Impact of Traffic Calming Zones (TCZs) in Cities on Public Transport Operations

Abstract

Traffic calming zones (TCZs) are increasingly being implemented in urban areas to enhance road safety, reduce vehicle speeds, and support sustainable mobility. However, their impact on public transport (PT) operations, particularly bus services, remains underexplored. This study examines the impact of classifying streets into TCZs on bus transport performance in Poland’s ten largest cities. Geospatial analysis and a custom R algorithm delineated areas suitable for TCZs based on road class and administrative category. GTFS data were analysed for almost 1000 bus lines to evaluate the overlap of their routes with TCZs. The findings reveal that in several cities, a significant portion of bus operations would run through TCZs, with the average route segment affected notably by city and zone classification methods. Differences in TCZ size and shape across cities were also statistically significant. This study concludes that although TCZs contribute to safer and more liveable urban environments, their influence on bus speeds, which can lead to changes in fuel or energy consumption, and route design must be carefully managed. Strategic planning is essential to find a balance between the benefits of traffic calming and the operational efficiency of PT. These insights offer valuable guidance for integrating TCZs into sustainable urban transport policy without compromising PT performance.

1. Introduction

The WHO estimates that 1.19 million road traffic deaths occurred in 2021 (the most recent data available in 2025), representing a 5% decrease compared to the 1.25 million deaths in 2010. The European Region reported the most significant drop—36%—in fatalities since 2010 among all regions worldwide. As the key to safety, in Europe, the development of safe road infrastructure was indicated [1]. According to a European Commission study, only 8% of road fatalities happen on motorways, 37% in urban areas, and 54% in rural areas [2]. Further decreases are still sought in both urban and rural areas. Overspeeding is one of the main risk factors for road traffic safety [3]. Speed management is necessary in urban areas, where the road user mix includes a high volume of vulnerable road users, such as pedestrians and cyclists, resulting in reduced speed and increased compliance.

Traffic calming (TC) is a combination of speed management policies and their implementation, mainly through physical measures, which reduce the negative effects of motor vehicle use, change driver behaviour, create a safer road environment for all users, and improve conditions for non-motorised users [4]. TC is based on strategies to reduce traffic accidents and includes various design features to lower vehicle traffic speeds and volumes [5].

Street networks are connected to the zones to force an area-wide effect of TC, and a consistent set of measures has been implemented to achieve the selected goals. In addition to reducing vehicle speed and improving safety for all road users, it can enhance the living environment for residents and reduce pollution and noise [6].

Although traffic calming zones (TCZs) support the realisation of urban strategies to increase traffic safety, their operation could influence the competitiveness of public transport (PT), for example, by lowering the commercial speed of PT services or reducing passenger comfort. Spending funds and time to improve traffic safety would be half the success, but it would compromise the use of sustainable forms of transport. Some cities (e.g., London, UK; Warsaw, Poland) have even made guidelines for introducing TC on streets with bus traffic [7,8]. The authors of this article identified a lack of studies on PT operations in TCZs and aimed to address this research gap.

This article aims to assess how the automatic classification of streets for TCZs based on selected features will affect bus transportation. For this purpose, an answer was sought regarding the relationship between the hierarchy of the road network and the layout of the bus route network, as well as whether there are differences between large cities in Poland in this regard. For this purpose, QGIS 3.40 software was used for the assessments, and a self-made R 4.4.3 algorithm was used for calculations.

The authors of this study, based on their experience in road traffic engineering and the economics of urban mobility, collected i.a. during the project “Introduction of traffic calming zones in Warsaw—guidelines and directions of development”, which was prepared by the Authors for the Road Traffic Management Office of the capital city of Warsaw in Poland in 2022. The project aimed to create design guidelines and designate areas for introducing traffic calming zones (Tempo 30 or residential zones) on networks of D, L, and, exceptionally, Z classes—the lowest classes in the 7-level road classification system in Poland. The method had three stages:

• Stage 1—Review of the available literature and regulations on TCZs and speed management [9]. In total, 229 sources were reviewed, including scientific articles in Polish (40 pieces) and English (41), materials from conferences (18), trainings (15), Polish and EU regulations (22), guidelines and standards from Poland (22), and other countries, including the UK and USA (29), programmes for traffic calming (24), and other sources (18).
• Stage 2—Development of a catalogue of good practices—guidelines for introducing traffic calming zones in Warsaw, and a set of typical solutions was prepared [8]. The guidelines were consulted with city hall clerks responsible for roads, safety, and PT, as well as traffic engineers designing traffic organisation projects.
• Stage 3—A GIS map of the target implementation of traffic calming zones in Warsaw, along with a technical description for the map, was made [10]. The map of the traffic calming zones was subject to public consultations in the city, allowing a socially acceptable solution to be developed.

The project is currently in the implementation phase, with the introduction of TCZs in the specified areas of the city.

2. Literature Review

First, the objectives of TCZs were determined, followed by their introduction methods. Among the reasons for introducing the zones, it was checked whether any purposes related to PT existed. The place of TCZs in the road hierarchy has also been specified, as have various understandings of their implementation in different countries. Next, the review examined the impact of TCZ implementation on PT.

2.1. Objectives of Traffic Calming Zones

TCZs aim to create more balanced, sustainable, and people-friendly urban environments [11]. In the SUMP (Sustainable Urban Mobility Plan) methodology, they are one of the typical actions to be implemented in the cities [12]. In particular, TCZ introduction can have multiple objectives to achieve, which are listed in

Table 1. Objectives of traffic calming zones in urban areas.

No.	Objective	Description	Mentioned in the Literature
1	Safety improvement	Reduce accident and fatality rates, especially for pedestrians and cyclists.	[1,3,4,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43]
2	Speed reduction and compliance	Ensure drivers adhere to posted reduced speed limits through physical and visual measures.	[1,3,4,7,8,9,10,11,13,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,42,43,44]
3	Active mobility promotion	Encourage walking and cycling by making streets more attractive for non-motorised users.	[1,4,8,9,10,11,12,13,14,15,16,17,18,20,21,22,23,24,25,34,35,36,37,38,39,43,44]
4	Environmental benefits	Reduce vehicles’ emissions of noise and pollution, as well as fuel consumption, by lowering speeds and smoothing traffic flow.	[4,7,8,9,16,17,18,20,21,23,24,25,28,30,31,32,34,35,37,38,39,40,43]
5	Traffic flow management	Discourage through-traffic in residential areas and local streets.	[1,4,8,9,10,11,13,15,18,20,22,23,24,28,37,38,39,43,44]
6	Quality of life enhancement	Create more liveable spaces through street transformation and a reduction in traffic dominance in neighbourhoods.	[4,7,8,9,11,17,20,23,25,29,34,37,38,43]
7	Accessibility improvement	Enhance mobility for all users, including children, older people, and those with disabilities.	[1,4,8,9,10,11,12,17,18,19,35,37,38,43]
8	Improvement in street aesthetics	Street transformation by space of higher quality (with so-called street furniture) and more greenery (also as a calming measure).	[4,8,9,10,11,20,25,28,29,30,31,37,38,43]
9	Community cohesion	Foster a sense of place and community by reclaiming street space for social interactions.	[8,9,11,16,17,18,20,28,37,38,43]
10	Revitalizations	Literally, it is revival or bringing back to life; it is the process of bringing degraded areas of the city out of a state of crisis.	[4,8,9,20,25,37,39]
11	Economic benefits	Increase in property values and a boost to local businesses.	[4,9,20,37,43]
12	Increase in public transport usage	Encouraging a shift in transport modal split to PT.	[3,15,17,20,38]
13	Crime reduction	Potentially decrease crime rates by increasing street activity and enhancing natural surveillance through the presence of people.	[4,9,38,43]

. In addition to the description of the objectives, they are listed in order of frequency of occurrence in the literature reviewed during the study.

The most common reasons for TCZs’ introduction are safety improvement (1), speed reduction, and compliance (2). These two reasons were highlighted in almost all of the reviewed literature. Safety improvement is connected with lowering the number of fatalities or crashes involving different road users. Speed reduction was never indicated as a goal by itself, but it was always associated with safety improvement, which is the primary goal of the TCZ.

More than half of the analysed sources indicated that the goals of TCZ were also active mobility promotion (3) and environmental benefits (4). In zones with reduced vehicle speeds, pedestrian and bicycle traffic become more competitive with cars and have more space. Lowered speeds and increased traffic smoothness, which means less acceleration and braking, are intended to reduce noise and pollution from car traffic and fuel consumption.

Less than half of the analysed sources indicated further goals of TCZs: traffic flow management (5), quality of life enhancement (6), accessibility improvement (7), aesthetics of street improvement (8), and increased community cohesion (9).

TCZs were rarely indicated as a tool for boosting the economy (11), increasing the use of PT (12), or reducing crime (13). TCZs should attract private investment, support other programmes involving homeownership and historic preservation, and assist downtown businesses [4]. With more attractive walking and cycling in TCZs, it is also believed that the use of PT should increase [15]. The most surprising goal of TCZs is crime reduction, achieved because of a more liveable community created by neighbours socialising in public streets—the effect is that more eyes on the street discourage antisocial behaviours [43].

2.2. Methods of Introducing Traffic Calming Zones

Urban areas are basic TCZs introduced by countries’ traffic laws, i.e., with a general speed limit of 50 km/h, as reported by the WHO in 58 out of 163 countries, including most European Union countries. Three countries have tightened their national laws to implement a general speed limit of 30 km/h in urban areas [1]. The WHO recommends taking action in areas with dense pedestrian and cyclist traffic [36]. That level of speed is considered safe for interactions within streets used by children and other unpredictable behaviours [37] or in higher-order activity centres (town centres, historical places, local centres) and designated residential areas [16].

In Spain, as of 11 November 2020, generic speed limits were established on urban roads: 20 km/h on single-carriageway one-lane roads, 30 km/h on single-carriageway two-lane roads, and 50 km/h on carriageways with two or more lanes per direction. Spanish researchers studied that reducing the speed limit from 50 km/h to 30 km/h on all urban roads is not recommended for the whole urban area. A higher speed on part of the network should help redirect traffic to regions with fewer pedestrians and inhabitants [22].

The simplest TCZ is when a road passing through a rural area enters an urban area for a short section, i.e., one kilometre. Speed limits are introduced to maintain calm traffic and improve road safety, for example, by allowing pedestrians to cross the road safely, and to save fuel by reducing the number of constant speed increases and decreases. Drivers tend to speed up when approaching the end of an urban area, which is characterised by densely located buildings [24].

In larger urban areas, engineering practice recommends creating a multi-level hierarchy of roadways, consisting of at least three levels: transit, local, and target roads (Figure 1). Transit traffic should be routed through roads with a higher speed limit than 50 km/h, located further from buildings and with a minimised number of intersections. The goal is to minimise the travel time between distant points of the network and attract traffic from lower levels of the hierarchy. Local traffic is subject to a speed limit of 40–50 km/h and should be used for medium-distance travel, particularly in connection with the target and transit roads. In urban areas (i.e., residential areas and city centres) with high cyclist and pedestrian movement and target traffic to destinations or from travel origins, speed is limited to 30 km/h or lower. Based on this hierarchy, local authorities are adapting speed limits on the streets to the context of their function [35].

Solutions for TCZs in Poland include zones 40–50 (for local traffic), Tempo 30 zones, residential zones (20 km/h), and shared spaces (for target traffic) [25]. Their introduction is often associated with the implementation of metered parking systems, e.g., in various spa towns in Poland or city centres [40].

Considering the function of the street, a specific TCZ may be introduced in the vicinity of a particular building, such as a school. There are various methods for traffic calming in such areas, for example, introducing a zone only during periods of school opening and closing times, as well as when students are spending time around school areas during breaks [19]. The vicinities of schools are not the only specific TCZs introduced globally. In Singapore, there is a so-called Silver Zone programme, which introduces road safety features that are friendly for older inhabitants. They are especially introduced in areas with a high proportion of senior residents and where there have been past road traffic injuries involving seniors [17].

To find the most suitable solution between full pedestrianisation or reconstruction to a woonerf-type TCZ—level of inhabitant zone—the Analytic Hierarchy Process (AHP) was tested using the example of a Greek municipality [41].

The introduction of TCZs means not only putting a speed limit traffic sign and police enforcement measures in urban areas, which would be inefficient in larger areas, but managing speed by self-explanatory road design [16] or engineering traffic calming measures (TCMs), which means to physically reduce speed by creating obstacles, changing car trajectories, or narrowing road cross-sections [38]. Guiding principles for implementing TCZs are developed based on multiple research tools, including the literature reviews, traffic measurements, consultation surveys, and practitioners’ knowledge, among others [16]. Before implementation, a simulation study could be conducted (i.e., PTV Vissim software). Video recordings and trajectory extraction software are the basis for creating a microsimulation model of the traffic-calmed street [44].

The most effective TCMs at the entrance to the TCZs were devices that changed the road surface elevation [30]. The drivers are warned of the traffic-calmed area ahead by the specific street and junction geometry, as well as the street furniture, compelling them to reduce their driving speed [26].

Some of the most common TCMs are concrete block-paved vertical traffic calming devices, including speed tables, speed humps, speed bumps, and raised intersections or pedestrian crossings. Because TCMs have a local effect on vehicle speed and do not create a memory effect for drivers, they must be combined appropriately to provide a TCZ in an urban area [27]. The idea of locating the measures is to ensure a stable, low-speed environment for vehicles. A few vertical traffic calming devices should be put in a small interval (100–200 m) rather than one high vertical traffic calming device every 500 m [31].

Other effective measures include adjusting the corridor geometry or narrowing a lane [29]. Effective measures are also chokers applied in combination with mid-block crosswalks, bollards, and concrete planters. Streets featuring more access points to courtyards and public areas can also lead to an effective reduction in speed [28]. Traffic calming zones can also be introduced with the strict organisation of parking places. In some countries, in residential zones, it is obligatory to mark all parking spots. The configuration of the parking lanes can also be a TCM to reduce speed [32]. Particular attention needs to be paid to the directions of traffic on one-way streets in zones with a speed limit of 30 km/h to avoid overcrowding on other streets [28].

A change in the roadway’s surface could be another example of TCMs. The surface made of cobblestone was one of the most effective, lowering the share of drivers exceeding the maximum speed on the road to only 10%. Drivers try to maintain driving comfort on road sections and must significantly reduce their speed. Cobblestones are often connected with historical heritage zones and are subject to the conservator’s supervision. Roads in TCZs should meet the high functional, engineering, and visual standards and be a harmonious part of the historic surroundings [29].

A comparative analysis showed that traffic calming should be introduced and work fairly similarly in other parts of the world (for example, India) as in European countries [34]. Still, there are examples like Australia, where area-wide 30 km/h speed limits are not generally accepted, partly due to regulatory barriers [16]. An extraordinary solution to support the enforcement of TCZs is the introduction of nighttime police operations, similar to those in Mexico City. This solution prevents driving at a higher speed than police vehicles travelling in parallel on all roadway lanes. That level of enforcement has not yet been needed in EU countries [39].

2.3. Traffic Calming Zones Influence the Operations of Public Transport

In the case of public transport (PT), TCZs mainly influence urban bus operations. Suburban, regional, or long-distance bus PT is primarily directed through the transit or local network of roadways and enters low-speed TCZs for only a short section of the route to access the main bus stations. For example, in the city of Hobart (Australia), 30 km/h zones, forming part of a streetscape upgrade, have been implemented to support the Central Bus Interchange [16]. Trams can also be influenced by the introduction of TCZs, particularly on sections of the rails that are combined with roadway and car traffic. However, in their own right-of-way sections, they are exempt from TCZ speed regulations [8]. Low-speed TCZs (30 km/h or lower) can affect regular bus routes, access roads to depots, and diversionary routes, mainly by potentially slowing down bus journeys, leading to increased travel times, disrupted schedules, and generally lower attractiveness of PT.

Before TCZ implementation, the network’s intelligent design must be carried out to achieve the required traffic calming without a negative effect on bus operation. The share of bus routes in TCZs and their impact on the schedule should be calculated. During the project for Warsaw, it was assumed that bus lines longer than 10 km should run through TCZs for less than 40% of the route, and lines longer than 15 km for less than 20% of the route. The exceedances of this indicator resulted from the line running through the strict city centre, which is of tourist and historic nature, and from the course of existing roads [10]. In some cases, the introduction of TCZs can lead to adjustments in bus routes or stop locations to minimise their impact [7].

Generally, buses operating in TCZs should be designated as ‘local services’ and be capable of operating at a speed of 25–30 km/h. Bus transport organisers should consider an operating speed of 25 km/h or lower only when crossing through vertical TCMs, to minimise discomfort to drivers and passengers. The measures used should be suitable for the types of buses in service, i.e., special measures need to be adapted when low-floor or articulated buses operate in TCZs. Care should also be taken to site parking at a sufficient distance from TCMs to avoid hindering bus operation [7]. On the street of TCZs with bus traffic, the density of TCMs should be much lower than in a typical street in the zone, limited, for example, only to the dangerous points of the street [8].

Generally, bus traffic is rarely provided in very-low-speed TCZs, especially in residential zones with a speed limit of 20 km/h or lower. In the city of Ljutomer (Slovenia), the research indicated that most vehicles on the roads in TCZs were passenger cars—99% in 2014 and 2017, and 98% in 2018. The 1% or 2% of vehicles were cars with trailers and two-axle lorries or buses [11].

TCZs require special attention during design to ensure they are compatible with PT operations. Buses, with their stiffer suspension and lower manoeuvrability compared to cars, demand smooth and well-planned layouts that protect passenger comfort—especially for elderly, disabled, standing, or moving passengers through the compartment section. TCMs can also accelerate vehicle wear and tear, causing an increase in maintenance costs, so their geometry and materials should be optimised to minimise such effects. All elements should be carefully designed to avoid conflicts with parking, access, and ground clearance, while maintaining high construction and maintenance standards to ensure long-term functionality and safety [7]. Road traffic and public transport organisers should consult all the solutions used [8].

Speed bumps, humps, and tables should not be used on bus routes because they influence the vertical movement of the bus. It is possible to use speed humps or tables only with sinusoidal on- and off-going ramps. A variable cross-section of ramp surfaces, different for the width of cars and buses, is also very effective [8]. Speed cushions are preferred, as they enhance road safety while causing minimal or no negative impact on bus operations and passenger comfort [7]. Because of their larger surface and approaches, raised intersections are allowed but are less comfortable for buses than cushions. Roundabouts should be designed to allow buses to pass easily [38]. In the case of mini-circles, the central island should be paved to allow the bus to turn using that space [43]. On intersections, it is necessary to consider prioritising the street with bus traffic to ensure the passengers’ comfort (avoiding frequent changes in acceleration) [10].

In London (UK), the most serious problems for bus operations have occurred on routes with excessive speed tables and cushions, particularly on high-frequency routes [7]. In Kraków (Poland), buses operate on roads with TCZs, connecting different city areas and its centre. On the streets, TCMs are installed—speed cushions to slow down car traffic without compromising the comfort level for bus passengers or slowing down the buses [23,44].

Public transport infrastructure can also be a TCM by itself, as exemplified by a bus stop without a bay. It can or cannot force drivers to reduce speed, depending on the width of the roadway, and whether it is possible to avoid a bus without any additional manoeuvre [29]. Buses typically stop for 10–20 s, which in TCZs should not significantly affect the traffic conditions. The solution reduces the costs associated with building additional infrastructure just for PT. It allows the bus to continue its journey after serving the stop without waiting to leave the bay, reducing the traffic influence on the bus journey [8]. Special care is needed at bus stops to ensure that access and egress are not compromised and to reduce the risks to passengers waiting to alight. Bus stops are also identified as areas that can restrict visibility for pedestrians crossing the street behind or in front of a stopped vehicle, so they require special attention during the design phase [22].

A different pavement structure can also be used where buses are in operation. Eight to ten centimetres thick, anthracite-coloured interlocking concrete blocks are recommended, particularly on the on-ramps, to provide a clear contrast with pavement markings. To reinforce the structure, a concrete base layer should be applied beneath the surface and extended through the one-metre-long approach and departure sections of the vertical traffic calming devices on streets used by buses [31]. Plateaus longer than 10 m allow all tyres to simultaneously be on the plateau, improving comfort [38]. All the measures and their influence on bus traffic are shown in Figure 2.

To sum it up, researchers have primarily focused on examining the influence of various TCMs on bus operations but have not investigated the overall impact of TCZs on the performance of bus networks. We found a research gap that will be explored further in this research.

3. Materials and Methods

The first stage of the research involved collecting road network data for the 10 largest cities in Poland. The data were obtained from Poland’s official geoportal [45] as a topographic objects database. Subsequently, data related to road carriageways were extracted. City boundaries were sourced from the National Register of Boundaries [46].

Roads in Poland are classified into classes and categories [47]. The classification by class depends on the technical standard of the road and its infrastructure. The following classes are distinguished: motorways (A), expressways (S), main roads of accelerated traffic (GP), main roads (G), collector roads (Z), local roads (L), and access roads (D). The category classification pertains to the road administrator and includes national, voivodeship, county, and commune roads. Classes and categories are interrelated but not unequivocally assigned to each other. TCMs are applied to roads at the lower end of the hierarchy. According to the assumptions [10], local roads (L) and access roads (D) should be located within TCZs. Consequently, three scenarios were adopted for this article:

• Streets of class Z or higher delineate borders of TCZs;
• Streets of county category or higher delineate borders of TCZs;
• A combination of the two above delineation criteria.

The carriageway layer was combined with the city boundary to determine TCZs, and the resulting network of lines was converted into polygons. It was assumed that the boundary of a TCZ runs 30 m from the carriageway that forms the boundary of the zone. The assumed value of 30 m is arbitrary. This value was adopted for all cities studied. This distance allows a vehicle to accelerate before reaching an intersection with traffic lights. The value was adopted due to the limited detail of the study. In reality, the precise location of a traffic sign depends on various factors, including road infrastructure, bus stop locations, road geometry, and many other elements. A value greater than 100 m does not significantly affect the number of TCZs thus designated. Only increasing this distance to 100 m results in many of the smallest zones disappearing, as shown in Figure 3.

For the identified TCZs, the following parameters were calculated:

• Surface area;
• Diameter of the inscribed circle;
• Diameter of the circumscribed circle;
• Shape coefficient is the ratio of surface area (in km²) to the diameter of the circumscribed circle (in km).

Zones with a surface area smaller than 0.02 km², a circumscribed circle diameter smaller than 70 m, or a shape coefficient smaller than 0.05 were removed. This eliminated minimal, narrow, and elongated areas, such as wide medians of dual carriageways. In this way, using the R language [48] and the packages [49,50,51,52,53], areas potentially designated for traffic calming were automatically identified and used in further research.

In the next stage of the study, GTFS data for PT in the selected cities were collected [54]. This data includes information about timetables, stop locations, and vehicle route paths. The data on vehicle route paths is optional [55]. Unfortunately, among the 10 largest cities in Poland, these data were available for only 7 cities, and the study was conducted for these cities. Due to structural differences, the data were converted into a uniform format. Data for 11 July 2023 were selected for this study. Subsequently, only bus lines were chosen from the dataset and grouped according to the type of line:

• Regular lines (marked in black);
• Accelerated, fast, and express lines (marked in red);
• Suburban and zone lines (marked in green);
• Night lines (marked in blue).

For each bus line, the following were determined: route lengths, route lengths within traffic-calmed zones, and the share of routes within TCZs. For each route variant, the number of trips per day was calculated. Then, the average trip length was determined due to differences in the course of variant routes for selected lines.

Descriptive statistics were then calculated for the results, and comparisons were made using nonparametric statistical tests due to the lack of normal distribution of the results. The detailed scope of the computed statistics is presented in the “Results” Section. The scope of the descriptive statistics includes the mean, median, standard deviation, and the minimum and maximum values of each variable. Statistically significant differences were found between the variable values for individual cities, as verified using the Kruskal–Wallis rank sum test.

The research workflow implemented in R is as follows:

• Download data from the BDOT10k database, unpack it, and select the roadway layer (SKJZ). Download city administrative boundaries from the National Register of Boundaries.
• Create road network layers in three variants: roads with class Z and higher, roads with the district category and higher, or roads with class Z and higher and also with the district category and higher.
• Merge the selected road layer created in step 2 with the city boundary.
• Convert the line layers obtained in step 3 into a polygon layer, then reduce the polygons by 30 m using the buffer function. Remove empty polygons.
• Determine polygon parameters: area, perimeter, circumscribed circle diameter, inscribed circle diameter, and area-to-circle diameter ratio.
• Remove small and very narrow polygons (e.g., wide median on multi-lane roads).
• Load route data into the GTFS structure.
• Load trip information from GTFS data.
• Select timetable data for a specific day and limit the data scope to buses (removing tram, metro, and rail routes and trips).
• Manually assign line numbers to line types.
• Determine the route length of lines in the TCZ and the share of routes. Due to the variants in some lines, the average route length across all trips during the day was taken into account.
• Perform steps 5–11 for all three layers of the TCZs determined in step 2.
• Calculate descriptive statistics for the TCZs and bus lines for all variants.
• Perform comparisons using statistical tests.
• Generate tables with results, graphs, and maps.

4. Results

Descriptive statistics for traffic-calmed areas determined based on the road network of class Z and higher are presented in

Table 2. Descriptive statistics of traffic calming zones designed by road class—short table.

TCZ Area Parameter	Białystok N = 70	Wrocław N = 97	Bydgoszcz N = 67	Gdańsk N = 66	Kraków N = 87	Lublin N = 77	Łódź N = 210	Poznań N = 133	Szczecin N = 136	Warsaw N = 430	p-Value 1
surface area [km2]											<0.001
Mean	1.31	2.80	2.43	3.77	3.50	1.75	1.26	1.81	2.06	1.06
Median	0.78	0.80	0.53	1.05	1.34	0.90	0.40	0.45	0.28	0.31
diameter—circumscribed circle [m]											<0.001
Mean	1753	2262	2159	2716	2723	1924	1559	1814	1579	1385
Median	1483	1542	1350	1864	1975	1518	1120	1222	916	929
shape [km]											<0.001
Mean	5.7	6.9	6.2	7.1	8.1	6.4	4.9	5.8	4.9	4.7
Median	5.1	4.8	4.5	5.0	6.3	5.5	3.6	4.1	2.9	3.4
diameter—inscribed circle [m]											<0.001
Mean	778	856	799	929	1027	828	643	771	631	608
Median	670	652	579	726	853	730	502	545	391	447
circuit [m]											<0.001
Mean	4843	7269	6567	8688	8115	5464	4285	5288	5056	3806
Median	3851	3898	3443	5121	5381	4195	2916	2997	2495	2436
circuit—surface area [m/km2]											<0.001
Mean	6805	8610	8131	7329	5777	6896	9220	8386	11,818	9333
Median	5220	5767	6617	5693	4248	5129	7091	6581	9280	7757

¹ Kruskal–Wallis rank sum test.

(full tables with all parameters are at the Appendix A). The mean surface area of TCZs ranges from approximately 1 km² in Warsaw and Łódź to nearly 4 km² in Gdańsk and Kraków. Across cities, median diameters of the circumscribed circles range from roughly 900 m (Szczecin) to over 1.9 km (Kraków). At the same time, shape indices and circuit lengths are generally higher in the compact historic cities of southern Poland (Kraków, Wrocław) than in those with more orthogonal street grids in central Poland (Warsaw, Łódź).

Table 3. Descriptive statistics of traffic calming zones designed by road category—short table.

TCZ Area Parameter	Białystok N = 58	Wrocław N = 29	Bydgoszcz N = 73	Gdańsk N = 83	Kraków N = 131	Lublin N = 85	Łódź N = 261	Poznań N = 143	Szczecin N = 132	Warsaw N = 431	p-Value 1
surface area [km2]											<0.001
Mean	1.60	9.75	2.23	2.98	2.30	1.59	1.00	1.67	2.13	1.06
Median	0.83	3.36	0.55	0.50	0.88	0.67	0.31	0.45	0.29	0.32
diameter—circumscribed circle [m]											<0.001
Mean	1902	4173	2077	2355	2266	1812	1360	1794	1628	1382
Median	1599	3623	1351	1235	1629	1446	965	1222	916	944
shape [km]											<0.001
Mean	6.1	13.4	5.9	6.1	6.4	6.0	4.4	5.6	5.0	4.7
Median	5.3	9.7	4.5	3.9	5.2	5.2	3.2	4.0	3.0	3.5
diameter—inscribed circle [m]											<0.001
Mean	803	1689	778	800	865	782	572	744	641	625
Median	717	1411	593	541	712	615	428	538	403	451
circuit [m]											<0.001
Mean	5562	13,136	6112	7602	6069	4984	3860	5220	5210	3720
Median	4318	9443	3570	3410	4303	4113	2570	2997	2560	2496
circuit—surface area [m/km2]											<0.001
Mean	7188	4805	8813	9064	7258	7015	10,526	8478	11,929	9263
Median	4929	2690	5578	6826	5145	5470	8195	6722	8957	7509

¹ Kruskal–Wallis rank sum test.

presents descriptive statistics for zones defined based on road categories, using the network of county roads and higher categories. The dispersion of values increases markedly—especially for Wrocław, which exhibits the largest mean area and circuit. Warsaw again shows the smallest and most numerous zones.

Table 4. Descriptive statistics of traffic calming zones designed by road class and category—short table.

TCZ Area Parameter	Białystok N = 86	Wrocław N = 101	Bydgoszcz N = 80	Gdańsk N = 99	Kraków N = 137	Lublin N = 93	Łódź N = 291	Poznań N = 145	Szczecin N = 140	Warsaw N = 457	p-Value 1
surface area [km2]											<0.001
Mean	1.06	2.69	2.03	2.48	2.20	1.44	0.89	1.65	2.00	1.00
Median	0.57	0.80	0.48	0.39	0.83	0.63	0.27	0.45	0.29	0.30
diameter—circumscribed circle [m]											<0.001
Mean	1527	2243	1943	2110	2215	1712	1265	1793	1597	1348
Median	1304	1542	1171	1232	1546	1354	902	1232	998	925
shape [km]											<0.001
Mean	5.0	6.9	5.6	5.6	6.2	5.7	4.1	5.6	4.9	4.6
Median	4.3	4.8	4.0	3.2	5.2	4.7	2.9	4.0	3.0	3.4
diameter—inscribed circle [m]											<0.001
Mean	667	857	736	732	846	737	536	739	640	598
Median	576	659	548	483	704	599	377	538	403	443
circuit [m]											<0.001
Mean	4214	7042	5692	6707	5970	4808	3529	5185	4983	3665
Median	3523	3898	3216	3345	4062	3642	2300	3063	2560	2366
circuit—surface area [m/km2]											<0.001
Mean	8676	8484	8919	9340	7446	7678	11,262	8451	11,760	9568
Median	6156	5408	6656	7681	5295	5590	8874	6722	8957	7845

¹ Kruskal–Wallis rank sum test.

contains data for areas identified using both road classes and categories. Combining both criteria yields intermediate results: the mean areas and diameters decrease relative to those in Table 3, but the variation among cities remains. Combining both hierarchies reduces extreme outliers observed in the single-criterion variants.

Next, descriptive statistics were calculated for the characteristics of PT lines for the same three scenarios. The results are presented in

Table 5. Descriptive statistics of bus lines—traffic calming zones designed by road class—short table.

Line Parameter	Białystok N = 46 1	Gdańsk N = 133 1	Kraków N = 172 1	Poznań N = 149 1	Szczecin N = 78 1	Warsaw N = 287 1	Wrocław N = 101 1	p-Value 2
Line colour
Black (regular)	30 (65%)	99 (74%)	76 (44%)	45 (30%)	37 (47%)	135 (47%)	53 (52%)
Red (accelerated)	0 (0%)	0 (0%)	7 (4.1%)	0 (0%)	2 (2.6%)	25 (8.7%)	4 (4.0%)
Blue (night)	0 (0%)	17 (13%)	14 (8.1%)	15 (10%)	22 (28%)	41 (14%)	17 (17%)
Green (suburban)	16 (35%)	17 (13%)	75 (44%)	89 (60%)	17 (22%)	86 (30%)	27 (27%)
Distance travelled [km]								<0.001
Mean	888	635	752	640	753	1176	1034
Median	891	423	580	556	561	887	682
TCZ distance travelled [km]								<0.001
Mean	103	96	148	71	54	79	256
Median	109	17	97	26	16	18	97
Mean trip length [km]								<0.001
Mean	13	13	14	15	12	16	14
Median	12	12	14	13	10	15	15
Trip count [−]								<0.001
Mean	72	53	58	51	72	85	78
Median	73	47	51	37	60	74	60
TCZ share [−]								<0.001
Mean	0.10	0.16	0.25	0.10	0.10	0.07	0.22
Median	0.10	0.04	0.19	0.06	0.04	0.02	0.22
Drive through the TCZ	38 (83%)	76 (57%)	169 (98%)	120 (81%)	71 (91%)	216 (75%)	101 (100%)

¹ n (%). ² Kruskal–Wallis rank sum test.

,

Table 6. Descriptive statistics of bus lines—traffic calming zones designed by road category—short table.

Line Parameter	Białystok N = 46 1	Gdańsk N = 133 1	Kraków N = 172 1	Poznań N = 149 1	Szczecin N = 78 1	Warsaw N = 287 1	Wrocław N = 101 1	p-Value 2
Line colour
Black (regular)	30 (65%)	99 (74%)	76 (44%)	45 (30%)	37 (47%)	135 (47%)	53 (52%)
Red (accelerated)	0 (0%)	0 (0%)	7 (4.1%)	0 (0%)	2 (2.6%)	25 (8.7%)	4 (4.0%)
Blue (night)	0 (0%)	17 (13%)	14 (8.1%)	15 (10%)	22 (28%)	41 (14%)	17 (17%)
Green (suburban)	16 (35%)	17 (13%)	75 (44%)	89 (60%)	17 (22%)	86 (30%)	27 (27%)
Distance travelled [km]								<0.001
Mean	888	635	752	640	753	1176	1034
Median	891	423	580	556	561	887	682
TCZ distance travelled [km]								<0.001
Mean	219	93	102	64	59	90	682
Median	165	13	58	20	19	18	349
Mean trip length [km]								<0.001
Mean	13	13	14	15	12	16	14
Median	12	12	14	13	10	15	15
Trip count [−]								<0.001
Mean	72	53	58	51	72	85	78
Median	73	47	51	37	60	74	60
TCZ share [−]								<0.001
Mean	0.20	0.14	0.17	0.09	0.10	0.08	0.59
Median	0.18	0.04	0.10	0.06	0.04	0.03	0.66
Drive through the TCZ	43 (93%)	77 (58%)	168 (98%)	120 (81%)	71 (91%)	216 (75%)	101 (100%)

¹ n (%). ² Kruskal–Wallis rank sum test.

and

Table 7. Descriptive statistics of bus lines—traffic calming zones designed by road class and category—short table.

Line Parameter	Białystok N = 46 1	Gdańsk N = 133 1	Kraków N = 172 1	Poznań N = 149 1	Szczecin N = 78 1	Warsaw N = 287 1	Wrocław N = 101 1	p-Value 2
Line colours
Black (regular)	30 (65%)	99 (74%)	76 (44%)	45 (30%)	37 (47%)	135 (47%)	53 (52%)
Red (accelerated)	0 (0%)	0 (0%)	7 (4.1%)	0 (0%)	2 (2.6%)	25 (8.7%)	4 (4.0%)
Blue (night)	0 (0%)	17 (13%)	14 (8.1%)	15 (10%)	22 (28%)	41 (14%)	17 (17%)
Green (suburban)	16 (35%)	17 (13%)	75 (44%)	89 (60%)	17 (22%)	86 (30%)	27 (27%)
Distance travelled [m]								<0.001
Mean	888	635	752	640	753	1176	1034
Median	891	423	580	556	561	887	682
TCZ distance travelled [m]								<0.001
Mean	78	74	97	61	52	74	253
Median	47	11	50	20	16	17	97
Mean trip length [m]								<0.001
Mean	13	13	14	15	12	16	14
Median	12	12	14	13	10	15	15
Trip count [−]								<0.001
Mean	72	53	58	51	72	85	78
Median	73	47	51	37	60	74	60
TCZ share [−]								<0.001
Mean	0.07	0.12	0.17	0.09	0.09	0.07	0.22
Median	0.05	0.03	0.10	0.06	0.04	0.02	0.22
Drive through TCZ	36 (78%)	76 (57%)	168 (98%)	120 (81%)	71 (91%)	216 (75%)	101 (100%)

¹ n (%). ² Kruskal–Wallis rank sum test.

. The exact values for TCZs were determined for numerical variables, while for categorical features, the number of occurrences and the percentage share were calculated. In all three tables, Kraków and Wrocław have the highest share of bus lines operating within TCZs (98–100%) and the mean share of trip length inside TCZs (0.17–0.59). In contrast, Warsaw and Poznań exhibit the lowest mean share of line trip lengths passing through the TCZs (0.07–0.10).

Figure 4 presents the share of bus lines in the TCZ, illustrated as a graph, showing the proportion of all bus lines operating within the area restricted by the TCZ across three scenarios of road division. Line type, number of trips per day, and mean trip length were considered, and each graph shows the share of trip length within the TCZ in percent.

The study using Pearson’s χ² test showed that the surface areas of the regions do not follow a normal distribution for all cities and all types of main street networks (based on both road classes and categories). Therefore, comparisons were conducted using nonparametric tests—the Kruskal–Wallis test and Dunn’s test.

It was also examined whether the choice of road classes, categories, or both classes and categories significantly affects the area of the regions. The results are presented in

Table 8. Verify the hypothesis regarding the equality of median surface areas for different main street networks using the Kruskal–Wallis test.

City	p-Value	Normal
Kraków	0.08	TRUE
Poznań	0.7	TRUE
Warsaw	0.08	TRUE
Białystok	0.04	FALSE
Gdańsk	0.33	TRUE
Wrocław	0.99	TRUE
Szczecin	0.99	TRUE
Łódź	0.77	TRUE
Lublin	<0.01	FALSE
Bydgoszcz	0.01	FALSE

. Significant differences were found in almost all cities. No statistically significant differences were identified for individual main street networks in Białystok, Lublin, and Bydgoszcz.

In the next step of the study, Dunn’s test examined whether there are differences in the distribution of TCZ surfaces between individual cities. This analysis was also conducted for the three types of main road networks. The results are presented in

Table 9. Statistical significance testing of the differences in the distributions of traffic-calmed area surfaces determined based on road category for individual cities.

City	Bydgoszcz	Gdańsk	Kraków	Lublin	Łódź	Poznań	Szczecin	Warsaw	Wrocław
Białystok	FALSE	FALSE	FALSE	FALSE	TRUE	FALSE	TRUE	TRUE	FALSE
Bydgoszcz		FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	TRUE
Gdańsk			FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	TRUE
Kraków				FALSE	TRUE	FALSE	TRUE	TRUE	FALSE
Lublin					TRUE	FALSE	TRUE	TRUE	FALSE
Łódź						TRUE	FALSE	FALSE	TRUE
Poznań							FALSE	FALSE	TRUE
Szczecin								FALSE	TRUE
Warsaw									TRUE

,

Table 10. Statistical significance testing of the differences in the distributions of traffic-calmed area surfaces determined based on road class for individual cities.

City	Bydgoszcz	Gdańsk	Kraków	Lublin	Łódź	Poznań	Szczecin	Warsaw	Wrocław
Białystok	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	TRUE	TRUE	FALSE
Bydgoszcz		FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE
Gdańsk			FALSE	FALSE	TRUE	FALSE	TRUE	TRUE	FALSE
Kraków				FALSE	TRUE	TRUE	TRUE	TRUE	FALSE
Lublin					TRUE	FALSE	TRUE	TRUE	FALSE
Łódź						FALSE	FALSE	FALSE	FALSE
Poznań							FALSE	FALSE	FALSE
Szczecin								FALSE	TRUE
Warsaw									TRUE

and

Table 11. Statistical significance testing of the differences in the distributions of traffic-calmed area surfaces determined based on road category and class for individual cities.

City	Bydgoszcz	Gdańsk	Kraków	Lublin	Łódź	Poznań	Szczecin	Warsaw	Wrocław
Białystok	FALSE	FALSE	FALSE	FALSE	TRUE	FALSE	FALSE	FALSE	FALSE
Bydgoszcz		FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE	FALSE
Gdańsk			FALSE	FALSE	TRUE	FALSE	FALSE	FALSE	FALSE
Kraków				FALSE	TRUE	FALSE	TRUE	TRUE	FALSE
Lublin					TRUE	FALSE	TRUE	TRUE	FALSE
Łódź						TRUE	FALSE	FALSE	TRUE
Poznań							FALSE	FALSE	FALSE
Szczecin								FALSE	TRUE
Warsaw									TRUE

. Most substantial differences were observed when zones were defined by road category. Łódź, Warsaw, Wrocław, and Szczecin most often differed from other cities, indicating distinct spatial patterns of traffic calming, while Gdańsk and Bydgoszcz showed relatively homogeneous distributions. When both road class and category were applied simultaneously, the number of significant differences decreased the most, suggesting that this combined approach yields more consistent and comparable delineations of traffic-calmed areas across cities.

The share of bus lines passing through traffic-calmed zones varies in individual cities and for different methods of defining the main street network. The number of lines passing through and not passing through traffic-calmed zones is presented in Figure 5.

Comparisons between cities were made using Fisher’s Exact Test. The comparison results are presented in

Table 12. Results of comparing the shares of bus lines travelling through traffic-calmed zones in different cities using various criteria for defining the main street network.

City 1	City 2	Number of Lines	Category		Class		Class and Category
			P.adj	P.adj.signif	P.adj	P.adj.signif	P.adj	P.adj.signif
Białystok	Gdańsk	179	4.17 × 10−5	****	0.0205	*	0.104	ns
Białystok	Kraków	218	0.656	ns	0.00275	**	0.000455	***
Białystok	Poznań	195	0.272	ns	1	ns	1	ns
Białystok	Szczecin	124	0.743	ns	1	ns	0.327	ns
Białystok	Warsaw	333	0.0353	*	1	ns	1	ns
Białystok	Wrocław	147	0.234	ns	0.000762	***	6.01 × 10−5	****
Gdańsk	Kraków	305	4.03 × 10−18	****	1.28 × 10−19	****	1.36 × 10−18	****
Gdańsk	Poznań	282	0.000637	***	0.000409	***	0.00038	***
Gdańsk	Szczecin	211	2.55 × 10−6	****	1.23 × 10−6	****	1.3 × 10−6	****
Gdańsk	Warsaw	420	0.00478	**	0.00277	**	0.00277	**
Gdańsk	Wrocław	234	8.24 × 10−16	****	3.24 × 10−16	****	3.24 × 10−16	****
Kraków	Poznań	321	5.66 × 10−6	****	1.08 × 10−6	****	5.66 × 10−6	****
Kraków	Szczecin	250	0.272	ns	0.0826	ns	0.272	ns
Kraków	Warsaw	459	8.82 × 10−11	****	1.42 × 10−11	****	8.82 × 10−11	****
Kraków	Wrocław	273	0.693	ns	1	ns	0.924	ns
Poznań	Szczecin	227	0.272	ns	0.327	ns	0.327	ns
Poznań	Warsaw	436	0.693	ns	1	ns	0.924	ns
Poznań	Wrocław	250	2.94 × 10−6	****	2.76 × 10−6	****	2.94 × 10−6	****
Szczecin	Warsaw	365	0.0197	*	0.0179	*	0.0179	*
Szczecin	Wrocław	179	0.0255	*	0.0205	*	0.023	*
Warsaw	Wrocław	388	1.28 × 10−9	****	1.28 × 10−9	****	1.28 × 10−9	****

**** p-value less than 0.0001, *** p-value between 0.0001 and 0.001, ** p-value between 0.001 and 0.01, * p-value between 0.01 and 0.05, ns—no significant.

. Depending on the method used to define the main street network, the number of lines passing through traffic-calmed zones in individual cities does not differ significantly. This analysis was also conducted using Fisher’s Exact Test.

Another measure of the impact of defining traffic-calmed zones on bus traffic is the share of the route length of lines running through these zones. This parameter usually does not follow a normal distribution, as shown in

Table 13. Results of testing the hypothesis of the conformity of the distribution of the shares of line routes within traffic-calmed zones with a normal distribution using Pearson’s χ² test.

Name	Test	p-Value	Normal
Białystok	Category	0.86	TRUE
Białystok	Class	0.77	TRUE
Białystok	Class and category	0.7	TRUE
Gdańsk	Category	<0.01	FALSE
Gdańsk	Class	<0.01	FALSE
Gdańsk	Class and category	<0.01	FALSE
Kraków	Category	<0.01	FALSE
Kraków	Class	<0.01	FALSE
Kraków	Class and category	<0.01	FALSE
Poznań	Category	<0.01	FALSE
Poznań	Class	<0.01	FALSE
Poznań	Class and category	<0.01	FALSE
Szczecin	Category	<0.01	FALSE
Szczecin	Class	<0.01	FALSE
Szczecin	Class and category	<0.01	FALSE
Warsaw	Category	<0.01	FALSE
Warsaw	Class	<0.01	FALSE
Warsaw	Class and category	<0.01	FALSE
Wrocław	Category	<0.01	FALSE
Wrocław	Class	0.07	TRUE
Wrocław	Class and category	0.04	FALSE

.

Next, using the Kruskal–Wallis and Dunn’s tests, it was examined whether there are differences between cities in the shares of route lengths within TCZs for different methods of zone definition. The results of the analysis are presented in

Table 14. Results of testing the hypothesis on differences in the shares of route lengths within traffic-calmed zones between different cities.

Comparison	Category		Class		Class and Category
	P.adj	P.adj.signif	P.adj	P.adj.signif	P.adj	P.adj.signif
Białystok–Gdańsk	0.82	FALSE	<0.01	TRUE	0.02	TRUE
Białystok–Kraków	1	FALSE	<0.01	TRUE	0.03	TRUE
Gdańsk–Kraków	0.42	FALSE	1	FALSE	1	FALSE
Białystok–Poznań	0.02	TRUE	1	FALSE	1	FALSE
Gdańsk–Poznań	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Kraków–Poznań	0.08	FALSE	<0.01	TRUE	0.02	TRUE
Białystok–Szczecin	<0.01	TRUE	1	FALSE	1	FALSE
Gdańsk–Szczecin	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Kraków–Szczecin	0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Poznań–Szczecin	1	FALSE	0.38	FALSE	1	FALSE
Białystok–Warsaw	<0.01	TRUE	1	FALSE	1	FALSE
Gdańsk–Warsaw	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Kraków–Warsaw	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Poznań–Warsaw	1	FALSE	0.09	FALSE	0.76	FALSE
Szczecin–Warsaw	1	FALSE	0.99	FALSE	0.85	FALSE
Białystok–Wrocław	<0.01	TRUE	0.02	TRUE	<0.01	TRUE
Gdańsk–Wrocław	<0.01	TRUE	1	FALSE	1	FALSE
Kraków–Wrocław	<0.01	TRUE	1	FALSE	0.33	FALSE
Poznań–Wrocław	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Szczecin–Wrocław	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE
Warsaw–Wrocław	<0.01	TRUE	<0.01	TRUE	<0.01	TRUE

. The results show that the integration of public transport with TCZs varies substantially across urban contexts.

The hypothesis was also verified regarding whether the share of a route within a TCZ depends on the type of bus line. The study analysed regular lines (stopping at all stops along the route), lines skipping selected stops (accelerated, fast, express), night lines, and lines extending beyond the city limits (suburban, zone, and supplementary lines). The distribution of shares is presented in Figure 6, and the results of Dunn’s test are shown in

Table 15. Verification of the hypothesis regarding differences in the shares of bus route lengths depending on the city.

Comparison	Black	Red	Blue	Green
Białystok–Gdańsk	TRUE			FALSE
Białystok–Kraków	TRUE			FALSE
Gdańsk–Kraków	FALSE		FALSE	FALSE
Białystok–Poznań	FALSE			FALSE
Gdańsk–Poznań	TRUE		FALSE	FALSE
Kraków–Poznań	TRUE		FALSE	FALSE
Białystok–Szczecin	FALSE			FALSE
Gdańsk–Szczecin	TRUE		TRUE	FALSE
Kraków–Szczecin	TRUE	FALSE	TRUE	FALSE
Poznań–Szczecin	FALSE		TRUE	FALSE
Białystok–Warsaw	FALSE			FALSE
Gdańsk–Warsaw	TRUE		TRUE	FALSE
Kraków–Warsaw	TRUE	FALSE	TRUE	FALSE
Poznań–Warsaw	FALSE		TRUE	FALSE
Szczecin–Warsaw	FALSE	FALSE	FALSE	FALSE
Białystok–Wrocław	TRUE			FALSE
Gdańsk–Wrocław	FALSE		FALSE	FALSE
Kraków–Wrocław	FALSE	FALSE	FALSE	FALSE
Poznań–Wrocław	TRUE		FALSE	FALSE
Szczecin–Wrocław	TRUE	FALSE	TRUE	FALSE
Warsaw–Wrocław	TRUE	FALSE	TRUE	FALSE

.

5. Discussion

The third scenario marked the most zones, where their boundaries were roads of the appropriate Tables (2, 3, or 4). Most zones were marked out in Warsaw (up to 457), and the fewest were in Bydgoszcz (80 zones). The fewest high-category roads are in Wrocław, which resulted in only 29 zones, while the fewest high-class roads are in Gdańsk, with 66 zones. The most similar divisions in terms of road class and category were obtained in Bydgoszcz (67 vs. 73 TCZs), Lublin (77 vs. 85 TCZs), Poznań (133 vs. 143 TCZs), Szczecin (136 vs. 132 TCZs), and Warsaw (430 vs. 431 TCZs), which may indicate the proper hierarchy of the road network in these cities. Significant differences were observed in both divisions in Wrocław (97 vs. 29 TCZs), Gdańsk (66 vs. 83 TCZs), Kraków (87 vs. 131 TCZs), and Łódź (210 vs. 261 TCZs).

The largest zones in terms of area were designated in Szczecin (maximum 103.93 km²) and Gdańsk (maximum 79.13 km²), which results from the large areas of these cities and the extensive forest or water areas within their borders. The smallest of the largest zones was designated in Białystok (8.58 km²), the smallest city in terms of total area. The average area of the zone was the largest in Wrocław and Gdańsk (2.69 and 2.48 km²), while the smallest was in Łódź (0.89 km²) and Warsaw (1.00 km²), which may indicate, on the one hand, a shortage, and on the other hand, a significant development in the network of high-class or category roads. This means the method of TCZ division could lead to different influences on traffic in cities, while creating the zones concept in the planning framework, and should also be linked to the level of road network development in a specific city.

Diverse road networks characterise large cities in Poland. Potential traffic-calmed zones, determined based on the road network of class Z and higher for the six largest cities (excluding Warsaw), are presented in Figure 7. It confirms the above observations regarding the density and size of the zones designated by the algorithm. In addition, the road network of cities must also be viewed from the perspective of the areas served, terrain, and natural obstacles, such as rivers, lakes, other water bodies, railway lines, forests, and industrial areas. In all cities, the railway network has a significant influence on the road network. In addition, the Vistula River determines the Warsaw and Kraków networks. In Wrocław–Odra channels; in Poznań, the Warta River; in Szczecin–Odra channels and Dąbie lake; in the Gdańsk–Martwa Wisła channel and Gdańsk bay. Only in Białystok is the river’s influence minor, due to the small size of the Biała river.

The smallest share of bus lines running through the TCZ was obtained in the variant of designating zones according to road classes and categories, which, in the context of reducing the impact of zones on bus traffic, should be considered the most favourable scenario (Table 5, Table 6 and Table 7). Despite this, in Wrocław, all bus lines ran through TCZs (100%), in Kraków, 98%, and in Szczecin, 91%. The smallest share was obtained in Warsaw (75%) and Gdańsk (57%)—Figure 5. This may indicate how bus lines penetrate the city area and provide coverage of the high-hierarchy street network (Warsaw and Gdańsk) or enter local access streets (Kraków or Szczecin). Pearson’s test showed that the passage of bus lines in Białystok and Wrocław through TCZs may have a significant impact.

In the case of Poland, the division of zones by road category (administrator) has a significantly more serious impact on the share of PT line routes in TCZs than the division by road class. According to the above principles, the division brings about the smallest share of travel routes in zones, which means the highest density of zones and their largest number.

TCZs have the most significant impact on regular lines, to a lesser extent on express and night, and the least on suburban lines. This results from the routing of these connections, where fast, night, or suburban connections usually constitute cross-city or radial lines. Among regular lines, there are many circular, feeder, and meandering lines within, among others, housing estates. Short lines (up to 10 km) also have a significantly greater share of routes in zones than long lines (over 20 km), which is again a result of the nature of the routing of connections (Figure 3). Short routes are characteristic of circular or feeder lines, while long routes are characteristic of cross-city or radial lines. This dependence does not apply to Wrocław and Białystok. City policies and design principles for bus lines should include the road class and category while preparing a new line.

The most significant number of suburban lines with a large share of the route in the TCZ was counted in Kraków and Gdańsk, similar to night and regular lines. The smallest shares are found in the cases of Szczecin, Warsaw, and Białystok. In four cities, there are fast lines, where Warsaw and Szczecin’s share of journeys in the TCZ is small, and larger in Wrocław and Kraków (Figure 6).

Estimating changes in buses’ fuel or energy consumption can be supported by the data of vehicles’ daily kilometres inside and outside TCZs (

Table 16. Share of daily distance of bus operations inside and outside of TCZs, divided by road class and category—data for 11 July 2023.

City	Daily Distance in TCZ [km]	Daily Distance Outside of TCZ [km]	Share of Daily Distance in TCZ [−]
Kraków	16,669	112,664	0.128
Poznań	9031	86,294	0.094
Warsaw	21,267	316,340	0.062
Białystok	3589	37,256	0.087
Gdańsk	9871	74,598	0.116
Wrocław	25,573	78,899	0.244
Szczecin	4032	54,728	0.068

). In the division by road class and category, for a hypothetical working day—11 July 2023—the most significant share of bus-driven kilometres in the mean value would be influenced by the TCZs in Warsaw (0.062). Around 12% of bus-driven kilometres in the mean value would be influenced by the introduction of TCZs.

A limitation of this study is that the analysis is based on GTFS timetable data. This study could be expanded to include data showing actual public transport vehicle traffic. However, not all cities provide such data. It should be noted that static GTFS data as of the study date were only available for seven of the ten largest Polish cities. Data on actual vehicle traffic would be available for even fewer cities. Additionally, errors and omissions in the data (e.g., lack of vehicle position transmission) must be taken into account. Similarly, the lack of GTFS data for many smaller cities will prevent such a study from being conducted for many county cities.

Considering the impact of TCZs on vehicle speeds is challenging at the general level used in this article. Analysing bus speed profiles for individual bus sections between stops would be necessary. Such data is not available even for the largest cities. Vehicle position is theoretically recorded every 10 s, but practice shows that these readings are often two or three times per minute. This measurement resolution does not allow for determining the instantaneous speed to analyse the impact of TCZs on speeds and travel times.

Lowering the maximum speed of buses due to TCZ implementation would also mean a longer duration of travel for passengers, which could result in a negative impact on lower bus patronage and undermine the goal of promoting sustainable urban transport. Additionally, it could lead to the need for an increase in the bus fleet to service the same network. The potential decrease in PT ridership resulting from lower vehicle speeds or reduced route directness within TCZs could be mitigated through complementary measures that improve service attractiveness. These include bus priority at intersections, whether with right-of-way on directions with bus traffic or in the traffic lights programmes (but in Warsaw, only 3% of traffic lights are inside designed TCZs) or cushion speed bumps. Further study should be conducted to determine how the implementation of TCZs influences the commercial speed of buses, based on real service data before and after TCZ implementation.

6. Conclusions

The main reason for implementing traffic calming zones (TCZs), as shown in the literature review, is to improve traffic safety (decrease in fatalities, injuries, accidents, and collisions), reduce speed, enhance compliance with speed limits, and promote active mobility. Still, the goals also include increasing the use of public transport (PT). Meanwhile, the attractiveness of ground-based PT, especially buses, may decrease due to lower travel speeds through the zones. This study aimed to test the methodology for designating TCZs according to a three-level hierarchy of roads in cities, concerning their impact on buses. The hypothesis was that TCZs can significantly affect city PT bus traffic.

A method was developed based on GIS map data, an algorithm written in R and GTFS timetable data, which allowed for checking the share of bus routes running through TCZs designated according to unified principles. The study’s results indicate that the more TCZs we designate, the less they will impact bus traffic. The zones have the most significant impact on regular lines, less on express and night lines, and the least on suburban lines. This is happening because of the nature of designing each type of bus line—regular lines often have routes that wind through residential streets, and suburban lines in city areas are routed straight through main road corridors leading to the city borders. If PT is to remain a competitive means of transport, TCZs should not be introduced in entire urban areas, as was the case in Spain.

The research methodology enables the determination of TCZs using GIS tools, GTFS data, and an algorithm written in R. These are the three essential elements for dividing the area into potential TCZs. The simplest division may concern individual classes or categories of roads (e.g., unified nomenclature, such as that used by OpenStreetMap). Still, more advanced methods, such as those used in the project for Warsaw, may include the distance to the nearest road of a higher class or category, the type of development near the road, and the presence of agricultural and tourist areas, among others.

This research contributes to a relatively underexplored area in the literature by providing a data-driven framework to evaluate the impact of TCZs on PT. It identifies a critical trade-off between urban safety initiatives and the operational effectiveness of sustainable transport modes. One of the study’s limitations is that it primarily focuses on static route data and has not yet incorporated dynamic performance metrics, such as real-time travel delays and passenger satisfaction indicators. Furthermore, the availability and completeness of GTFS data varied across cities, which limited the feasibility of an entirely uniform comparison.

Despite the documented benefits of TCZs in terms of road safety and environmental improvements, their interaction with PT infrastructure requires careful planning and design. This study emphasises the importance of tailoring TCMs to accommodate buses, particularly by avoiding vertical deflections, such as speed humps, in favour of more bus-friendly solutions, such as speed cushions or raised intersections with appropriate geometry. Moreover, it underscores the need for differentiated strategies based on the length and character of bus routes, ensuring that extensive low-speed segments do not disproportionately burden long-distance or high-frequency lines.

As the best scenario for the introduction of TCZs, minimising their influence on bus traffic involves dividing zones by road class and category. However, the number of lines going through TCZs was still significant, with all cities having a percentage higher than 57%. The shortest lines are more commonly found within TCZs than longer connections. It should be noted that TCZs are desired by residents living along the roads; however, other functions of the transport corridors should also be taken into consideration, especially on transit and local roads within the road hierarchy.

Creating urban road network policies based on a three-level hierarchy of roads and TCZs at the lowest level should also lead to the update of design guides and principles for bus lines. The process should be connected, and the lines of the highest importance should be routed outside of the TCZs. Those of lower meaning, such as feeder or meandering lines, can be routed through TCZs.

As a side effect of the method, significant differences in the results were observed in the case of four out of nine cities when comparing the number of TCZs marked using the technique based on road class or category. This indicates an inappropriate road structure, which is currently managed in Wrocław, Gdańsk, Kraków, and Łódź. Significant development of the road networks was found in Łódź and Warsaw, because the method divided them by the average value in the smallest zones.

A further area of research may be the assessment of the actual change in bus travel speed after implementing the TCZs and the potential for estimating the decrease in similar projects. The maximum speed of buses in TCZs is reduced, which has a negative effect, resulting in a lower commercial speed [56]. Lower speeds in models result in higher fuel consumption (for diesel buses) [57] or energy consumption (for electric buses) [58] per kilometre driven. However, in TCZs, there is also a positive effect on commercial speed brought about by solutions such as deleting traffic lights, transforming intersections to give right-of-way to bus routes, and constructing roundabouts. Those solutions can lead to fewer stoppages and smoother traffic. The influence on fuel and energy consumption is unclear and requires further studies, but this study highlights that around 12% of bus kilometres would be affected by TCZ implementation.

It should be noted that zones may refer to the introduction of Tempo 30 zones, residential areas, or shared spaces. Another TCZ study could address the impact of these three specific solutions. The literature review only indicated that there are TCMs that more or less affect bus traffic in the TCZ, especially preferred to use with bus traffic are speed cushions over humps or bumps, but giving the roads with bus traffic right-of-way is also an additional solution that could reduce the TCZ’s influence on bus traffic. Future research should explore more dynamic data sources, such as AVL (Automatic Vehicle Location) data or smart card transaction logs, to evaluate how TCZs affect travel time reliability and user experience in real operational contexts. Additionally, further investigations could address the design of integrated policies that harmonise TCZ goals with PT efficiency, fostering both safety and a modal shift toward sustainable urban mobility.

Tram tracks separated from the roadway should not be included in TCZs. Traffic signs used in Poland allow for increased tram speeds on separated tracks, which enables maintaining high tram speeds. The impact of TCZs on cycling and other micromobility modes is minor. Electric bicycles are assisted up to 25 km/h, and few urban cyclists travel faster than 30 km/h. Electric scooters in Poland can travel at speeds of up to 20 km/h, so the 30 km/h speed limit does not affect their use.

The main contribution of this study is the integration of an automated GIS-based algorithm for delineating traffic-calmed zones with transport analysis, applied consistently across multiple major Polish cities. The methodology was initially developed in advance of Warsaw’s local approach and subsequently tested in other urban contexts to evaluate its broader applicability and transferability. The method was then adapted to local conditions in Warsaw. With additional specific parameters, it already formed the basis for the adoption of TCZs in the capital city of Poland.

In conclusion, while TCZs play a vital role in creating safer and more liveable urban environments, a balanced and data-driven approach is essential to ensure they do not unintentionally compromise the functionality and appeal of PT systems. Coordinated planning between traffic engineers, urban planners, and transport authorities is necessary to achieve urban mobility that is both safe and efficient.