Question 1 of 7: Definitions and discussion

Question 1 — Definitions and discussion (a) to (e), 4 marks each — 20 marks

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a) Bicycle lanes versus bicycle paths

Both are cycling facilities, and the distinction between them is one of physical separation from motor traffic, which in turn governs their capacity, their safety performance and the design standards that apply. The MUTCDC and the TAC Geometric Design Guide classify cycling facilities on exactly this axis.

A bicycle lane is a portion of the roadway itself, designated for the exclusive or preferential use of cyclists by pavement marking and signing, but sharing the same paved surface and the same vertical plane as the adjacent motor-vehicle lanes. A typical Canadian urban bike lane is 1.5 m to 1.8 m wide, is delineated by a solid white line (dashed through intersection approaches and at driveways where turning traffic must cross it), and carries one-way flow in the same direction as the adjacent traffic. Because it lies within the roadway, a bicycle lane is subject to the roadway's design speed, its drainage and its intersection control; cyclists in it are legally vehicle operators, they interact continuously with turning, merging and parking manoeuvres, and the facility offers no protection against a lateral encroachment. Its advantages are low cost, ease of retrofitting into an existing cross-section, continuity through intersections and the fact that cyclists remain visible to drivers in the drivers' normal scan pattern.

A bicycle path (a shared-use or multi-use path) is a facility physically separated from the roadway by an open space, a boulevard or a barrier, or located on an independent right-of-way altogether. Widths are larger — commonly 3.0 m to 4.0 m for two-way operation — because the path must accommodate two-way cycling and, in the shared-use case, pedestrians, runners and other users at very different speeds. A path removes the continuous conflict with motor traffic and is therefore far more attractive to inexperienced and recreational cyclists, but it introduces a different and often under-appreciated risk: at every point where the path crosses a roadway or a driveway the cyclist re-enters the traffic stream at an unexpected location and frequently from an unexpected direction, and drivers do not scan for a two-way cycle movement on what they read as a sidewalk. Intersection treatment is consequently the critical design element for paths, and Canadian practice requires explicit crossing control (crossrides, signal detection, or a transition back into the roadway) rather than allowing the path simply to end.

In summary: a bicycle lane is a marked part of the roadway that buys continuity and low cost at the price of continuous exposure to traffic; a bicycle path is a separated facility that buys comfort and mid-block safety at the price of complex, higher-risk intersection crossings.

(b) Cordon counts versus screenline counts

Both are area-wide volume studies in which counting stations are placed on a line rather than at isolated points, and in both the quantity of interest is the traffic crossing that line rather than the traffic at any one station. They differ in the geometry of the line and therefore in the question they answer.

A cordon count places stations on a closed line completely enclosing an area — a central business district, a campus, a hospital precinct or a special-generator site. Every roadway crossing the cordon is counted, directionally, so the survey yields the total inbound and outbound volume for the enclosed area. Its characteristic products are the accumulation of vehicles or persons within the cordon as a function of time (obtained by integrating inbound minus outbound), the peak accumulation, and the parking demand and turnover implied by that accumulation. A cordon count is the standard basis for downtown parking studies, for sizing transit and pedestrian facilities serving a district, and for establishing trip-generation totals for a special generator. Because the line is closed, a cordon count is self-checking: over a full day, inbound and outbound totals must very nearly balance, and a large imbalance signals a missed roadway or a counter failure.

A screenline count places stations on an open line — typically an imaginary line following a natural or artificial barrier such as a river, a rail corridor or a ridge — that divides a study area into two parts without enclosing either. Every roadway crossing the line is counted directionally, giving the total interchange of traffic between the two halves. Screenlines answer a travel-demand question rather than an accumulation question: they measure the aggregate flow across a corridor, they are used to detect long-term shifts in travel patterns, and above all they are the primary tool for validating and calibrating a travel-demand model, because a model's assigned volumes summed across a screenline can be compared directly with a single robust field measurement. Barriers are chosen as screenlines precisely because they limit the number of crossings, so a small number of stations captures the entire interchange.

In short, a cordon is closed and measures what enters, leaves and accumulates inside an area; a screenline is open and measures the interchange between two areas. The two are complementary and are frequently run together, the screenline providing the corridor-level control totals against which the cordon's district-level detail is checked.

(c) Leading protected phase versus lagging protected phase

A protected phase gives a movement an exclusive right-of-way indication — conventionally a green arrow — so that no conflicting movement may legally proceed. For left turns the design question is not whether to protect but where in the cycle to place the protection relative to the adjacent through movement, and that placement is exactly the leading-versus-lagging distinction.

A leading protected phase serves the left turn before the adjacent through movement. The phase opens with a green arrow for the left turn while opposing through traffic is held on red, and only when the arrow terminates does the through green begin. This is by far the more common arrangement in Canadian practice, so drivers expect it; it empties the left-turn bay at the moment its queue is longest; and it separates the turn cleanly from the opposing through platoon. Its principal drawback is the exposure it creates to the yellow trap if the opposing left turn is permitted while the subject left still shows a permissive indication.

A lagging protected phase serves the left turn after the through movement: both through movements run first, and the arrow is displayed at the tail of the phase once the opposing through movement has been terminated. Lagging protection favours progression along an arterial, because the through band can be released without an interruption at the head of the phase and the left turns are cleaned up afterwards; it also permits a lead–lag arrangement in which one direction leads and the other lags, which is a powerful degree of freedom in time–space band optimisation on a coordinated corridor. Its weaknesses are driver unfamiliarity, greater yellow-trap exposure when the opposing direction is permissive, and left-turn queues that must wait out the whole through interval, which increases the storage length the bay must provide.

(d) Protected phase versus permissive phase

This pair concerns whether a movement is given the right-of-way exclusively or must find it, and it is the single most consequential choice in left-turn signal design because it changes the movement's saturation flow by roughly an order of magnitude.

Under a protected phase the movement proceeds on its own indication with all conflicting movements stopped. A protected left turn therefore discharges as a saturated queue at close to the through-movement saturation flow — on the order of 1600 to 1800 vehicles per hour of green per lane — because no gap search is involved. Delay is predictable, the capacity is a deterministic function of the green time, and the crash types associated with gap misjudgement are eliminated. The cost is paid in cycle length: each additional protected phase adds its own lost time and, at an intersection with pedestrians, roughly a further pedestrian clearance interval, so the cycle grows and every other movement's delay grows with it.

Under a permissive (permitted) phase the movement shares its green with a conflicting movement and may proceed only through gaps in that conflicting stream — a permitted left turn shows a circular green and must yield to opposing through traffic and to pedestrians in the crosswalk it crosses. The capacity available to it is therefore not the green time but the gap availability in the opposing stream, so it falls sharply as opposing volume rises and can approach zero when the opposing through movement is itself near saturation. Permissive operation is efficient at low opposing volumes because it consumes no separate phase, and it is what makes a compact two-phase plan possible; it becomes unsafe and ineffective at high opposing volumes, and it is unsuitable where sight distance to the opposing stream is restricted or where a heavy conflicting pedestrian movement continually blocks the turn.

Intermediate arrangements exist and are common in Canada: protected-permissive operation displays an arrow for part of the phase and then a circular green for the remainder, buying the protected discharge of the initial queue while retaining the sneaker capacity of the permitted interval. Question 2 of this paper is a case where the choice is forced by arithmetic rather than preference — the protected option turns out to be infeasible.

(e) Time mean speed versus space mean speed

Both are averages of vehicle speeds in a traffic stream, but they average different populations and therefore answer different questions. Time mean speed is the arithmetic mean of the spot speeds of the vehicles passing a point during a stated interval,

and it is what a radar gun, a laser or a single inductive loop measures. Because the sampling is at a point in time-flow, fast vehicles pass the point more frequently than slow ones and are therefore over-represented in the sample.

Space mean speed is the average speed of the vehicles occupying a length of roadway at an instant, and is computed as the harmonic mean of the individual speeds, or equivalently as the length divided by the mean travel time over that length,

It is the average that travel-time runs, floating-car studies and probe-vehicle data produce, and it is the speed that appears in the fundamental relation of traffic flow, so it — and never the time mean speed — is the value to use with

The two are related by

where $\sigma_s^{2}$ is the variance of the space-mean-speed distribution. Because a variance cannot be negative, the time mean speed always equals or exceeds the space mean speed, and the gap widens as the speed distribution spreads — it is negligible on a free-flowing facility with disciplined speeds and can be several kilometres per hour in congested or mixed-traffic conditions. Substituting a measured time mean speed into $q=ku_s$ therefore understates density, which is the practical reason the distinction matters.