the time known as SALTAR. The specifications for this unit also took into consideration requests from Road Authorities in Minnesota and Iowa. Three prototypes were supplied by the company Norsemeter in April 1999, one to Norway and two to the USA. Testing

The units delivered in April 1999 have been tested both in Norway and in the USA. Results to date show that the unit satisfies the requirements laid down and that it is reliable. The unit will be further tested this winter. It is hoped that these tests will produce good results, enabling the SALTAR unit to satisfy future requirements and goals, and to become an important tool for the Public Roads Administration in achieving better friction on the roads in the years ahead.

0.9 Bare and dry

0.8 0.7

Bare and wet

Friction

0.6 0.5 0.4

Slush

0.3

Loose snow

0.2

Packed snow Wet Ice

0.1 0 0%

20%

40%

60%

80%

100%

% slip

Figure 1. Typical graphs describing the friction conditions for different road surfaces depending on the wheel slip.

12

Goals: • The Public Roads Administration will review their standards regarding friction and, if necessary, adjust these in accordance with today s cars and modern maintenance methods and techniques with stringent demands for traffic safety.

• Friction requirements must be based on specific measuring methods and tested friction measuring equipment.

• Friction measuring equipment will be used that will register and document whether the road surfaces and maintenance satisfy the friction standards.

• Friction measuring units must be calibrated against a standard and approved for use in control and documentation.

• The friction measuring units should not be expensive.

• Friction measuring units must be stable in use and give comparable results.

The SALTAR friction measuring device • Small in volume, but extremely robust and can work under the most difficult climatic conditions in winter, with low temperatures, as well as in wet and salty environments.

• Easy to use, can take continuous measurements without applying the vehicle s brakes and can measure at any speed.

• Can be attached to any type of vehicle as a trailer and can, if required, also be mounted on the frame in front of the rear wheels on salting trucks and snowploughs.

• Registers the friction on the road in five levels shown on a display, but can be connected to a computer or a data logger for presenting the friction coefficient and further processing of measured data.

• Can be connected to maintenance equipment for winter measurements, but is so robust that it can also be used during the summer for all types of measurement.

Article is written by chief engineer Arnulf Ingulstad, Road Technology Department, Norwegian Public Roads Administration.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

INSTITUTE OF TRANSPORT ECONOMICS (TØI), NORWAY

Developments of performance contracts in Norway Performance contracts instead of competitive tendering seem to be the new tool to make public transport more efficient in Norway. In these contracts the transport companies get more freedom and financial incentives to improve the service.

Two different performance contracts have been introduced in Norway in recent years, one in Oslo starting from 1999 and one in Hordaland County from 2000. There are also plans for similar contracts in the city of Kristiansand. In total, this will represent around 50 percent of the local public transport market in Norway, measured in terms of passengers.

Tendering is extensively used within local public transport in Scandinavia.

Around 80 percent of publicly served routes in Sweden and Denmark are open to tender. Norway is the exception, as the proportion of tendered services is below 10 percent. Nevertheless, there is no indication that public transport operations in Norway are less cost-effective than those in the other Scandinavian countries, and the level of subsidies is much lower (ISOTOPE 1997, JANES Urban Transport Systems 1999-2000).

On a national level, annual subsidies to the public transport companies have been

Competitive environment

The main reason for this development is extensive use of threatening competition, based on the experience from Sweden and Denmark and the possibility to use tendering in Norway. As from 1991, efficiency agreements have dominated the contract form in Norway. In these contracts, specific cost cutting or subsidy reductions have been agreed as a condition for the routes not to be tendered. An examination

Population

Subsidy rate %

Savings Mill (NOK) 1997

(1000)

1986 1997

Subsidy Net social1 Prod2 Cost3 Pass3

Oslo Bergen Trondheim Kristiansand Tromsø Sum

reduced by about NOK 1.2 billion (1997 level) in the period 1986—1997, a reduction of 42 percent in fixed prices (Norheim and Carlquist 1999). In a number of towns, public transport receives almost no subsidy at all (table 1). The reduced subsidies have placed the focus on cost-effectiveness and been passed on to passengers in the form of higher fares and reduced service levels, but also increased cost-effectiveness and productivity. In total is there a net social saving of NOK 166 million (1997 level), or about 28 percent of the total reduction in subsidies. There is a clear trend of reduced productivity gain over the latter part of the period, and a higher proportion of the cut in subsidies is passed on to the passengers. For the last period, 1991-97, are there no net social savings for the cities as a whole.

500 225 145 70 60

50 28 59 20 25

31 8 4 27 11

-353 -115 -120 10 -14

-141 +51 -84 +30 -21

-592

-166

Changes 1986-97 %

47 12 33 21 -2

-21 -13 -22 -22 0

-2 -24 -3 +2 -19

1 Net social savings is included the effect on fares, service level and increased car traffic 2 Productivity is defined as number of vehicle km per employee 3 Cost is defined as operating cost per vehicle km 4 Passengers is defined as number of passenger per inhabitant Table 1: Development of subsidies, cost and productivity in five different cities in Norway 1986-97 Source Norheim and Carlquist 1998 NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

of the effects of these efficiency agreements reveals that the public transport companies have achieved a cost reduction of between 1.5 and 4.5 percent annually, but that even companies which did not have specific agreements also experienced considerable increases in efficiency in the period. There appears to be broad consensus that the main weakness of the agreements is a lack of market/revenue incentives.

Simultaneously, the public transport system is dependent on a continuos and targeted product development in order not to lose its market share. An analysis of the development of public transport in the 10 largest urban areas in Norway revealed that if the services were maintained at current levels the number of passengers would decline by 1.6 percent per annum (Norheim and Renolen 1997). In Bergen the competition was even more severe with a declining market share of 2.4 percent annually if nothing was done with the services or frameworks for public transport.

This underline the importance of a subsidy contract, which provides the opportunity for a targeted product development.

Development of performance contracts

The Institute of Transport Economics has undertaken analyses of the new subsidy contracts for Oslo, Hordaland County and Kristiansand. The main objective of these studies was to examine the potential for developing an alternative subsidy model, compared to both the current model and traditional tendering procedures. Such a model could provide an incentive to develop more market-efficient and cost-effective public transport. The performance contracts imply that market initiatives including responsibility for long-term planning and product development will become the responsibility of the public transport companies, while stringent terms will be imposed by the authorities in respect of conditions and the achievement of goals. Commencing with the basic char13

PHOTO: HARALD AAS

INSTITUTE OF TRANSPORT ECONOMICS (TØI), NORWAY

Hordaland County Council in Western Norway has decided that the principle of quality contract should form the basis for all contracts relating to public transport.

acteristics of the public transport market, we have developed a subsidy model which both encourages increased readership as well as increasing the benefit for existing passengers.

The proposed contracts, both in the long and the short term, are dependent on the level of freedom in the contract. Those analyses which have been carried out suggest that it is possible to develop a subsidy contract that can combine commercial profit and socio-economic efficiency. The performance contract contains of the following elements: 1. Framework and quality demands for the authorities 2. Performance-related subsidies 3. Conditions for fulfilment of the contract 4. Proposals for transfer to new arrangements.

A transitional arrangement was intro14

duced for the public transport company in Oslo starting from 1999, and a full-scale contract is being introduced in Hordaland in 2000.

Transitional arrangements in Oslo

The City Council in Oslo has agreed to the main principles of this form of quality contract . As a first step in this direction, the City Council has approved a clause in the 1999 budget, whereby a specific proportion of the operating subsidy to Oslo Public Transport is to be performance-dependent. In addition to a fixed subsidy of NOK 472 million (1998 level), a performance-dependent bonus of NOK 37.3 million has been allocated dependent upon either increased route production or an increased number of passengers. This allocation amounts to about 8 percent of the initial subsidy level.

The City Council has attached particular importance to measures during the peak

period, based on the following criteria: • Increased frequencies during rush hours • Increased capacity (seat-kilometres) • Marketing costs • Increased total number of passengers In the contract, a bonus of NOK 1,000 is obtained for every additional departure in rush hours, NOK 0.5 per extra seat-kilometre, and NOK 4 per extra passenger. A further NOK 1.5 million has been allocated to extended marketing/start-up costs. The largest proportion of the bonus is related to total journeys (56%), while increased frequency and seating capacity are each apportioned a 20% bonus.

Production targets achieved

The effect of the performance bonus after one year is that the production, peak departure and seat-km targets were achieved, while the target for passengers was almost

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

INSTITUTE OF TRANSPORT ECONOMICS (TØI), NORWAY

bonus remains at NOK 4 per passenger, the new bonus of around NOK 3.3 million will represent a full passenger bonus.

It is important to emphasise that this transitional agreement is a first step, with the aim being to further develop the arrangement.

Hordaland County

On 25 March 1999, Hordaland County Council in Western Norway decided that the principle of quality contract should form the basis for all contracts relating to public transport in Hordaland as from 1 January 20001. In this respect, a quality contract signifies a performance contract, similar to the model developed in Oslo.

The authorities in Hordaland have defined a framework for the contract for minimum quality of service, which has to be satisfied. The framework for the contract comprises overall quality require-

ments with regard to price, service and accessibility that the authorities require to be included.

The following elements can be included within such a framework: • Net contract with income responsibility for the companies.

• Establishment of an ongoing travel quality survey used as a quality indicator and where the contract may be terminated or renegotiated if the indicator falls below 90 percent of the initial index. The surveys carried out in the first six months of 2000 provide the basis for the initial index.

• The operators are responsible for proposing new differentiated fares within an upper fare level.

• The companies are to be responsible for providing ongoing reports on central key statistics for the quality of the services every half-year concerning route

PHOTO: HARALD AAS

achieved. The total passenger increase was 4.2 million passengers (2.7 percent) and the total bonus NOK 33.3 million. There have been some reductions in the service lately indicating that we need to know more details about the cost and revenue for the company before we can evaluate the total effect of the contract. However, we can already conclude that this is a demanding contract where the company needs an indepth understanding of the demands for public transport.

Negotiations are currently (March 2000) being conducted between the operator and the company regarding new elements in the contract. This contract is still based on the 1998 passenger level, but a greater emphasis will be placed on passenger increase. In total, the new maximum bonus for 2000 will be NOK˚35˚million, based on NOK 30 million from the increase in passengers and NOK 5 million from a quality index. If the

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

15

INSTITUTE OF TRANSPORT ECONOMICS (TØI), NORWAY

kilometres, vehicle kilometres (basic service and extra peak services), frequency, regularity, reliability, patronage and revenue.

Performance-dependent subsidies

The contract between Hordaland County and the operators is a full scale performance contract, where most of the subsidies are based on the service level and the operators have greater responsibility for developing the service level and fare structure. Without going into all the details, there are a number of important principles and elements in the contracts we would like to emphasise. This is linked both to

the decentralised responsibility of market development and competitive pressure in the contract.

The performance-based subsidies are based on four elements: route km, vehicle hours for basic operation and extra in peak periods, and number of peak passengers. The level of the different elements varies between the companies due to differences in the cost and fare structure. The fixed amount is basically a scaling factor calculated to get the initial subsidy level for the initial service level produced. This is to some extent a transitional arrangement to ensure that the starting point of the contract is close to the

initial financial situation of the companies, and if they do not change the service level, the subsidies will remain unchanged2.

The companies are free to decide the optimal service level based on the revenue and output-based subsidies, according to their commercial profit. The agreement between the county authorities and the companies is initially valid for 4 years, but with a clause about index regulation and subsidy rates during this period of agreement.

Article written by B r d Norheim

Reports: (1) Norheim B rd and Carlquist Erik (1999) Market efficient public transport? An analysis of Developments in Oslo, Bergen, Trondheim, Kristiansand and Troms¿ TflI-report 428/1999 Language of report Norwegian with English summary (2) Norheim B rd and Johansen Kjell Werner (1998) Quality contracts —Alternative to competitive tendering in Oslo TflI-report385/1998 Language of report Norwegian with English summary (3) Norheim B rd and Renolen Heidi (1997) The Demand for Public Transport in Norway 1982-94 An analysis of the differences and development in the 10 largest cities TflI-report 362/1997 Language of report Norwegian with English summary (4) ISOTOPE (1998) Improved structure and organisation for urban transport operations of passengers in Europe Final report EC 1998 (5) Johansen Kjell W and Norheim B rd (1999) Quality Contracts- Discussion of output-based subsidy contracts TflI-report 455/1999 Language of report Norwegian with English summary (6) Carlquist Erik et al (1999) Quality Contracts in Hordaland County An evaluation of alternative contract forms TflI-report 452/1999 Language of report Norwegian with English summary (7) Carlquist Erik and Johansen Kjell W (1999) Local Public Transport Systems- Financial and Organisational Frameworks in Norway and Abroad TflI-report 451/1999 Language of report English

16

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

What roundabout design provides the highest possible safety?

In Sweden, as in several other countries, the number of roundabouts is on the increase.

According to a study recently carried out by the VTI, roundabouts with a maximum permissible speed of 50 km/h are typically safer for motorists than grade-separated intersections. Single-lane roundabouts can be just as safe for cyclists as other types of intersection, and for pedestrians they are perhaps safer than any other type.

At the beginning of the 1980s, there were only about 150 roundabouts in Sweden.

Now, there are almost 1000, and within the next 10-year period the number may rise to at least 1500.

It has been known for some time that a roundabout is a very safe type of intersection, eliminating most serious injuries involving motor vehicle occupants. As far as the safety of vulnerable road users, i.e. cyclists and pedestrians, is concerned, earlier results were not as unambiguous and positive. There were worries, especially in regard to cyclists, that safety may even deteriorate as a result of introducing roundabouts. There were also different views on detail design and whether cyclists should be allowed to mix with motorists on the carriageway or segregated on separate cycle paths with special cycle crossings adjacent to the pedestrian crossings near the roundabouts.

In an endeavour to improve the state of knowledge in this field, VTI was commissioned by the Swedish Road Administration to study accident and injury risks at roundabouts with different layouts in different traffic environments.

Additional data on the number of accidents and the number of injured over the period 1994—1997, as well as the number of vehicles, cyclists and pedestrians passing through the roundabouts, were later collected for some of the roundabouts surveyed. Three studies were subsequently performed; speed analysis in 536 roundabouts, analysis of safety of cyclists and pedestrians in 72 roundabouts and analysis of motorist safety in 182 roundabouts.

In conjunction with the survey, one run was made to measure the speed towards, through and out of each roundabout. The run was performed slightly more aggressively than would be the case with a normal motorist. The objective was to obtain indirect measures of the factors (layout, regulation, etc) affecting safety at roundabouts.

Separate relationships for speed when entering, circulating in and exiting the roundabout were estimated using multiple non-linear regression analysis: Speed = a x b X1 x c X2 x d X3 x e X4 x where X1, X2, X3, X4 etc are dummies (0/1 variables) indicating different properties.

The regression coefficients b, c, d, e show the estimated effects for each property.

The results obtained indicate the following: • Speed is higher at a speed limit of 70 km/h than at 50 km/h, and higher when the general speed limit is higher than the local limit.

• Speed is on average higher if the roundabout has multiple lanes.

• Speed is lower if the radius of the central island is 10-20 m than if it is smaller or larger.

• Provision of additional trafficable area around the central island has no effect on speed.

• Flaring the approach to the left reduces speed into and through the roundabout.

The results appear in all respects to be very reasonable, and have also been found to be in very good agreement with the relationships obtained in the accident and injury analyses described below.

More accidents to cyclists if there is more than one lane

The pedestrian and cyclist (PC) analysis comprises 72 roundabouts, most of those which in the preliminary survey were judged to have at least 100 cyclists per day.

At these 72 roundabouts, there were a total of 67 police reported accidents involving cyclists, 58 of which resulted in inju-

Three different studies

According to a preliminary survey, there were about 700 roundabouts in Sweden in autumn 1997. All but 50 of these were surveyed in the field to classify them with respect to geometric design, speed level, etc.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

17

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

Single-lane lane roundabout with centre island radius 10-20 m reduces speed

ries, during the period 1994—1997.There were no fatalities in the accidents, but 10 persons had serious injuries and 48 slight injuries.

At 52 of the roundabouts, there were no accidents involving cyclists. Eight roundabouts account for 48 accidents to cyclists, with 9 serious and 32 slight injuries. Four of the roundabouts with the most accidents have two lanes, and at three roundabouts cyclists use the road without separate bicycle crossings.

Statistical processing entailed simple tabulations, correlation and regression analyses, etc. The factor which was initially found to have the greatest effect on the number of accidents involving cyclists, apart from the number of motor vehicles and cyclists, was the number of lanes. From then on, the accident material was therefore

treated separately for two-lane and singlelane roundabouts.

The observed numbers of accidents and injured are compared to predicted values according to previous VTI studies of conventional intersections, including those controlled by signals.

As can be seen in Table 1, at the two-lane roundabouts the total observed numbers of accidents and injuries are more than twice those predicted. For the single-lane roundabouts, on the other hand, there is very good agreement between the observed numbers of accidents and injuries and the predicted numbers.

The above prediction models for the number of bicycle (and pedestrian) accidents contain only the number of entering

motor vehicles and the number of crossing cyclists (and pedestrians) as explanatory variables. According to previous studies (of conventional or signal controlled intersections), these factors were found to explain almost the entire variation in the number of accidents, apart from the purely random variation.

According to the prediction models, for example, an intersection with 10,000 motor vehicles and 1000 crossing cyclists per day is expected to have approximately 0.2 bicycle accident annually, or in other words one bicycle accident every 5 years. For an intersection with an equal number of motor vehicles and 1000 pedestrians, the number of accidents involving pedestrians is expected to be half the above number.

Observed number of Predicted number of No. of lanes No. of Veh. Cycles No of Acci- Injury Fatali- Serious Light Acci- Injury Fatali- Serious Light round- per day, per day, accident dents acci- ties injuries injuries dents accities injuries injuries abouts average average years dents dents 1 lane ≥ 2 lanes

58 14

11066 23375

1075 1397

163 53

28 39

24 34

0 0

5 5

19 29

35,7 19

30,2 16

0,1 0,3

9,6 5

19,5 9,9

* At least one leg with a PC crossing has two lanes

Table 1. Bicycle accidents at single-lane and two-lane* roundabouts.

18

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

It may be noted that no bicycle accidents have occurred at roundabouts which have the lowest number of both motor vehicles (< 10,000 per day) and cyclists (< 1000 per day). See Table 2.

Flow class

Veh. per day, average

Lower flows 6463 Larger flows 13488

Safe for pedestrians, particularly at single-lane roundabouts

is about 1.7 times the number travelling through them on the carriageway. If we construct the risk index as No. of accidents per the sum of the products no. of motor vehicles x no. of cyclists , the risk

Cycles per No. of day, roundaverage abouts

No. of accident years

Observed No. of accidents

Pred.

No. of accidents

476 1390

50 113

0 28

4.9 30.7

20 38

Table 5 (see page 20) sets out the observed and predicted numbers of accidents and injuries involving pedestrians. Overall, during the investigation period there were 15 pedestrian accidents with one killed, 2 seriously injured and 10 slightly injured.

Twelve of the accidents occurred at the two-lane roundabouts and only 3 on single-lane roundabouts.

For the two-lane roundabouts, there is almost perfect agreement between the observed and predicted values. For the single-lane roundabouts, however, the observed values are substantially (3-4 times) lower than those predicted.

The results suggest that roundabouts pose no problems for pedestrians compared to conventional or signal controlled intersections. The results also clearly show that for pedestrians single-lane roundabouts are much safer than two-lane roundabouts.

Table 2. Bicycle accidents at single-lane roundabouts with different flows.

Fewer bicycle accidents if radius of central island > 10 m and if there are special bicycle crossings

is about 2.5 times greater for cyclists on the carriageway than for cyclists on a bicycle crossing.

Regression analyses concerning single-lane roundabouts indicate that a central island radius (including any additional trafficable area around the central island) >10 m is most beneficial for reducing the number of bicycle accidents. This is also shown

by Table 3. In addition, roundabouts with

Location of bicycle crossings?

An issue that is often discussed is what distance is desirable between the bicycle crossing and the roundabout itself. The study provides no answer to this. According to earlier VTI studies concerning con-

Accidents involving motor vehicles Radius of central island

Veh. per day, average

< 10 m > 10 m

9464 12782

Cycles per No. of day, roundaverage abouts 1050 1101

30 28

No. of accident years

Observed No. of accidents

Pred.

No. of accidents

87 76

19 9

18.4 17.2

During the period 1994—1997 (a total of 563 accident years) the 182 roundabouts studied had a total of 456 police reported accidents involving motor vehicles only, of which 88 (19%) resulted in injuries.

There were no fatalities in these accidents, but 17 persons (15%) sustained serious injuries and 93 (85%) slight injuries.

The material is dominated by roundabouts in the half-central or outer areas of towns. They typically have four legs and a posted speed limit of 50 km/h, but there are also about twenty roundabouts in the countryside where the speed limit on some approaches is 70 km/ h. Half the roundabouts have more than 10,000 entering motor vehicles per day.

The largest number is 44,500. More than 10 roundabouts have more than 25,000 entering motor vehicles per day.

Table 3 Bicycle accidents at single-lane roundabouts of different radii.

a central island radius greater than 10 m have both a lower number of bicycle accidents per year and a lower bicycle accident rate (number of bicycle accidents per million crossing cyclists).

The regression analysis indicates that for cyclists it is safer to bypass a roundabout on a bicycle crossing than to travel on the carriageway. Nine of the 58 single-lane roundabouts can be characterised as cyclists mainly on the carriageway . These nine roundabouts account for 15 of the total 28 accidents (see Table 4). These raw numbers also indicate that cyclists face a greater risk when travelling through a roundabout on the carriageway.

Overall, at single-lane roundabouts there were 18 accidents involving cyclists on the carriageway and 10 accidents with cyclists on a bicycle crossing. At the same time, the total number of cyclists bypassing the roundabouts on a bicycle crossing

ventional or signal controlled crossings, however, there are indications that a distance of approximately 2-5 m might be the best. The entering motorist would then as a first stage be able to pay attention to cyclists on the crossing, and in the second stage he would have space beyond the bicycle crossing to give way to vehicles on the roundabout. However, this should be investigated further.

Cyclists travelling on

Veh. per day, average

Cyclists per day, average

No. of roundabouts

No. of accident years

Observed No. of accidents

Pred.

No. of accidents

Bicycle crossing 10478 Carriageway 14268

1011 1422

49 9

134 29

13 15

26.9 8.6

Table 4. Bicycle accidents at single-lane roundabouts with cyclists on bicycle crossings and on the carriageway.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

19

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

Traffic volume and speed have the greatest influence on accidents

The factors which usually have the greatest influence on the numbers of accidents and injured, given a certain type of intersection, are the volume of traffic and the speed of vehicles. When it comes to accident rate and injury rate (numbers of accidents and injured per million entering vehicles) the volume of traffic obviously has less effect since these indices are already normalised with respect to the volume of traffic.

The regression equations drawn up for accident rate and injury rate use the factors

eral speed limit. It should be noted that the safe travelling speed through a roundabout may be lower than the posted speed limit.

Recommended speeds are not posted in Sweden.

Radius of 10-25 m best?

The introductory simple tabulations showed that the factors most likely to influence accident rate and injury rate are speed, number of legs, number of lanes and radius. Here, the term radius denotes the radius of the central island itself, including any additional trafficable area

greater than about 25 m, are the most dangerous, but the pattern is fragmented. The reason seems to be that the larger the radius, the more often there is a speed limit of 70 km/h and two lanes. It is likely that there is an optimum situation with the lowest accident and injury risks for radii somewhere between 10 and 25 m. A shorter radius than 10 m often gives a straight driving path for straight through traffic with potentially high travel speeds. A very long radius also gives a straighter path than a moderate radius.

Observed number of Predicted number of No. of lanes No. of Veh. Pedestrians No. of Acci- Injury Fatali- Serious Light Acci- Injury Fatali- Serious Light round- per day, per day, accident dents acci- ties injuries injuries dents accities injuries injuries abouts ave. ave. years dents dents 1 lane ≥ 2 lanes

58 14

11066 23375

486 1338

163 53

3 12

2 10

0 1

0 2

2 8

11,2 10,4

9,3 8,9

0 0,1

2,3 2,8

5,7 6

* At least one leg with a PC crossing has two lanes

Table 5. Pedestrian accidents at single-lane and two-lane* roundabouts.

total number of entering motor vehicles per day and the speeds measured when entering, circulating and exiting the 182 roundabouts surveyed.

The results are in very good agreement with previous investigations. The number of accidents is directly proportional to speed, while the number of injured has an even greater and more quadratic relationship with speed. The accident and injury rates exhibit a slight positive relationship with total traffic. The fact that this relationship is weak for roundabouts compared to intersections is also in good agreement with previous results.

The lower the speed limit, the lower the risk and the number of injured per accident

Table 6 further bears out the significance of speed for accident risk and the number of injured per accident at roundabouts. In the table, a breakdown has been made by the highest local speed limit on the roundabout and also the highest general speed limit within 600 m of the roundabout. It is evident from the table that the risk of injury and the number of injured are greatly influenced by both the local and the gen20

around this. The results indicate that the accident and injury rates are higher if the speed limit is 70 km/h, if there are four legs, if there are two lanes and if the radius is very large.

Analyses at greater depth in which several factors were considered at the same time have shown that it is primarily the factor 50 or 70 km/h that explains the differences in risk, and possibly also the factors 3 or 4 legs and the number of lanes.

As regards the radius, there appear to be indications that the very largest radii, Speed limit, Speed limit, local general

New prediction model

Several approaches were tested in constructing new prediction models or standard values for accident rate, number of injured per accident and injury rate. The model finally chosen for the accident rate is as follows: Predicted accident rate = 0.1353 x 0.86 3leg x 1.88 speed70 x 1.20 2lanes where the dummy variables 3leg, speed70 and 2lanes are equal to 1 if there are three legs (0 if there are four), if the maximum

No. of roundabouts

Accident rate*

No. of injured per accident

Injury rate**

50 50 50

50 70 90

81 49 20

0.15 0.17 0.18

0.12 0.23 0.43

0.018 0.039 0.078

70 70 70

70 90 110

14 16 2

0.17 0.27 0.46

0.28 0.38 0.35

0.047 0.103 0.162

182

0.17

0.24

0.041

Totals

* Number of accidents per million entering vehicles ** Number of injured per million entering vehicles Table 6. Accident rate, number of injured per accident and injury rate for different speeds.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

local speed limit is 70 km/h (0 if 50 km/h) and if there are two lanes on the roundabout (0 if there is one). According to estimated regression coefficients, the accident rate is reduced by about 14% if there are three legs, increased by about 88% if the maximum local speed limit is 70 km/h, and increased by about 20% if there are two lanes on the roundabout. The speed factor is the only significant variable. The coefficient of determination with respect to the number of accidents is below 55%.

Alternatively, the following model, which is somewhat more complicated for practical purposes but has a somewhat better coefficient of determination, might be used: Alt. predicted accident rate = 3legs speed70 loclow 0.1130 x 0.92 x 1.84 x 1.40 2lanes x 1.17 The dummy variable loclow is equal to 1 if the maximum general speed limit within 600 m of the roundabout is higher than the local maximum speed limit. According to this prediction model, the accident rate increases by about 40% if the general speed limit is higher than the local limit.

Designing a prediction model for injury rate is much more difficult. The number of injured is only a quarter of the number of accidents, and also has a larger random spread in relative terms. In view of this, it

was decided to build a prediction model for injury rate in a slightly different way. The observed injury rates were quite simply fitted to a function of the predicted accident rate. The level was fitted to the function with a multiplicative factor and the dispersion with an exponential factor. The following results were obtained: Predicted injury rate = 1.6871 0.8178 x (Predicted accident rate)

per day, of which 30% is secondary road traffic. The general speed limit is assumed to be 70 km/h. The roundabout is assumed to have one lane. The roundabout is further assumed to have a local speed limit of 50 km/h, while the grade-separated intersection maintains a limit of 70 km/h.

In the second case, the comparison is made for an intersection in the countryside, with 5000 motor vehicles entering per day, of which 20% is secondary trafAlt. Predicted injury rate = fic. The general speed limit is assumed to 0.7215 x (Alt. Predicted accident rate) 1.6119 be 90 km/h. The roundabout is assumed The coefficients of determination for the to have one lane and to have a speed limit number of injured are just below, and just of either 50 or 70 km/h. The grade-separated intersection is assumed to have a above, 40%.

The predicted number of injured per ac- speed limit of either 70 or 90 km/h.

As seen in the above comparisons, cident is obtained by simply dividing the predicted injury rate by the predicted ac- from the traffic safety standpoint a cident rate — or by first predicting the roundabout is even somewhat better than number of injured and the number of acci- a grade-separated intersection, at least if it is assumed that a speed higher than 50 dents, and performing the division. km/h is not permitted locally on the Safer than a grade-separated roundabout. intersection For an average roundabout, with 10,000 motor vehicles entering per day Table 7 sets out a number of comparisons and 30% secondary road traffic, it may be between the expected numbers of accidents expected that there will be one police reand injured in one year at roundabouts and ported accident every 2 years, one injured at grade-separated intersections. person every 10 years and one seriously A comparison is first made for an interinjured person every 50 years — and pracsection with four legs in a town outer tically never any fatalities. area , with 10,000 motor vehicles entering

Type of intersection

General speed limit

Local speed limit

Motor vehicles entering per day

Percentage Pred. of no. of secondary accidents road traffic

Roundabout* Roundabout** Grade sep.

70 70 70

50 50 70

10 000 10 000 10 000

0.30 0.30 0.30

Roundabout* Roundabout* Roundabout** Roundabout** Grade sep.

Grade sep.

90 90 90 90 90 90

50 70 50 70 70 90

5 000 5 000 5 000 5 000 5 000 5 000

0.20 0.20 0.20 0.20 0.20 0.20

Pred. no. of injured

Pred. no. of seriously injured or killed***

0.49 0.58 0.51

0.10 0.13 0.16

0.015 0.020 0.032

0.25 0.46 0.29 0.53 0.23 0.26

0.05 0.15 0.07 0.18 0.08 0.10

0.008 0.022 0.010 0.027 0.016 0.020

VTI Meddelande 864 and 865, 1999 Ulf Br de and J r gen Larsson Photographs: Cover illustrations for the two VTI Bulletins

* Single-lane roundabout ** Single-lane roundabout (alternative prediction model) *** For roundabouts, the percentage of seriously injured or killed is assumed to be 15% Table 7. Comparison between expected numbers of accidents and injured in one year at a roundabout and at a grade-separated intersection.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

21

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

Optimisation of clothoid lengths simulations of dynamic vehicle reactions on horizontal curves with track irregularities This report was presented at the international conference Railway Engineering — 2000 held at the Commonwealth Institute, Kensington, London on 5th-6th July 2000, and contains an extension of the alignment studies in VTI Rapport 424A. In this study, different combinations of radii and clothoid lengths (lengths of transition curves), within predefined terrain corridors are evaluated .

Although the simulations are focused on the influence of the (design) alignment, the track model includes lateral, vertical, cant and gauge irregularities measured with the Swedish track recording vehicle, STRIX. Two levels of irregularities have been used; one set of data has been sampled from a track for 165 km/h operation with tilting trains, and another set of data from track for 190 km/h tilt operation. Both sets of

data represent poor track, close to the maintenance limits.

An X2000 power car, a tilting X2000 coach and a non-tilting Eurofima coach have been studied. Vehicle reactions have been simulated in the time domain with non-linear vehicle/track models in the GENSYS computer code, using 46 degrees of freedom for each vehicle. The tilting train is simulated at a speed of 250 km/h and the Eurofima coach at a speed of 200 km/h. Hence, the track irregularities are believed to represent a worst case.

Vertical wheel/rail forces, track shift forces and wheel climbing ratios have been used in boundary conditions, with limits suggested by UIC and CEN. Comfort on Curve Transitions (PCT) has been used as the object function, since it takes into account the combined effects of lateral acceleration, lateral jerk and roll velocity of the vehicle body.

The results from the present study confirm earlier observations and conclusions published in VTI Rapport 424A. The clothoid lengths of a single curve that minimise comfort disturbances (PCT) depend on: • The angle between the adjacent tangent tracks. A larger angle results in longer optimal clothoids.

• The position of the binding obstacles. A binding obstacle in the middle of the curve normally results in a longer optimal clothoid than a binding obstacle at the end.

• Vehicle characteristics. Tilting vehicles result in longer optimal clothoids than non-tilting vehicles.

The PCT values are generally higher for track with irregularities than without.

However, the clothoid lengths that minimise PCT are in most cases the same, independent of the track irregularities.

In planning new railways and realigning existing ones, the following two aspects should be taken into consideration: • The optimal alignments are not the same for tilting vehicles compared to non-tilting vehicles. The optimal transition curves are longer for tilting trains.

• For tilting trains, the desire for long transition curves (due to passenger comfort) must be balanced against the desire for large curve radii (due to wheel/rail forces).

Author: Series: Language:

22

Björn Kufver VTI särtryck/reprint 335 English

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

SWEDISH NATIONAL ROAD AND TRANSPORT RESEARCH INSTITUTE (VTI)

Ride comfort and motion sickness in tilting trains Human responses to motion environments in train ans simulator experiments This thesis presents a systematic study of human responses to different motions and strategies of car body tilt control regarding ride comfort, working/reading ability and motion sickness on high-speed tilting trains. Experiments with test subjects were performed in a tilting train on curved track as well as in a moving vehicle simulator. The study is multi-disciplinary, combining knowledge and methods from the fields of railway technology, human factors and vestibular science.

The main experiment in a tilting train was performed with about 75 seated test subjects, mainly students from Link ping University, making three test runs. In total, these subjects participated in about 210 individual test rides, each with a duration of about 3 hours. Additional tests on comfort disturbances with pushbutton technique have been reported in the project. The simulator experiments used a total of about 75 subjects, making some 320 test rides each of about 30 minutes duration.

Test motions consisted of combinations of horizontal (lateral) acceleration and roll acceleration, together with either roll or horizontal acceleration. Rate of change of horizontal acceleration (jerk) and roll velocity were of the same order of magnitude as in a tilting train environment, but horizontal acceleration alone was about half the magnitude. Horizontal and vertical vibrations from a tilting train were added to the test motions, and train seats and interior train noise were also introduced to create a train feeling .

Test designs and methodology have been developed during the course of the experiments. The test subjects answered questionnaires, four times per test run in the train experiment and every 5 minutes in

the simulator experiment. The investigated variables were: estimated average ride comfort, estimated ability to work or read, and occurrence of symptoms of motion sickness (dizziness, nausea and not feeling well). Lateral and vertical accelerations together with roll motions were monitored and recorded for later evaluation.

Results from the train experiments show that the estimated average ride comfort was about 4 on a 5-degree scale, which indicates good . Results also show that a reduced tilt compensation of the lateral acceleration while curving, together with a reduced tilt velocity of the car body, reduces the provocation of motion sickness.

However, a reduction in tilt compensation may produce an increased number of comfort disturbances due to lateral acceleration in the car body.

Regression analysis shows that motion doses from roll acceleration may be used to predict the incidence of motion sickness.

The simulator experiments show that the primary sources of provocation of nausea and motion sickness are the motion doses from roll and lat-

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

eral acceleration in the horizontal plane.

The study proposes a hypothesis and a model of provocation of motion sickness. It is shown that motion sickness has a time decay, or leakage. A model for this leakage is proposed.

The determinative types of motion for provocation of nausea and motion sickness in tilting trains are identified and future tilting train and/or simulator experiments are proposed in order to further investigate their influence.

Doctorial thesis by Johan F rstber g, VTI

23

Annotated reports from the Danish Road Directorate (DRD)

Evolution and harmonization of evenness evaluation techniques During the past decade, the operational effectiveness of profiling road and airfield pavements has increased considerably due to the introduction of non-contact sensors, such as lasers and other optical equipment.

The sensors are capable of measuring the distance from a known reference point to the pavement surface at high speed. Modern data acquisition systems and computers make it possible to process and store the large amounts of data produced by these devices. This technology can measure both longitudinal and transverse pavement profiles at ordinary traffic speeds.

Thus, it is possible to avoid the road clo-

sures or traffic hold-ups earlier caused by slow-moving measuring vehicles.

Monitoring of evenness characteristics on road networks has become an easier task through the introduction of these high-speed systems. As a result, this work is now being carried out in a more systematic manner, which facilitates reliable pavement management.

The paper highlights the evolution in profiling techniques from static measurement techniques to response type devices and modern profiling techniques. The necessity of harmonization of modern profiling equipments and the importance of

reliable network monitoring of pavement evenness in connection with the Danish Pavement Management System BELMAN are also covered by the report.

The report is available in electronic form (PDF) from the Danish Road Directorate s web site www.vd.dk under Publications.

Title: Evolution and Harmonization of Evenness Evaluation Techniques Author: Bjarne Schmidt Series: Report 94 Language: English

Development of improved mechanistic deterioration models for flexible pavements This report includes a paper presented at the 4th International Conference on Managing Pavements. The paper describes a pilot study in Denmark with the main objective of developing improved mechanistic deterioration models for flexible pavements based on an accelerated full scale test on an instrumented pavement in the Danish Road Testing Machine.

The reader will gain an insight into the work necessary for successfully establishing mechanistic deterioration models based on Accelerated Load Testing (ALT) complemented with laboratory testing.

The work described is part of the International Pavement Subgrade Performance Study carried out under contract to the Federal Highway Administration, TurnerFairbank Highway Research Center. U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory were assigned overall management of the study, which should lead to input to the AASHTO 2002 Design Manual.

The report is available in electronic form (PDF) from the Danish Road Directorate s web site www.vd.dk under Publications.

Title: Development of improved mechanistic deterioration models for flexible pavements Author:

Hans Jørgen Ertman Larsen, Per Ullidtz Series: Report No. 89 Language: English

Speed management in urban areas The Road Directorate has published a new handbook in Danish on speed management for urban areas. The handbook describes why - and how - speed management should be used to improve safety, security and the environment in urban areas.

24

The report is also available in electronic form (PDF) from the Danish Road Directorate s web site www.vd.dk under Publications.

Title: H ndbog i hastighedsplanl gning for byomr der .

Author:

Poul Greibe, Puk Kristine Nilsson and Lene Herrstedt.

Series: Report No. 194 Language: Danish

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

Annotated reports from VTT, Communities and Infrastructure, Finland

Household activity and mobility in a suburb of Helsinki MA The main objective of VTT s recent activity and mobility research is to determine the effect of people s daily and weekly activities on their travel patterns. The focus is on households activities and household members roles in activity scheduling, in addition to individual persons activities. The idea is that people s travel behaviour is greatly influenced by their social environment, in particular by people with whom they share their daily life.

Factors influencing the decision by a person or household to take part in a certain activity can be grouped as follows: structural factors (e.g. housing, children s day care and transport services), social factors (e.g. household members and their roles), and personal factors (e.g. attitudes, interests, education and profession). In this study, the main interest was on the social factors determining people s activity and mobility.

The research was carried out as a postal questionnaire addressed to entire households, including children aged at least 10.

The questionnaire was divided into three parts: background information on the household and its members, individual

Wednesday, family with one child activity other activity pick up etc. recreation visit (friend, neighbour) hobby other business shopping work related school / study work home 4

6

8

10

father mother daughter

12

14

16

18

20

22

24

time

Figure 1. Example of a household activity path.

activity diaries for one week and a detailed one-day activity/travel diary. The research was carried out in three suburbs in Helsinki MA. The main categorisation used in the analysis was grouping of households into nine groups according to their structure. As an example of the results, Figure 1 represents households daily routines displayed

as activity paths on a time-activity axis of coordinates.

Author: Virpi Britschgi Series: Contractor Report 547 Language: Finnish

Price Elasticity of Public Transport in Helsinki MA The aim of the study of price elasticity of public transport in Helsinki MA, which was funded by the Metropolitan Area Council, was to determine the effects of small fare changes on demand, i.e. the number of public transport users. The research was limited to regular tickets for adults, both regional tickets and the cities internal tickets. However, if changes in the entitlement to a concessionary pass had affected the demand for regular tickets, this was taken into account.

The research period covered the years 1987-1997, i.e. the period since the introduction of the regional ticket. During the period, new ticket types, such as daytime tickets, exclusive tram tickets, single tickets purchasable in advance and team tickets, were introduced.

The main methodological effort focused on determination of elasticities using time series. The other methods tested were the logit model and a before-after survey. The research was complemented with a literature review of research on price elasticities in other EU countries.

In view of the data available, ETLA s consumption model was chosen for estimation. The 30-day travel pass was regarded as the most important ticket type.

For the other ticket types, price elasticities were defined, but were found to be more diffuse owing to the interchangeability of these ticket types in practice. In addition, especially for ticket types sold freely on board vehicles, it is impossible to determine the exact number used by citizens themselves and that by visitors.

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

For the regional 30-day travel pass, six different models were estimated, of which the best model in terms of its variables gave a short-term elasticity of -0.32. A short-term elasticity of -0.25 was obtained from the best of three models estimated for the internal 30-day travel passes.

Author:

Tuuli Järvi-Nykänen, Heikki Kanner, Virpi Britschgi, Anthony de Carvalho, Kari Alho Series: Tuuli Järvi-Nykänen, Heikki Kanner, Virpi Britschgi, Anthony de Carvalho, Kari Alho Language: Finnish with English abstract

25

Annotated reports from the Institute of Transport Economics (TØI), Norway

Road pricing in Oslo: Effects for the travellers

It can be argued that road pricing is an effective and attractive form of taxation. The picture is from the toll ring in Oslo.

Marginal cost road pricing has the double effect of discouraging congestion and raising public revenue. To the extent that public funds are a scarce resource, the latter effect may be well the more important as seen in an economic efficiency perspective. This is at least the case in a less heavily congested city like Oslo, the capital of Norway.

This is one of the main conclusions in the TflI report Road pricing in Oslo: Effects for the travellers . The conclusion, however, depends on how the road pricing revenue is used. If it is used to step down distortionary taxation somewhere else in the economy, or to extend the supply of a public good for which the willingness-to-

larger than zero, the benefit of marginal cost pricing is very substantially reduced.

In the project first and second best marginal cost road pricing is studied by means of the RETRO model for the greater Oslo area. Special emphasis is put on equity effects, as described by changes in the Lorenz curve or in the Gini coefficient, both of which are defined in terms of household income per consumption unit.

The economic efficiency of marginal cost pricing appears quite sensitive to assumptions regarding the shadow value of public funds, much larger gains being calculated under a 0.25 value of the shadow price than when the shadow price is zero.

The income distribution impact of marginal cost road pricing is generally unfavourable, not so much because lower income groups have their accessibility reduced, but because they end up paying a larger share of their income in road charges than do families in the upper income brackets.

This unfavourable income distribution effect may, in prinicple, be neutralised if the revenue is redistributed to the consumers in the form of a poll transfer. But in this case a major part of the economic benefit, due to the provision of additional public funds, will most probably be wasted.

pay exceeds the marginal cost of production, then a ˙double dividend¨ accrues. If, on the other, the revenue is redistributed to the private sector in a way that does not improve the incentive structure faced by economic agents, there is no extra dividend to be accounted for.

The use of a non-zero cost of public funds implicitly assumes that a double dividend somehow does arise.

A bit simplified, one might say that in Authors: Lasse Fridstrøm, Harald Oslo, second best marginal cost pricing is Minken, Arild Vold socially profitable first and foremost beSeries: TØI report 463/1999 cause it is — we assume — an attractive form Language: Norwegian with English of taxation. If, on the other hand, the marsummary ginal opportunity cost of public funds is not

Assessment of annoyance caused by vibrations in dwellings from road and rail traffic. Proposal for a Nordtest Method Socio-vibrational studies for assessing annoyance due to vibrations are to designed and conducted differently by different researchers and in different countries. This lack of standardisation is an important obstacle for authorities and researchers wishing to profit from research and results in other countries or who wish to perform analyses on extended data sets.

26

This report describes a project to to establish a common Nordic Method for assessing annoyance due to vibrations in dwellings from road and rail traffic in socio-vibrational and social surveys.

Important common aspects and differences between a survey in Sweden 1992 and a socio-vibrational survey in Norway 1998 were summarised and evaluated as a

departure point for a first draft proposal.

After discussions the proposal has been revised. The finalized proposal is included as part 2 of the report.

Authors:

Ronny Klæboe, Evy Öhrström Series: TØI report 458/1999 Language: English

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

Annotated reports from the Institute of Transport Economics (TØI), Norway

Wintertime speed limits An experiment with reduced wintertime speed limits in parts of Norway are found to have a strongly significant effect on speed, especially when the road is not covered by snow or ice. When the road is covered by snow or ice, however, road users voluntarily adjust their speed, regardless of the speed limit. Reduced wintertime speed limits are found to have a larger effect on speed the higher is the initial speed level on a given road link.

The experimental study on temporarily reduced wintertime speed limits described in this report was carried out during the winters of 1998 and 1999 and involved a number of different road links in rural Norway. The aim has been to assess the effect of reduced wintertime speed limits on speed, under varying conditions.

A data set consisting of 16 786 observations was gathered on hourly mean and 85th percentile speed (dependent variables), as well as on traffic volumes, road surface conditions, luminosity and visibil-

When the road is covered by snow or ice, road users voluntarily adjust their speed, regardless of the speed limit.

Authors:

ity (control variables). By means of a BoxCox regression model, the partial effects on speed of reduced wintertime speed limits, as well as of the control variables, were estimated.

Arild Ragnøy, Lasse Fridstrøm Series: TØI report 462/ 1999 Language: Norwegian with English summary

Mid-term Review of the Tanzania Road Sector Programme - Final Report The current Road Sector Programme in Tanzania as supported by NORAD focuses on institutional strengthening of the Ministry of Works and institutional co-operation between the Ministry and the Norwegian Public Roads Administration.

The institutional reform process in the road sector in Tanzania is fairly advanced and a new semi-autonomous roads agency is scheduled for operation July 2000. The

Mid-term review concludes that the programme is running well. Progress is mainly in accordance with plans and programme effectiveness and sustainability is acceptable at mid-term.

The institutional co-operation has been useful in building up competence in Tanzania. It is recommended that the programme continue as planned with emphasis on implementation of preparations al-

ready made and on support to the new roads agency, now being implemented, in road sector management.

Authors:

Henning Lauridsen, Terje Assum, Immanuel N. Kimambo, Sønneve Ølnes Series: TØI report 476/2000 Language: English

Six years old children and crossing of roads. Results of a training experimet The health risk of children in traffic has changed little during the last 10 to 15 years.

To reduce the risk, more extensive and new countermeasures are needed. Children at early school age are mainly involved in road accidents as pedestrians and in particular when crossing the road.

This report describes and presents the re-

sults from an experiment of training six-yearolds. The age for starting school in Norway is 6 years. For training, a model representing various traffic environments was used.

The effectiveness of training was measured by comparing the performance on a before test and an after test in real traffic using a test group and a control group. The

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

results show that training improved the children s ability to select safe spots for crossing the road.

Authors: Alf Glad, Kari Midtland Series: TØI report 473/2000 Language: Norwegian with English summary

27

Annotated reports from the Norwegian Public Roads Administration

The Fourth Symposium on Strait Crossings - call for papers In Norway you will find the deepest and the longest sub-sea road tunnels in the world. Norway has also built the longest cantilever bridge in the world with a main span of 301 m, located not far from Bergen.

Two floating bridges, longest of their kind without side anchors, are in operation.

There is, however, a great need for additional strait crossings and for new and improved solutions, both for fixed links and ferries. The economic, environmental and social aspects are all of great importance, and must be given equally high attention.

The Symposium will take place in Bergen, Norway, on 2-5 September 2001.

The city is located close to several major strait crossings, including the Triangle Link Project which will be completed in April 2001. Bergen is an old and beautiful city, playing an active role in Norwegian trade and shipping activities for centuries, and has been appointed one of the EU s

towns of culture in 2000.

The Fourth Symposium on Strait Crossings will highlight the following areas: • Traffic safety • Environmental issues • Aesthetics • Social effects • Maintenance and operation • Life-cycle costs/Life-cycle assessment • Technical solutions.

Call for papers The call for papers addresses scientists, engineers and others working with or interested in those subjects covered by the Symposium. Authors are invited to submit an abstract of not more than 400 words (one page) typed in English.

Abstracts written in Microsoft WORD can be sent by e-mail or by post.

The abstracts should have been sent to the Symposium secretariat before 1 September but due to this magazine«s schedule of

publication, the dead line is extended to 31 October 2000.

Please send to (email is preferred) the following address: Strait Crossings Secretariat Attain: Secr. gen. Jon Krokeborg P.O. Box 8142 Dep N-0033 Oslo Norway E-mail: sc01@vegvesen.no You will find the abstract form and the necessary information on the Symposium s homepage on the Internet: http://www.straitcrossings.com/ Various technical tours/excursions will be included in the programme. The tours will visit interesting sites demonstrating different types of strait crossings, and will also include an element of tourism, taking the participants to some of the most scenic places in Norway.

Stone materials for road, airport and railroad construction In 1993, the working group Arbeidsfellesskapet LPG was created in order to carry out the project Quality of products from the crushed rock and gravel industry . The primary goal for the project was to document the properties of stone materials for use as roadbase in constructing roads, railways and airports, as well as contributing to greater knowledge and correct use of the materials. The results are shown in several project reports and a final report. The Board decided - in coop-

eration with the Public Roads Administration - to produce a handbook on the use of such materials, and a committee was established in 1996. The main parts of the handbook deal with the following subjects: frames and guidelines, materials and solutions, production and handling of materials, and laying and compaction.

Series: Handbook 223 Language: Norwegian

Inspection of bridges This book is intended to safeguard the goals for management, operation and maintenance of bridges in a satisfactory way. Demands on routine inspections of all bridges are set out. The aim of the handbook is to provide guidelines for inspection of bridges so that they are inspected in a reliable manner and are treated uniformly throughout the country. The results from the inspections will be entered in 28

BRUTUS, the bridge management system used by the Public Roads Administration, and will then form the basis for planning operation and maintenance. Safety of the bridge inspectors during the inspections is also discussed.

Series: Handbook 136 Language: Norwegian

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

Annotated reports from the Swedish National Road and Transport Research Institute

Evaluation of the 16-year age limit for driver licensing In September 1993, a reform was implemented in Sweden through which the age limit for practising car driving was lowered from 17 years to 16 years while the licensing age remained eighteen.

The purpose of lowering the age limit was to give the learner drivers an opportunity to acquire more experience as drivers before being allowed to drive on their own. The primary aim of this study was to evaluate the effect of the reform and the process of the introduction and use.

The effects were evaluated in terms of register studies of accident involvement, questionnaire based self-reported behaviour and attitudes and observed behav-

iour in controlled experiments. The process was analysed through register studies of participation, questionnaire studies and diaries covering the educational content and process as well as background data of participants and non-participants.

The results show that the change has resulted in a general reduction of accidents per 10 million km among young novice drivers with approximately 15 percent during the evaluation period of the two first years with a licence. Those who have utilised the possibility to start early have also practised more and thus gained more experience. There is, however, a certain amount of uncertainty concerning the causal effects since the evaluation has not

been carried out with an experimental design. The conclusion is, in spite of these remaining uncertainties, that the reform has been beneficial for the safety of novice drivers in Sweden. The results also suggest a potential for additional safety improvements if more of the young learner drivers can be brought to utilise the low age limit.

and from school and less often as vulnerable road users. The risk run by secondary school pupils was twice as high and that by middle school pupils 50 % higher, than that run by primary school and upper secondary school pupils.

Half the journeys to school were on foot or by cycle. Journeys on foot were seldom longer than 2 km, and cycle journeys seldom longer than 3.5 km. Girls walked

more often than boys, boys cycled more often than girls. The journey time to school was on average 8-9 minutes on foot, by cycle or by car. It was three times as long by bus. Much more time was spent in the traffic environment on the journey home than on the journey to school. It was also usual for the mode to be changed; for instance, pupils travelled to school by car and walked home.

The project is part of the VTI topic T raffic safety analysis . The investigation on journeys to school is based on a traffic safety survey (TSU92-) and on the official statistics relating to accidents reported to the police.

Authors:

N.P. Gregersen, H-Y. Berg, S. Dahlstedt, I. Engström, S. Nolén, B. Nygaard and P. A. Rimmö Series: VTI rapport 452 Language: Swedish with English summary

Journeys to school The risk of school pupils being injured in a traffic accident was 50 % higher during school time than during a similar period when it was not a school day.

Primary school and higher secondary school pupils ran the least risk.

To a large extent, this was due to the fact that these two groups, more often than the two others, middle and secondary school pupils, travelled by bus or car to

Author: Series:

Hans Thulin KFB&VTI forskning/research 30 Language: Swedish with English summary

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

29

Annotated reports from the Swedish National Road and Transport Research Institute

School transport in relation to traffic safety This study focused on traffic safety aspects in connection with school transport. A literature study was carried out. Observations were made while travelling on several trips and attending an education day. Accident data were also of interest.

Previously, contracts had been compared from the aspect of traffic safety. This called for an extension resulting in a larger study entitled T raffic safety aspects in the pur-

chase of transportation at public expense, with particular reference to school transport . The aim was to identify and value the traffic safety aspects concerned in order to minimise the total costs to the community. The first part involves literature from the 90 s, excluding vehicle-related aspects, and has been documented in this report.

The most comprehensive investigations were found in the American literature. The importance of driver education was pointed out and accident crash studies had resulted in high standards for school buses.

In Europe, the EC directives also emphasise the importance of the driver. In Great Britain, the case for changing the present system was examined in terms of discretionary policies, expenditure and extent and means of provision. The congestion resulting from the increasing number of pupils being driven to school by their parents was also discussed.

In Sweden, responsibility is put on the school boards and the municipalities. Coordination between different types of community-financed journeys is advocated, sometimes without any regard for the consequences in terms of traffic safety.

In the USA, accidents have been reduced to an extremely low level. In Sweden, the number of accidents related to school transport is still unknown.

Authors:

Gunilla Sörensen, Anna Anund and Peter Wretling Series: VTI meddelande 885 Language: Swedish with English summary

Elderly pedestrians and cyclists are a high risk group in urban areas The aim of this investigation has been to describe the exposure and injury risk situation of different age groups of pedestrians and cyclists in different traffic environments.

The investigation is based on exposure data from TSU92-, which is a questionnaire based traffic safety survey. TSU92has been in progress continuously since 1992, with questionnaires sent out daily to the country s population.

One of the findings has been that the risk run by pedestrians and cyclists of being seriously injured or killed was about 30% lower at signal controlled crossings than at non-signal controlled ones, and that the risk was considerably lower on straight sections than at junctions.

30

Compared with other age groups, elderly (65—84) pedestrians and cyclists had a much greater risk of being killed or seriously injured in traffic. This applied primarily in urban areas where two thirds of pedestrians who had been killed were 65 or older. The situation for the elderly was especially problematic when crossing a road or street. The risk that elderly pedestrians and cyclists will be seriously injured or killed when crossing a street in an urban area was 30 times as high as that for the age group that was at least risk, the 15— 44 age group. Generally speaking, the elderly crossed roads/streets less often than the other age groups. Unnecessary crossings were presumably not made at all, or it may be said that crossings were better

planned. Elderly pedestrians crossed the street at junctions more often than other groups, presumably because pedestrian crossings were generally sited at these points. The elderly cycled more often than other age groups in mixed traffic, i.e. they used cycle tracks to a lesser extent than the other groups. The same applied to grade separated and signal controlled junctions.

Authors:

Hans Thulin and Henrik Kronberg Series: VTI meddelande 886 Language: Swedish with English summary

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

Annotated reports from the Swedish National Road and Transport Research Institute

Forms of parking and engine heaters – data for describing exhaust emissions on engine start A considerable proportion of the total exhaust emissions by road traffic consists of excess emissions on cold starting.

In order that this proportion of emissions may be estimated, it must be possible to describe engine temperature on starting. In turn, engine temperature depends on forms of parking and the occurrence and use of engine heaters. In this study, based on a questionnaire survey, these conditions are investigated.

For temperatures below –10° C, 86 % use electric heaters when both heaters and electricity supply are available.

In the most northern parts of Sweden, where a supply of electricity is provided,

the occurrence of electric heaters is the same, 91 %, for parking both at work and near the dwelling.

On a national level, parking with provision of electricity supply is used by 30 % at work and by 50 % near the dwelling.

Author:

Ulf Hammarström and Per Henriksson Series: VTI meddelande 891 Language: Swedish with English summary

The use of steel wire barriers on motorways reduces the number of injured By commission of the Swedish National Road Administration, VTI has investigated the motorways on which steel wire barriers were installed in the central reserve during the period 1994—1997.

The investigation comprises both the be-

fore and after periods and just over 10 % of the vehicle mileage on motorways.

The number of accidents reported to the police in which only vehicles were involved has increased by 20 %. The number of injury accidents has however

decreased by 16 % and the number of injured by 11 %. On the other hand, it seems that the serious accidents occur to nearly the same extent even when the almost total absence of head-on collisions is taken into consideration. The number of fatalities has increased, to some extent because the number during the before period had been lower than expected.

The number of head-on collisions in which vehicles in both directions were involved constituted only 3 % of accidents during the before period; these were almost eliminated during the after period, when the steel barrier was installed in the central reserve.

Authors:

Göran Nilsson and Li Ljungblad Series: VTI rapport 442 Language: Swedish with English summary

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000

31

Questions concerning the content of the articles, or orders for the publications referred to, should be directed to the publishing institution, see addresses below.

REQUESTS FOR BACK ISSUES, AND NOTIFICATION OF ADDRESS CHANGES: Readers outside the Nordic countries: see Swedish address.

Readers in the Nordic countries: see addresses below.

Web site: www.vti.se/nordic

Denmark Helen Hasz-Singh Danish Road Institute P.O. Box 235 DK-4000 Roskilde Phone: + 45 46 30 70 00 Fax: + 45 46 30 71 05 E-mail: hhz@vd.dk

Norway Harald Aas Institute of Transport Economics (TØI) P.O. Box 6110 Etterstad N-0602 Oslo Phone: + 47 22 57 38 00 Fax: + 47 22 57 02 90 Order phone: + 47 22 57 39 13 E-mail: harald.aas@toi.no 32

Finland

Iceland

Kari Mäkelä VTT, Communities and Infrastructure P.O. Box 1901 FIN-02044 VTT Phone: + 358 9 45 64 586 Fax: + 358 9 46 41 74 E-mail: kari.s.makela@vtt.fi

Hreinn Haraldsson Public Roads Administration Borgartún 7 IS-105 Reykjavik Phone: + 354 563 1400 Fax: + 354 562 2332 E-mail: hrh@vegag.is

Norway

Sweden

Helge Holte Norwegian Public Roads Administration P.O. Box 8142 Dep. N-0033 Oslo Phone: + 47 22 07 39 00 Fax: + 47 22 07 34 44 E-mail: helge.holte@vegvesen.no

Sigvard Tim Swedish National Road and Transport Research Institute (VTI) SE-581 95 Linköping Phone: + 46 13 20 40 00 Fax: + 46 13 14 14 36 Order phone: + 46 13 20 42 24 E-mail: sigvard.tim@vti.se

NORDIC ROAD & TRANSPORT RESEARCH NO. 2 · 2000