Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 47
36 I N N O VAT I O N S I N T R AV E L D E M A N D M O D E L I N G , V O L U M E 2
The only nontrivial transformation that requires I I
explanation is the aggregation of non-transit- )
exp (Vi = exp (C ) × P (i ) = exp (C )
exponentiated utility, for which the following notation is i =1 i =1
Wn
introduced: N I W (5)
= exp (Vin )
i = 1, 2, . . . , I = choice alternatives; n =1 i =1
n = 1, 2, . . . , N = individual records to be aggre-
gated within the group;
Vin = known individual utilities; By combining Equations 2 and 5, it is possible to obtain
Pn(i) = known individual probabilities; the expression for aggregate exponentiated utilities that
Wn = person trips (1, for individual tours; party size, would exactly reproduce the target market shares and
for joint tours); user benefits:
Vi = unknown aggregate utilities; and
W = total person trips for the group.
)
exp (Vi = P ( i ) × exp (C )
Wn
The purpose of utility aggregation is to find a utility I
N W
expression that will exactly replicate (a) aggregate mode = P (i ) × exp (Vin ) (6)
shares and (b) total composite utility across all individ- n =1 i =1
ual records and consequently replicate a UB calculation.
The aggregate mode shares for the group of records can
be readily calculated as follows: Equation 6 has a simple intuitive interpretation. The
aggregate exponentiated utility of each mode is propor-
N
tional to a product of two factors. The first one is equal
W n × Pn ( i ) (1) to the aggregate mode share and reflects the improve-
P (i ) = n =1
ment of each mode in comparison with the other (com-
W
peting) modes. The second one is equal to the weighted
The first condition (replication of aggregate shares) geometric average of the individual (exponentiated) log
leads to the following expression: sums; this component is sensitive to the overall improve-
ment of all modes.
Vi = ln P ( i ) + C
(2) This aggregation calculation should be implemented
separately for the base scenario and the build scenario
where C denotes a utility scale constant that has to be and for each group of aggregated records. In the same
determined. way that this aggregation can be applied for mode utili-
The second condition (replication of the composite ties, it can be applied for the composite nontransit utility
utility) leads to the following expression (for simplicity it as well as it can be generalized for any nested structure
is assumed that the choice model is a simple multinomial by using a full mode choice log sum for each record LSn.
logit model and the composite utility is calculated as a The aggregate exponentiated nontransit utility can be
simple one-level log sum): calculated as follows:
I N I P ( SOV ) + P ( HOV )
W × ln exp (Vi
) = Wn × ln exp (Vin ) (3) exp (VNT = )
i =1 n =1 i =1 + P ( NM ) + P ( SB)
N Wn
(7)
Equation 4 results from an equivalent transformation × n =1
exp ( LSn ) W
of Equation 3:
I N
Wn I where
)
exp (Vi = exp × ln exp (Vin ) SOV = single-occupancy vehicle,
i =1 n =1 W
i =1
Wn HOV = high-occupancy vechicle,
(4)
N I W NM = nonmotorized vehicle, and
= exp (Vin ) SB = school bus.
n =1 i =1
The described methodology of UB calculation has
By substituting the expression for aggregate utilities from been applied for the analysis of the recent LRT project in
Equation 2 to Equation 4, the necessary formula for the Columbus, Ohio, and approved by FTA. The results are
utility scale C is obtained: provided in a companion paper.
