16-Civ-B7 Transportation Planning and Engineering · May 2018
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Paper format. National Examination, May 2018 — 16-Civ-B7, Transportation Planning and Engineering. Three hours, closed book (one two-sided aid sheet permitted; Casio or Sharp approved calculator). Seven questions of 20 marks each; any five constitute a complete examination, and only the first five as they appear in the answer book are marked. The per-sub-question mark split is printed on page 7 and is reproduced beside each part below. All seven questions are solved here, because the complete set is the more useful study resource.
Reference texts for this subject.
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.
Trip generation is estimated separately for the two ends of a trip because the two ends are driven by different things. Production is a property of the households that live in a zone: it answers “how many trips do the residents of this zone make?” Attraction is a property of the activities located in a zone: it answers “how many trips end here?” A zone in a Canadian regional model is therefore described by two quite different data sets, and the split matters because the four-step model must later balance the two.
The production variables that consistently carry explanatory power are household size, household income, vehicle availability and the number of employed residents, together with the life-cycle stage of the household and the residential density of the zone. Household size raises production almost linearly, because each additional person carries an independent set of work, school, shopping and recreation obligations. Income raises it because travel is a normal good: a higher-income household consumes more discretionary out-of-home activity and can afford the vehicle that makes it convenient. Vehicle availability raises it for the same reason and is usually the single strongest predictor, since it converts a latent desire to travel into an actual trip at low marginal cost. Employed residents fix the number of obligatory work trips, which are the peak-hour trips that matter most for capacity design. Residential density and accessibility work in the opposite direction on a per-capita basis: dense, mixed zones let one walking trip satisfy several purposes, and trip chaining suppresses the count of vehicle trips even where the count of activities is unchanged.
The attraction variables are measures of the opportunities a zone offers. Total employment, disaggregated by type, is the primary variable, because retail, office and industrial jobs generate very different numbers of visitor trips per employee; retail floor area, school and post-secondary enrolment, hospital beds and the supply of parking are the usual supplements. A zone with a large retail floor area attracts a large number of short, off-peak, discretionary trips, while a zone dominated by office employment attracts a sharply peaked flow of work trips in the morning. Parking supply is a genuine attraction variable and not merely a consequence of demand: an office zone that supplies no parking will attract the same number of person trips but far fewer vehicle trips.
The reason these variables behave as they do is that trip generation is the demand side of the transport market. Production variables measure a household’s resources (income, vehicles) and its needs (people, workers); attraction variables measure the supply of destinations that can satisfy those needs. Neither is exactly linear — production per household saturates above roughly four or five persons, and vehicle availability saturates above two vehicles — which is exactly why cross-classification models cap the explanatory variables, as Question 3 of this paper does.
Average vehicle occupancy (AVO) is person trips divided by vehicle trips. Raising it attacks congestion at the vehicle end without suppressing the person trip, which is why it is politically easier than trip suppression. Because vehicle trips vary inversely with AVO, raising commuter AVO from a typical Canadian value of 1.15 to 1.30 removes \(1 - 1.15/1.30 = 11.5\) per cent of peak vehicle trips — a reduction comparable to adding a lane, at a small fraction of the cost.
The measures that actually move AVO fall into four families. Incentives to share the vehicle: employer-based carpool and vanpool matching, guaranteed-ride-home programs that remove the main behavioural objection to ridesharing, and dynamic ride-matching applications that make the matching cost negligible. Priority for high-occupancy vehicles: HOV lanes and HOT lanes with a 2+ or 3+ occupancy threshold, queue-jump lanes at ramps and signals, and priority access to bridges and tunnels — these convert occupancy into a travel-time saving the traveller can feel every day. Pricing: priced or cash-out parking (paying the employee the cash value of the parking stall they do not use), congestion or corridor pricing with an HOV exemption, and employer-subsidised transit passes such as the U-Pass arrangements used across Canadian universities. Complementary measures: compressed work weeks, staggered start times and telework, which strictly speaking reduce or reschedule trips rather than raise occupancy, and should be described as such if full marks are wanted.
The effect on travel patterns is a mode shift out of the drive-alone vehicle into carpools, vanpools and transit, together with a measurable spreading of the peak where staggered hours or pricing are used; origin–destination patterns themselves change little, because the underlying activity locations are fixed in the short run. The effect on travel time is twofold: HOV users gain directly from the priority facility, and all users gain indirectly because removing vehicles from an over-saturated approach relieves the queue non-linearly — near capacity, a ten per cent volume reduction can cut delay by far more than ten per cent, as the queueing analysis in Question 2 shows. The effect on fuel consumption is the largest of the three: vehicle-kilometres fall roughly in proportion to the fall in vehicle trips, and the remaining vehicles operate at higher, steadier speeds instead of in stop-and-go conditions where fuel use per kilometre is at its worst. Both effects push the same way, so a fuel and greenhouse-gas saving somewhat larger than the vehicle-trip saving is the normal outcome.
Low-density suburban development pushes work trips toward the automobile and makes them longer, and both effects follow from the same cause: density determines what the other modes can offer.
On mode choice, transit service is supply-constrained by density. Conventional bus service needs roughly 30 dwelling units per hectare in the walking catchment to support a useful frequency; typical single-detached suburban development runs well below that, so the service that can be justified is infrequent, indirect and uncompetitive. The curvilinear street patterns and cul-de-sacs that characterise these subdivisions raise route circuity for both the bus and the pedestrian, so the effective walking distance to a stop is much greater than the straight-line distance. Land uses are separated by zoning, so no work destination is within walking or cycling range. Parking is free and abundant at both ends. Household vehicle availability is consequently high, which in the utility framework of Question 7 means a large positive constant on the automobile alternative and a large out-of-vehicle time penalty on transit. The predictable result is a drive-alone share for work trips well above the regional average, a transit share that is non-trivial only for trips to a strong central business district served by express or park-and-ride service, and a walk and cycle share for work trips near zero.
On travel distance, low-density development consumes land, so the same population occupies a larger area and the average separation between any home and any job grows. Suburban growth also tends to outrun employment growth locally, producing a jobs–housing imbalance that forces long radial commutes to the central business district or to a suburban employment centre; where jobs have themselves dispersed, the commute becomes a long suburb-to-suburb trip that no transit line can serve. Because the gravity model of Question 5 is driven by a deterrence function of distance, a settlement pattern that raises inter-zonal distances directly raises the modelled — and the observed — average work-trip length. The measured consequence in Canadian census journey-to-work data is a higher vehicle-kilometres-travelled per worker, higher fuel use and higher emissions, which is precisely the outcome that regional growth strategies and transit-oriented development policies are written to reverse.
