The evidence of the impact of bus priorities, cycle lanes and traffic calming is covered in a separate section.

Nottingham Zones and Collar, UK

Display location

Context
The Nottingham Zones and Collar scheme was introduced experimentally in 1975-76 at the western sector Nottingham by Nottinghamshire County Council, and aimed to selectively delay non bus traffic where queues would cause least disruption to buses, thereby encouraging car users to transfer bus. The main scheme features were:

• Zone exit controls - traffic wishing to leave the two residential zones during the morning peak was restricted to certain exit roads most of which were controlled by traffic lights. The exit controls were designed to limit the rate at which traffic joined the main roads so that buses and other main road traffic could run more freely.
  
• Collar controls - traffic lights on each of six main radial roads formed a partial collar of control signals around the city. The signals were timed to limit the traffic entering, and particularly passing through, the inner city area. It was intended that the traffic flows during the morning peak hour should be reduced by an average of about 10% below the August 1973 levels.
  
• Bus priorities - to prevent bus passengers being delayed in the control queues, bus priority lanes were provided on the approaches to all the collar signals and to some zone exits. Exemptions from certain traffic regulation orders allowed buses to bypass delays at other zone exits by making turns banned to private vehicles or by using short sections of bus only street.
  
• Park and ride - travellers for whom the conventional bus services were inconvenient could travel by car to one of four specially provided peripheral car parks and from there continue their journey by coaches operating a frequent, limited stop service into the city.

The zone exit controls with their associated bus priorities and the collar controls, operated within the morning peak between 7:30 am and 9:30 am; certain of the bus lanes approaching the collar sites and other two bus lanes on the main roads operated also during the evening peak. Coach services to park and ride sites ran every 7.5 minutes during both peak periods, and it was possible to use nearby, normal service buses throughout the day.

Impacts on demand
Vincent and Layfield (1977) concluded that results showed the scheme failed to achieve its two main objectives - on the demand side discouraging travel by private vehicles, and on the supply side making travel by bus more attractive.
The main impacts on demand side are:

• Average numbers of vehicles passing through six collar controls between 7:30 and 9:30 were 13,380 before and 13,150 after, in a decrease of only 1.7%.
  
• Only about half the traffic leaving the two controlled residential zones experienced extra delays; these averaged just 1.0 to 2.5 minutes depending on location during the morning peak. On main radial roads between the outskirts and the city centre, the average journey times of traffic increased by no more than 1.5 minutes during both the morning and evening peak hours.
  
• No significant changes were observed in transport mode by residential zones for journeys through the zone and collar controls to the inner city area, since the small improvements in bus journey times relative to those by car were largely counteracted by an increase in bus fares of 20-25% and a decrease in fuel prices of 20% between the surveys (both changes being in real terms after allowance for inflation).
  
• It was found difficult to impose very long delays on traffic approaching the collar signals due to the restricted space for storing queues of vehicles. The maximum delays produced were about three or four minutes depending on site.
  
• The scheme produced a redistribution of traffic between the main radial roads leading into the city, but this resulted in little increase in vehicle kilometres.
  
• There was a decrease in through traffic taking shortcuts through the residential zones, but the effect on total traffic in the zones was small.
  
• Park and ride had little effect on traffic congestion. The removal of cars from inner city roads was negated by the Park and Ride buses, each of which removed only 3.3 cars on average.

Impacts on supply
The main impacts on supply side are:

• The scheme reduced bus journey times between the residential zones and the city centre by less than on a minute on average during the period of enforcement.
  
• The punctuality of buses was not improved, but bus priorities largely protected buses from the effects of the restricted measures.
  
• Two bus priority lanes during the evening peak, produced savings averaging 1.5 to 2 minutes per bus. The overall saving in bus time due to these two evening peak bus lanes was greater than that observed during the morning peak period when a total of ten bus lanes and various other priority measures were operational on the surveyed routes.

Display location

Context
Fitzroy and Smith (1993) report the performance and main features of the operation of road reallocation in Zurich. During the 1950s and 1960s the trams and buses had been criticised as slow and unreliable, despite the long history of tram use. Delays arose from private vehicle obstruction such as cars turning at junctions and being forced by parked vehicles to drive along the tramlines. To augment public transport speeds and improve reliability, a package of measures was introduced, based on approval of a city referendum in 1977. The aim was to ensure that the road space allocated to public transport was well defined and unobstructed. To facilitate the smooth operation of the segregated track, several complementary actions were initiated:

• prohibition of parking and waiting on 15 road stretches;
  
• 28 no-left-turns at junctions with roads used by trams;
  
• the banning of private cars and taxis from selected zones;
  
• the presence of a sufficient number of uniformed police to enforce restrictions.

The road space allocated to public transport consists mainly of the 117.3 km, 13-route tram network. There are also 19.2 km of reserved lanes for diesel buses (compared to 2.6 km in 1976) out of a city network of 89.7 km, and 36.3 km of trolleybus lines. The benefits of segregated track lie in the length of the exclusive lanes and the strictness with which they are enforced.

After the several years of this implementation, a computer guided signaling system was installed to ensure that punctuality is maintained and waiting time is minimised for public transport. In addition, all public transport mode season tickets (Rainbow) were introduced throughout the city in 1985, freely transferable across family members or friends.

Impacts on demand
From 1960 to 1977, the number of trips by public transport was fairly stable at just below 200 million per year. Thereafter this grew steadily to around 210 million per annum between 1982 and 1984. From 1985 trips rose dramatically to about 280 million in 1990, which represents 33% growth in six years. This was mirrored by a stabilisation in number of car trips on main roads since 1981. However, Rainbow ticket introduction was chiefly responsible for the rapid climb between 1985 and 1990.

Impacts on supply
The combined effect of these measures reduced average journey durations. It was estimated that 8.2 km stretch of one tram route took 36 minutes to traverse during the evening peak period in 1970. By 1991, this had fallen to 32.5 minutes for the same journey at the same time of day.

Contribution to objectives
The Zurich scheme was introduced to improve public transport services as a package of measures and there was no evidence of an effect on the vehicle traffic volume (Fitzroy and Smith, 1993). The contribution to objectives shown in the table below is based on the assumption that the main objective was to improve public transport.

Gothenburg, Sweden

Display location

Gothenburg city centre divided into cells (Cairns et al, 1998)

Context
Cairns et al (1998) and Vuchic (1999) cover Gothenburg as a case study. In 1970, the closure of the CBD to unauthorised vehicles was implemented at weekends since severe traffic and environmental problems became apparent. In 1970-72, the CBD was divided into six cells, whose boundaries could not be crossed by private vehicles. The cells were surrounded by an inner ring road, which acted as a bypass for through traffic, and allowed entry and exit to and from the sectors. In 1973-74, it was planned to extend the zoning concept to the whole of the Central Urban Area (CUA) divided into two cells by a North-South screenline. Between 1970 and 1980, other policies were also put in place: three major central streets were pedestrianised and many streets were made one-way. The proportion of the tram network running on reserved tracks was increased from 65% to 90%, and other public transport was given some priorities on-street and at signals. In 1979, a comprehensive traffic policy was put into place, to manage traffic, to improve the attractiveness of other modes, to complement the existing cells, and to try and reduce vehicle flows, partly to facilitate implementing further cells. By 1988, the number of parking spaces in the CBD had been cut from 21,000 to 14,000, and parking charges were increased by up to 100% (an equivalent of about €1.0/h at 1988 prices).

Impacts on demand

(Cairns et al, 1998) (unit: vehicles/day)

There was obviously a substantial decrease in traffic entering the CBD and CUA. The North-South screenline across the CUA recorded 25,265 vehicles in 1971, and 26,690 vehicles in 1983, an increase of 5.6%, with a marked shift of traffic towards the completed route to the north of the CBD. This means that through traffic increased, transferring onto the remaining available routes. The increase in through traffic was much less than traffic increases in the rest of Gothenburg, so that the cells might have both suppressed car trips for which they were the destination, and suppressed traffic growth in the area generally.

As a result of the CBD cells, total vehicle kilometres for private traffic increased by about 7%, but with no increase in vehicle hours, as changes to junctions made flows more efficient. In addition, the introduction of cells stimulated public transport use and pedestrian movements in the CBD.

Impacts on supply
The cell system itself has not affected supply, but other measures such as the pedestrianisation of major streets, one-way streets, increases in the proportion of the tram system on reserved tracks and traffic in the traffic cells has reduced the road capacity for car use. In addition, parking space was reduced in the CBD.

Display location

Freiburg city centre and major roads (Cairns et al, 1998)

Context
Cairns et al (1998) covers Freiburg as a case study for good planning practice. Most roads in the city centre were closed to motor vehicles by the end of the 1970s, when parts of the major ring road around the city centre were widened from 2 to 4 lanes. The main city centre road was closed for car use, but trams and buses were allowed to enter the city centre. During the late 1980s, some of the capacity on the ring road was reallocated to cycle lanes and cycle paths. The width of the car lanes was reduced by 1-1.5 metres for the cycle lanes. In 1996 some parts of the ring road (Rotteckring and Werderring) were reduced from 4 lanes to 2 lanes, and 2 lanes were changed into two way bus lanes, at the same time a parallel road (Bismarckallee and Schnewlinstrasse) was widened to 4 lanes with a cycle lane.

Impacts on demand
The following table shows the changes to one day traffic flows on major city centre ring road.
(Cairns et al, 1998)

(unit: 1000 vehicles)
* In Nov. 1995 Bismarckallee and Schnewlinstrasse were closed.
** In Feb. 1996 Bismarckallee was opened with 2 lanes.
*** In Mar. 1996 Rotteckring was reduced to 2 lanes and Bismarckallee was extended to 4 lanes.

This table shows a 12% decline of traffic flows on the major city centre ring roads in ten years despite a significant growth in the number of vehicles in Freiburg from 350 per 1000 population in 1977 to 487 per 1000 in 1996, an increase of nearly 40% in 20 years. However, these changes resulted in a reduction in traffic on the treated roads of 34%, of which a large proportion was observed on the alternative routes provided. The overall effect was a reduction in traffic of about 7% of the traffic formerly using the treated sections.

In 1976, 22% of all trips into the city were made by public transport, 18% by bicycle and walking and 60% by vehicle, and in 1995 modal share has changed to decrease car use, with 26% by public transport, 28% by bicycle and walking and 46% by vehicle.

Impacts on supply
In Freiburg, overall road capacity has been reduced overall since 1970s for reallocation to bus or cycle, but some road capacity reductions were associated with expanding the capacity of alternative routes in order to maintain the supply of road space.

Gaps and weaknesses
Long distance diverted traffic may occur outside the studied area. Any omission of these diversions would be likely to result in an overestimate of traffic reduction.

Physical restrictions have not always been implemented to meet simple objectives: one is to reduce the traffic volume; the other is to improve the attractiveness of public transport. As a consequence, it is sometime difficult to isolate which specific measure is effective or ineffective within physical restrictions even though the main cause of the change is clear in the case studies. Individual measures covered in separate sections help to illustrate this.