Question 1 of 7: Land Development, Trip Production and Connected-Vehicle Traffic Management

Question 1: Land Development, Trip Production and Connected-Vehicle Traffic Management (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) Commercial land development and accessibility (6 marks)

Accessibility is the ease with which activity opportunities can be reached from a given location, and it is the single variable that couples the transportation system to the land-use system. It is usually written as a weighted sum of the opportunities available at every other zone, discounted by the cost of reaching them, $A_i = \sum_j O_j\, f(c_{ij})$, where $O_j$ is a measure of opportunity in zone $j$ (retail floor space, employment, population) and $f(c_{ij})$ is a decreasing function of the generalised travel cost between $i$ and $j$. Because $f$ falls steeply with cost, a site's accessibility is dominated by what lies within a short travel time of it, and improving the transportation service that serves the site raises $A_i$ without any change in the surrounding land use at all.

Commercial land uses are the most accessibility-sensitive of all land uses because their revenue is generated by people who must physically travel to them. A retailer's turnover depends on the size of its trade area, which is the set of households that can reach the store within an acceptable travel time; an office's ability to recruit depends on the size of the labour shed within a tolerable commute. Raising accessibility therefore enlarges the market that a given parcel commands, and competition among firms for that enlarged market is bid into the rent the parcel can charge. This is the classical bid-rent argument: commercial bid-rent curves are the steepest of all land uses, so commercial activity outbids residential and industrial uses precisely at the points of highest accessibility — historically the central business district, and in the contemporary Canadian metropolis also the freeway interchange, the regional transit station and the arterial intersection.

The association runs in both directions, which is why it is described as the land-use/transport feedback cycle. Improved accessibility raises land value, which attracts more intensive commercial development; that development generates additional trips; those trips load the network and, unless capacity or transit service grows with them, they raise travel costs and erode the very accessibility that attracted the development. The cycle is self-limiting in congested corridors and self-reinforcing where capacity keeps pace. In Metro Vancouver the mechanism is explicit policy rather than an accident: the Regional Growth Strategy and TransLink's regional transportation strategy concentrate commercial floor space in designated Urban Centres and Frequent Transit Development Areas, and the office and retail intensification at Metrotown, Brentwood and Lougheed followed the SkyTrain investment rather than preceding it.

Two practical qualifications matter to a transportation planner. First, accessibility is mode-specific and directional: a big-box retailer needs automobile accessibility plus parking supply, whereas a downtown office tower needs transit accessibility during a two-hour peak, so the same site can be highly accessible for one commercial use and poorly accessible for another. Second, the relationship is the reason a transportation study cannot take land use as fixed. A capacity improvement that is evaluated against a frozen land-use forecast will understate the traffic it eventually carries, because the improvement itself induces the commercial development that generates the new trips. Integrated land-use/transport models exist for exactly this reason.

(b) Factors that increase trip production at three levels (7 marks)

Trip production is the number of trips whose home end lies in the unit of analysis. The three levels below are nested — a zone is an aggregation of households, and a household is an aggregation of persons — so the factor that is meaningful at each level is different in kind, and that distinction is what the question is testing.

(i) Zonal level — the number of dwelling units (residential density) in the zone. Zonal production is by construction the sum of the productions of the households the zone contains, $P_z=\sum_{h\in z} p_h$, so any increase in the count of households raises the zonal total directly and almost linearly, even if no individual household changes its behaviour at all. This is why residential land-use intensification is the strongest single lever on zonal production, and why the trip-generation step of the four-step model takes the forecast dwelling-unit inventory as its primary input. A second, weaker zonal effect operates through accessibility: a zone that is well connected to employment and retail sees a modest increase in the trip rate per household, because more opportunities fall inside the acceptable travel-cost window. The two effects can oppose each other, since dense accessible zones often have smaller households, which is exactly why the model disaggregates to the household level rather than fitting a rate to zonal population alone.

(ii) Household level — household income (and the automobile availability it buys). Income raises production for two linked reasons. It relaxes the money budget within which the household chooses its activities, so discretionary shopping, personal-business and social/recreational trips — the trip purposes with the highest elasticity — become affordable. And it raises automobile ownership, which lowers the generalised cost and, more importantly, the effort of an additional trip, so trips that would previously have been chained onto another trip or foregone are made separately. The effect saturates: once a household has one vehicle per licensed driver, further income buys better vehicles rather than more trips, so the income response flattens at the top of the distribution. Household size and life-cycle stage act in the same direction, which is why the cross-classification tables used in Question 3 are indexed by persons per household and vehicles per household.

(iii) Person level — employment status, specifically being a full-time worker who holds a driver's licence. A worker makes two mandatory home-based work trips on a typical weekday before any discretionary travel is considered, and the fixed anchor of the work trip generates further chained stops on the return leg for child care, shopping and personal business. A non-worker — a pre-school child, a student served by a school bus, a person out of the labour force, a retiree — has no such mandatory anchor and produces materially fewer home-based trips. Licence holding matters independently of employment because an unlicensed adult in a car-owning household is a passenger rather than a driver and is far more constrained in when they can travel. This is why modern activity-based and tour-based models are built on persons rather than households: the same three-person household produces very different travel depending on whether it contains three workers or one worker and two dependants.

(c) V2V and V2I communications applied to traffic management (7 marks)

Connected-vehicle technology changes traffic management in one fundamental way before any specific application is considered: it replaces a sparse, fixed, point-based sensing network of loop detectors and cameras with a dense, mobile, vehicle-based one. Every equipped vehicle becomes a probe reporting its own position, speed and trajectory, so the traffic manager obtains network-wide travel times, densities and turning movements continuously rather than at a handful of instrumented cross-sections. That change moves traffic management from a reactive posture, in which a problem is inferred minutes after it forms, to a proactive one in which it is detected in seconds and, in some cases, anticipated.

Vehicle-to-infrastructure applications are principally about control and information. Signal controllers broadcasting signal phase and timing (SPaT) messages allow green-light optimal speed advisory, so an approaching driver is told the speed that will arrive on green — this suppresses the accelerate-and-brake cycle, cuts fuel use and emissions, and reduces rear-end conflicts on the approach. The same channel supports transit signal priority and emergency-vehicle pre-emption with far better position accuracy than a loop or an emitter. In the reverse direction, vehicle data feeds adaptive signal control, so split, cycle and offset respond to measured platoons rather than to a historical time-of-day plan; it improves ramp-metering and variable-speed-limit algorithms on freeways; and it supports queue-end warning, work-zone warning and road-weather advisories delivered directly into the cab, which are the three applications with the clearest collision-reduction case in Canadian winter conditions.

Vehicle-to-vehicle applications are principally about safety and capacity. Forward-collision warning, blind-spot warning, electronic emergency-brake lights and intersection-movement assist address the conflict types that dominate urban collision statistics. The capacity implication is more interesting to a traffic engineer: cooperative adaptive cruise control lets a following vehicle react to the lead vehicle's braking through a radio message rather than through the driver's perception–reaction time, which safely shortens headways. Because jam density is the reciprocal of the average space headway, shortening headways shifts the whole fundamental diagram — jam density rises, capacity rises with it, and the same pavement carries more vehicles per hour per lane. Truck platooning exploits the same mechanism for aerodynamic benefit on intercity corridors.

At the network level the two channels together enable dynamic route guidance, real-time origin–destination estimation and dynamic traffic assignment, incident detection within seconds of an event, and dynamic congestion pricing based on measured rather than assumed conditions. Route guidance is worth singling out because it bears directly on Question 7: drivers assigned by a system-aware advisory service can be steered away from the pure user-equilibrium outcome toward a system optimum, and because guidance removes much of the perception error in travel-time estimates, it also moves route choice from a stochastic user equilibrium toward the deterministic equilibrium the classical model assumes.

Three cautions belong in any professional answer. Benefits are strongly non-linear in market penetration — cooperative adaptive cruise control delivers little until a substantial fraction of the fleet is equipped, so the mixed-fleet transition period must be designed for explicitly. Cybersecurity and privacy are first-order engineering requirements rather than afterthoughts, since a spoofed SPaT message or a falsified probe report is a safety hazard and a continuous position trace is personal information under Canadian privacy legislation. And the institutional frame must keep pace: Transport Canada's work on connected and automated vehicles, the SAE J2735/J2945 message standards and provincial pilot regulations define what a Canadian agency may actually deploy, and an engineer recommending a connected-vehicle strategy should confirm that the intended application is within that frame before committing capital.