Traffic Flow Theory, Capacity and Operations

Introduction

Traffic engineering concerns the safe, efficient and equitable movement of people and goods through transport networks. At postgraduate level, traffic operations must be understood not merely as the adjustment of signal timings or the calculation of road capacity, but as the interaction of human behaviour, vehicle technology, infrastructure, demand, control systems and institutional policy.

Traffic flow theory provides models for explaining how vehicles distribute themselves in space and time. These models support the analysis of speed, flow, density, headways, queues, bottlenecks, intersections, public transport priority and congestion-management strategies. Because traffic is a dynamic system, a treatment that improves one location may transfer congestion elsewhere or change travel behaviour over time.

This chapter develops the fundamental relationships among speed, flow and density; examines car-following and shock waves; explains capacity and level of service; and applies these ideas to intersections, signal timing, roundabouts, corridors, parking and congestion management.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Define and distinguish traffic flow, speed, density, headway and spacing.

2. Apply the fundamental speed–flow–density relationship and interpret traffic-flow diagrams.

3. Explain car-following, lane-changing, shock waves and queue formation.

4. Estimate roadway and intersection capacity under stated assumptions.

5. Interpret level-of-service concepts and their limitations.

6. Analyse signal timing, roundabouts and corridor operations.

7. Explain the relationship between parking policy, demand and congestion.

8. Evaluate congestion-management strategies using operational, environmental and equity criteria.

1. Fundamental Traffic Variables

1.1 Flow

Traffic flow, denoted by q, is the number of vehicles passing a reference point during a specified time interval. Its usual units are vehicles per hour (veh/h). If N vehicles pass a point in time T hours,

q = N/T.

Flow is a rate, not a total count. A count of 600 vehicles in 15 minutes corresponds to a flow rate of 2,400 veh/h, assuming that the interval is used to represent the prevailing rate.

1.2 Speed

Speed is the distance travelled per unit time. In traffic engineering, two averages are especially important:

• Time-mean speed is the arithmetic mean of speeds observed at a point. It is appropriate for spot-speed observations.
• Space-mean speed is the harmonic mean of vehicle speeds over a road section. It represents the average speed experienced over space and is the appropriate speed for relating flow and density.

For n observed speeds v1, v2, ..., vn, the space-mean speed is

u_s = n / (1/v1 + 1/v2 + ... + 1/vn).

The harmonic mean is lower than or equal to the arithmetic mean unless all speeds are equal. Confusing these two means can produce biased estimates of travel time and density.

1.3 Density and spacing

Density k is the number of vehicles occupying a unit length of roadway, normally expressed in veh/km/lane. If N vehicles occupy a section of length L kilometres,

k = N/L.

Spacing is the distance between corresponding points on successive vehicles, while time headway is the elapsed time between their arrivals at a reference point. Under stationary and homogeneous conditions, average spacing and density are related by

s = 1/k,

when compatible units are used. For example, a density of 40 veh/km/lane corresponds to an average spacing of 0.025 km, or 25 m, between vehicles.

1.4 The fundamental identity

For a homogeneous traffic stream,

q = k u,

where q is flow, k is density and u is space-mean speed. Unit consistency is essential: veh/km multiplied by km/h gives veh/h. This identity is a conservation relationship rather than a particular behavioural model. It remains valid whether traffic is free-flowing or congested, provided the variables refer to the same stream and location or sufficiently consistent time-space conditions.

2. Speed–Flow–Density Models

2.1 The fundamental diagram

The fundamental diagram describes the relationship among flow, speed and density. Typical traffic states include:

• Free flow: vehicles operate with limited interaction and speed is primarily controlled by roadway geometry, speed limits, weather and driver preferences.
• Transitional or near-capacity flow: interactions become substantial and small disturbances may produce instability.
• Congested flow: speed decreases as density increases, queues form and disturbances may propagate upstream.

At zero density, flow is zero. At jam density, vehicles are closely packed and speed and flow approach zero. Between these extremes, flow reaches a maximum called capacity.

2.2 Greenshields’ linear speed–density model

A classical model assumes a linear relationship between speed and density:

u = u_f(1 - k/k_j),

where u_f is free-flow speed and k_j is jam density. Combining this with q = ku gives

q = u_f k(1 - k/k_j).

The flow–density relationship is therefore parabolic. Differentiating with respect to k gives the density at maximum flow:

k_c = k_j/2.

The corresponding capacity is

q_c = u_f k_j/4,

and the speed at capacity is

u_c = u_f/2.

These results are properties of the Greenshields model, not universal laws. Real traffic often exhibits asymmetric, unstable and hysteretic behaviour that the model does not represent.

Worked Example 1: Fundamental-diagram calculation

A lane is represented by a Greenshields model with free-flow speed 100 km/h and jam density 150 veh/km/lane.

The critical density is

k_c = 150/2 = 75 veh/km/lane.

Capacity is

q_c = (100 × 150)/4 = 3,750 veh/h/lane.

Speed at capacity is

u_c = 100/2 = 50 km/h.

The result is an idealized theoretical capacity. Field capacity would normally be lower or different because of heavy vehicles, access friction, lane changing, weather and driver response.

2.3 Alternative models

The Greenberg model assumes a logarithmic speed–density relationship and is sometimes useful for congested conditions, but it predicts unrealistic behaviour at low density. Underwood’s exponential model can represent free-flow conditions more plausibly but has its own limiting assumptions. Modern practice frequently uses empirical, piecewise or simulation-based models calibrated with detector, probe-vehicle and trajectory data.

Model selection should be based on the intended use, data quality and operating regime. A simple model may be preferable for planning sensitivity analysis, whereas microscopic simulation may be justified for complex intersections, multimodal interactions or connected-vehicle operations. Calibration and validation against independent observations are essential.

3. Car-Following and Traffic Stability

3.1 Car-following behaviour

Car-following describes how a driver adjusts speed in response to a leading vehicle. A driver generally attempts to maintain a desired speed and a safe gap while responding to the leader’s speed, acceleration and relative position. Important concepts include:

• Reaction time: time between perceiving a stimulus and initiating a response.
• Desired time headway: preferred time gap between successive vehicles.
• Net gap: distance from the front of the following vehicle to the rear of the leader.
• Minimum or standstill gap: space retained when vehicles are stopped.
• Perception uncertainty: variation caused by attention, visibility, workload and expectation.

A simple stimulus-response model may be expressed as

a_n(t + T) = α [u_{n-1}(t) - u_n(t)] / [x_{n-1}(t) - x_n(t) - l_{n-1}],

where a_n is the acceleration of vehicle n, T is reaction time, α is a sensitivity parameter, u is speed, x is position and l is vehicle length. This is illustrative of a delayed relative-speed model; actual models may include desired spacing, nonlinear response and bounded acceleration.

3.2 Stability and disturbances

A small disturbance, such as a leader braking briefly, may be damped, maintained or amplified as it travels through a platoon. Amplification produces stop-and-go waves even when no obstruction remains. Stability depends on reaction time, acceleration capability, desired gaps, heterogeneity and control systems.

Adaptive cruise control can reduce some disturbances by shortening response delays and regulating gaps, but mixed traffic containing conventional and automated vehicles may exhibit new instabilities. Consequently, claims about automated vehicles increasing capacity must be supported by assumptions concerning market penetration, control policy, lane use and safety margins.

4. Shock Waves and Queue Dynamics

A traffic shock wave is a boundary between two traffic states that moves through space and time. If state 1 has flow q1 and density k1 and state 2 has flow q2 and density k2, the wave speed is

w = (q2 - q1)/(k2 - k1).

The units are consistent: (veh/h)/(veh/km) = km/h. The sign indicates direction. A negative value denotes movement upstream relative to the roadway coordinate convention.

4.1 Queue formation and dissipation

When demand exceeds discharge capacity at a bottleneck, vehicles accumulate upstream. The upstream boundary of a queue often moves backward, although individual vehicles move forward. When demand falls below discharge capacity, the queue dissipates from its downstream end and its upstream boundary moves downstream.

For an initially empty queue with arrival rate λ and service rate μ, where both are approximately constant and λ > μ, the queue grows at λ − μ vehicles per unit time. If the queue lasts for duration t, the accumulated vehicles are

N_q = (λ - μ)t.

This expression is valid only while the arrival and service rates remain constant and the queue does not encounter another constraint.

Worked Example 2: Shock-wave speed

Before a bottleneck, traffic has q1 = 1,800 veh/h and k1 = 30 veh/km. In the congested state, q2 = 900 veh/h and k2 = 90 veh/km. Taking state 1 as free flow and state 2 as congested,

w = (900 - 1,800)/(90 - 30) = -900/60 = -15 km/h.

The queue boundary travels upstream at 15 km/h. This does not mean that vehicles move backward; it describes the motion of the traffic-state boundary.

5. Capacity and Level of Service

5.1 Capacity

Capacity is the maximum sustainable hourly rate at which persons or vehicles can reasonably be expected to traverse a point or uniform section under prevailing roadway, traffic and control conditions. It is not a fixed physical constant. Capacity varies with lane width, grades, curves, access points, heavy vehicles, weather, driver population, incident conditions, speed management and control delay.

Capacity analysis should distinguish:

• Basic or ideal capacity under standardized assumptions.
• Adjusted capacity under prevailing traffic and geometric conditions.
• Observed capacity, which may be affected by random demand and temporary bottlenecks.
• Person capacity, which accounts for vehicle occupancy and public transport.

For a signalized approach, an idealized lane-group capacity is commonly represented as

c = s(g/C),

where s is saturation flow in veh/h of green, g is effective green time and C is cycle length. This equation assumes approximately stable saturation discharge during effective green and does not by itself account for start-up lost time, clearance lost time, spillback or downstream restrictions. In practice, lost time is incorporated through the effective green and cycle definition or through a more detailed signal model.

5.2 Degree of saturation

The volume-to-capacity ratio is

X = v/c,

where v is demand flow and c is capacity. Values below 1 do not guarantee satisfactory operation: variability can still produce queues, and oversaturated conditions may persist after demand falls. Values greater than 1 indicate that demand exceeds sustainable service capacity for the analysis period.

5.3 Level of service

Level of service (LOS) is a qualitative or semi-quantitative description of operating conditions. Depending on facility type, measures may include travel speed, density, delay, queue, freedom to manoeuvre, reliability and pedestrian comfort. LOS categories are useful for communication and design screening, but they should not be treated as universal measures of transport quality.

Limitations include:

• Vehicle delay may penalize high-occupancy modes less or more than intended depending on the method.
• A road can achieve good vehicle LOS while creating poor conditions for pedestrians and cyclists.
• LOS is generally a snapshot and may not represent reliability or peak-period variability.
• Capacity-based targets can encourage road expansion that induces additional demand.
• Local cultural, environmental and accessibility priorities may not align with a single letter grade.

Postgraduate analysis should therefore report the underlying measures as well as the LOS category.

6. Signalized Intersections

6.1 Operational principles

A signalized intersection allocates right-of-way through phases, intervals and coordination. Important terms include cycle length, green interval, yellow change interval, all-red clearance, effective green, lost time, offset and saturation flow.

A signal plan must balance competing objectives: safety, delay, queue storage, progression, pedestrian service, public transport priority and emissions. Increasing green time for one movement reduces available time for another unless the cycle is lengthened or other phases are removed.

6.2 Cycle length and timing

For an isolated intersection, a classical minimum-delay approach relates cycle length to lost time and critical flow ratios. A commonly used form of Webster’s estimate is

C_0 = (1.5L + 5)/(1 - Y),

where C0 is estimated cycle length in seconds, L is total lost time per cycle in seconds and Y is the sum of critical flow ratios for the phases. The formula is an approximation derived under simplified assumptions, including relatively stable arrivals and a delay-minimization objective. It should not be used mechanically where pedestrian constraints, coordination, oversaturation or unusual phase structures dominate.

Effective green allocation is often based on critical flow ratios:

g_i = (y_i/Y)(C-L),

where yi is the critical flow ratio for phase i. Actual design must include minimum pedestrian times, clearance requirements, turn conflicts, queue storage and safety policy.

6.3 Coordination and progression

On an arterial, offsets determine the time relationship between signals. Progression can reduce stops for a platoon travelling at a target speed, but it may increase delay on side streets or for pedestrians. Dynamic coordination using detector and probe data can respond to changing demand, incidents and transit priority requests. Benefits should be assessed corridor-wide rather than at one intersection only.

7. Unsignalized Intersections and Roundabouts

At unsignalized intersections, capacity depends on priority rules, accepted gaps, follow-up headways, approach geometry and conflicting flows. Gap-acceptance models assume that a driver accepts a gap when it exceeds a critical threshold. The critical gap varies by driver, vehicle type, turning movement, visibility, speed and perceived risk.

Roundabout capacity is influenced by circulating flow, entry geometry, lane configuration, pedestrian crossings and driver yielding behaviour. Roundabouts generally reduce severe right-angle conflicts and may reduce delay at moderate, balanced flows. They are not automatically superior: high circulating demand, constrained pedestrian crossings, large turning movements, poor visibility or inadequate storage can cause failure.

A roundabout assessment should examine entry queues, circulating queues, pedestrian accessibility, bicycle treatment, heavy-vehicle swept paths, emergency access and spillback to adjacent junctions. Capacity models must be calibrated to local yielding behaviour where possible.

8. Corridor Operations

A corridor is a connected sequence of links and junctions whose performance is interdependent. Key measures include travel time, delay, stops, reliability, queue spillback, person throughput, transit speed, emissions and safety exposure.

Effective corridor management may combine:

• Signal coordination and adaptive control.
• Turn restrictions and access management.
• Transit signal priority and queue-jump lanes.
• Reversible or managed lanes where justified.
• Incident detection and rapid clearance.
• Dynamic route guidance and traveller information.
• Freight delivery management.
• Pedestrian and cycling priority.

The bottleneck principle is central: improving a non-bottleneck location may have little network benefit if the limiting constraint remains. Queue spillback can block upstream intersections, creating gridlock. A corridor model should therefore represent storage capacity, not only link capacity.

9. Parking and Its Relationship with Traffic Operations

Parking is part of the transport system, not a separate land-use issue. Searching for parking generates circulating traffic, while curbside loading, illegal stopping and parking manoeuvres reduce effective capacity. Parking supply and price influence destination choice, vehicle ownership, trip frequency and mode choice.

Operational measures include occupancy, turnover, duration, search time, illegal parking and revenue. A high occupancy rate is not always desirable: when occupancy approaches full utilization, search traffic and double parking may increase sharply. Pricing, time limits, residential permits, loading management and off-street supply should be designed together.

Parking policy also raises equity concerns. High charges may burden lower-income users unless alternatives and targeted concessions exist. Free parking supplied by employers or developers can function as a travel subsidy and may encourage automobile dependence. Curb allocation should explicitly consider pedestrians, public transport, freight, emergency access, accessibility spaces and micro-mobility.

10. Congestion Management

Congestion management seeks to reduce the social and economic costs of unreliable or excessive travel demand. Strategies can be classified as follows:

Demand management

Examples include congestion pricing, parking pricing, teleworking, flexible working hours, travel plans, ridesharing and improved public transport. Pricing is economically efficient when charges reflect congestion externalities, but it requires transparent objectives, reliable technology, privacy safeguards and measures to protect vulnerable users.

Supply and operational management

Examples include bottleneck removal, ramp metering, hard-shoulder running where safe, incident management, signal optimization, managed lanes and transit priority. Capacity expansion may be appropriate in selected locations, but induced demand can reduce long-term benefits.

Information and technology

Probe vehicles, connected infrastructure, automatic number-plate recognition, computer vision and digital twins can support detection and prediction. Artificial intelligence can estimate travel times, identify incidents and optimize signals, but deployment requires representative training data, cybersecurity, explainability, human oversight and protection against discriminatory outcomes. Video analytics should comply with privacy and data-retention requirements.

Evaluation framework

A congestion strategy should be assessed using multiple criteria:

• Total and person-hours of delay.
• Travel-time reliability.
• Safety and conflict exposure.
• Emissions and energy use.
• Public transport, walking and cycling performance.
• Distributional effects across income, disability and location.
• Capital, operating and enforcement costs.
• Network resilience under incidents and extreme weather.

A strategy that reduces vehicle delay but increases pedestrian risk or displaces congestion into a disadvantaged community is not operationally or ethically optimal.

11. Illustrative Case: Signal-Corridor Bottleneck and Transit Priority

Background

Consider an illustrative four-intersection urban corridor carrying 2,000 veh/h in the dominant direction during the evening peak. Buses operate every eight minutes. The downstream intersection has limited storage between signals, and queues occasionally block the upstream junction.

Problem

Field observations show that the downstream approach receives a demand of 1,050 veh/h against an estimated capacity of 900 veh/h during the peak. The approach is therefore oversaturated, with X = 1,050/900 = 1.17. Average bus travel time is unreliable because buses are trapped in the queue. Extending green at the upstream signal worsens downstream spillback.

Analysis

The principal constraint is downstream discharge and storage, not upstream green time. Detector data show that queues begin to block the preceding junction after approximately 12 minutes. The corridor team compares three alternatives: signal retiming, transit signal priority and a short managed bus lane approaching the bottleneck.

Approach and solution

The adopted strategy is staged:

1. Retiming reallocates green time to the bottleneck approach while preserving minimum pedestrian service.

2. A transit-priority request extends green by up to 8 seconds when a late bus is detected, subject to queue and safety constraints.

3. A bus-only approach lane reduces bus interference from general traffic.

4. Queue detectors inhibit priority when the downstream storage is nearly full.

5. Parking and loading restrictions are enforced near the junction to restore usable lane width.

The design is tested in a calibrated microsimulation and then implemented as a temporary field trial. Performance is measured using person delay, bus reliability, queue spillback, pedestrian delay, emissions and side-street delay.

Outcome

The trial reduces bus travel-time variability and prevents regular blockage of the upstream intersection, although some side-street delay increases. The queue-protection logic is essential: unconditional transit priority would merely transfer congestion into the downstream link. The final plan includes public reporting and periodic review rather than treating the initial settings as permanent.

Learning points

The case demonstrates that capacity is a network and storage phenomenon. It also shows why person throughput may justify priority for buses even when vehicle delay on a minor approach increases. Successful operations require field validation, adaptive controls and transparent evaluation.

12. Limitations, Governance and Best Practice

Traffic-flow models simplify heterogeneous human behaviour. Results may be sensitive to calibration, time interval, detector error, demand variability and assumptions about compliance. Models should not be presented with false precision. Analysts should report uncertainty ranges, test alternative demand scenarios and distinguish observed facts from model outputs.

Good practice includes:

• Use consistent definitions and units.
• Separate demand, capacity and realized flow.
• Check queue storage and spillback.
• Calibrate models with independent data.
• Analyse peak and off-peak periods, incidents and weather.
• Evaluate people movement, not vehicles alone.
• Include walking, cycling, public transport and accessibility.
• Document assumptions and sensitivity tests.
• Protect privacy in automated data collection.
• Engage affected communities before imposing pricing, access or enforcement changes.

Contemporary traffic operations increasingly combine fixed sensors, connected vehicles, public transport data, computer vision and predictive analytics. These tools improve situational awareness but do not replace engineering judgement, safety review or democratic accountability.

Chapter Summary

Traffic flow is governed by the conservation identity q = ku, while the observed relationship between speed, flow and density depends on behavioural and roadway conditions. Fundamental diagrams provide useful idealizations, but real traffic is heterogeneous, dynamic and often unstable. Car-following behaviour explains platoons, reaction effects and stop-and-go waves. Shock-wave analysis explains why queue boundaries move upstream and how bottlenecks form and dissipate.

Capacity is a prevailing-condition measure rather than a fixed physical constant. At intersections, saturation flow, effective green, lost time, phase structure and downstream storage determine performance. LOS provides a communication framework but must be supplemented by reliability, safety, person throughput, accessibility and environmental indicators. Signal coordination, roundabouts and corridor management require network-wide analysis because local improvements can transfer delay or create spillback.

Parking influences circulation, demand and curb capacity. Congestion management combines operational improvements, demand management, pricing, public transport priority, information systems and carefully governed technology. Professional practice requires calibration, uncertainty analysis, ethical consideration, privacy protection and evaluation of distributional impacts.

Key Terms

Flow: Rate at which vehicles pass a reference point, usually veh/h.

Speed: Distance travelled per unit time.

Time-mean speed: Arithmetic mean of spot speeds measured at a point.

Space-mean speed: Harmonic mean speed over a road section.

Density: Number of vehicles occupying a unit length of roadway.

Headway: Time interval between successive vehicles at a point.

Spacing: Distance between successive vehicles along a roadway.

Fundamental diagram: Relationship among traffic speed, flow and density.

Free-flow speed: Speed under low-density conditions with minimal vehicle interaction.

Jam density: Density at which traffic speed and flow approach zero.

Capacity: Maximum sustainable flow under specified prevailing conditions.

Saturation flow: Approximate discharge rate during a fully effective green interval.

Degree of saturation: Demand divided by capacity, v/c.

Level of service: Description of operating quality using measures such as speed, delay or density.

Car-following: Behaviour of a vehicle responding to a leader in the same lane.

Shock wave: Moving boundary between two traffic states.

Queue spillback: Extension of a queue that blocks an upstream link or intersection.

Critical gap: Minimum acceptable gap used in a gap-acceptance decision.

Signal offset: Time relationship between signal controllers along a corridor.

Congestion pricing: Charging strategy intended to manage demand and congestion costs.

Review Questions

Short Answer Questions

1. Distinguish time-mean speed from space-mean speed.

2. State the fundamental relationship among flow, density and speed.

3. What is meant by jam density?

4. Define degree of saturation and explain the meaning of a value greater than one.

5. Identify two limitations of using level of service as the sole performance measure.

Long Answer Questions

1. Explain the fundamental diagram and derive the critical density and capacity under the Greenshields model.

2. Discuss car-following behaviour, reaction time and the development of stop-and-go waves.

3. Explain shock-wave theory and its application to queue formation and dissipation.

4. Describe the principal factors that determine signalized-intersection capacity and cycle length.

5. Evaluate the role of parking policy and congestion pricing in urban traffic management.

Analytical / Application Questions

1. A lane has a free-flow speed of 90 km/h and jam density of 140 veh/km. Using the Greenshields model, calculate critical density and theoretical capacity, and discuss why field capacity may differ.

2. A traffic state changes from 2,400 veh/h and 40 veh/km to 1,000 veh/h and 100 veh/km. Calculate the shock-wave speed and interpret its direction.

3. A corridor has a heavily congested downstream bottleneck and frequent bus delays. Propose an operations strategy, identify possible adverse effects, and specify performance measures for evaluating the intervention.

Multiple Choice Questions with Answers

Question 1

Which equation expresses the fundamental relationship for a homogeneous traffic stream?

A. q = k/u

B. q = ku

C. q = u/k

D. q = k + u

Correct Answer: B. q = ku

Question 2

For the Greenshields model, at what density does theoretical maximum flow occur?

A. Zero density

B. One-quarter of jam density

C. One-half of jam density

D. Jam density

Correct Answer: C. One-half of jam density

Question 3

A shock wave has q1 = 1,800 veh/h, k1 = 30 veh/km, q2 = 900 veh/h and k2 = 90 veh/km. What is its speed?

A. −15 km/h

B. −2 km/h

C. 15 km/h

D. 45 km/h

Correct Answer: A. −15 km/h

Question 4

A demand of 1,050 veh/h is served by a capacity of 900 veh/h. What is the degree of saturation?

A. 0.86

B. 0.95

C. 1.05

D. 1.17

Correct Answer: D. 1.17

Question 5

Which speed measure is the harmonic mean of observed vehicle speeds?

A. Time-mean speed

B. Space-mean speed

C. Design speed

D. Operating speed

Correct Answer: B. Space-mean speed

Question 6

In the simplified signal capacity equation c = s(g/C), what does g/C represent?

A. Lost time per cycle

B. Effective green ratio

C. Critical flow ratio

D. Queue discharge speed

Correct Answer: B. Effective green ratio

Question 7

What normally causes a queue to grow at a bottleneck?

A. Arrival flow remains below discharge capacity

B. Arrival flow exceeds discharge capacity

C. Jam density becomes zero

D. Free-flow speed increases

Correct Answer: B. Arrival flow exceeds discharge capacity

Question 8

Which issue is especially important when evaluating corridor transit signal priority?

A. It can transfer delay to side streets or downstream links

B. It eliminates all intersection delay

C. It requires no detection or control logic

D. It always reduces total vehicle delay

Correct Answer: A. It can transfer delay to side streets or downstream links

Question 9

What is the primary purpose of a critical-gap concept at an unsignalized intersection?

A. To determine pavement thickness

B. To estimate whether a driver accepts an available gap

C. To calculate jam density

D. To set parking prices

Correct Answer: B. To estimate whether a driver accepts an available gap

Question 10

Why can very high parking occupancy increase congestion?

A. It eliminates parking search traffic

B. It encourages vehicles to circulate while searching for spaces

C. It guarantees higher public transport speed

D. It reduces curbside conflicts automatically

Correct Answer: B. It encourages vehicles to circulate while searching for spaces