Summary Mobilities, Travel and Networks
Lecture 1
Chapter 3: Random Utility Theory – Cascetta, 2009
Introduction
Transport flows result from the aggregation of individual trips. Each trip is the result of a number of
choices made by transport system users (by travellers in the case of personal transport or by
operators (manufacturers, shippers, carriers) in goods transport). Some choices are made
infrequently (e.g., where to reside and work and whether to own a vehicle), while others are made
for each trip (whether to make the trip, at what time, to what destination, which travel mode etc).
Each choice context, defined by the available alternatives, evaluation factors, and decision
procedures, is known as a choice dimension. In most cases, travel choices are made among a finite
number of discrete alternatives.
Random utility theory is the most widely used theoretical paradigm for modelling transport-related
choices, and choices among discrete alternatives random utility models (behavioural model).
Basic assumptions
Random utility theory is based on the hypothesis that every individual is a rational decision-maker,
maximizing utility relative to his or her choices. The theory is based on the following assumptions:
- The generic decision-maker, in making a choice, considers mutually exclusive alternatives that
constitute her choice set. The choice set may differ according to the decision-maker (e.g., in the
choice of transport mode, the choice set of an individual without a driver’s license or car obviously
should not include the alternative “car as a driver”).
- The decision-maker assigns to each alternative in his choice set a perceived utility or
“attractiveness” and selects the alternative that maximizes this utility.
- The utility assigned to each choice alternative depends on a number of measurable characteristics,
or attributes, of the alternative itself and of the decisionmaker.
- Because of various factors described later, the utility assigned by the decisionmaker to an
alternative is not known with certainty by an external observer (analyst) wishing to model the
decision-maker’s choice behaviour.
From the above assumptions, it is not usually possible to predict with certainty the alternative
that the generic decision-maker will select. However, it is possible to express the probability that the
decision-maker will select alternative j: this is the probability that the perceived utility of alternative j
is greater than that of all the other available alternatives.
The set of available alternatives (choice set) significantly influences the choice probabilities. If a
particular decision-maker’s choice set is known, the definition of choice probability can be applied
directly. However, oftentimes, the analyst has no exact knowledge of the generic decision-maker’s
choice set.
Systematic utility: the mean perceived utility perceived by all decision-makers having the same
choice context (alternatives and attributes) as decision-maker. The attributes can be classified in
different ways;
- level-of-service or performance attributes: those related to the service offered by the transport
system (times, costs, service frequency, comfort, etc.).
- activity system attributes: those related to the land-use characteristics of the study area (e.g., the
number of shops or schools in each zone).
,- socioeconomic attributes: those related to the decisionmaker or to his household (income, holding
a driver’s license, number of cars in the household, etc.).
Utilities are merely a way of capturing the preference ordering among alternatives, and so have no
intrinsic units of measurement; alternatively, they can be expressed in arbitrary dimensionless units
(utils).
The randomness of perceived utilities: Various factors account for the difference between the utility
perceived by an individual decision-maker and the systematic utility common to all decision-makers
with equal values of the attributes. They include:
- Errors in measuring the attributes that are included in the systematic utility.
- Attributes that are not included in the systematic utility because they are not directly observable or
are difficult to evaluate (e.g., travel comfort or total travel time reliability).
- Instrumental attributes that are included in the systematic utility specification but only imperfectly
represent the actual attributes that influence the alternatives’ perceived utility (e.g., modal
preference attributes replacing variables such as the comfort, privacy, image, etc. of a certain
transport mode; the total number of commercial establishments in a given zone replacing the
number and variety of shops).
- Variability among decision-makers, or variations in tastes and preferences among different decision-
makers and, for an individual decision-maker, over time. Different decision-makers with otherwise
identical attributes might have different utility functions or different values of the reciprocal
substitution coefficients according to personal preferences (e.g., walking distance is more or less
disagreeable to different people).
- Errors in the evaluation of attributes by the decision-maker (e.g., erroneous estimation of travel
time).
Rewarding rush-hour avoidance: A study of commuters’
travel behavior – Ben-Elia & Ettema, 2011
Introduction
Congestion on urban roads throughout the EU is increasing and is expected to worsen as the demand
for travel increases and supply of road infrastructure remains limited. It has external costs such as
pollution, noise and road user safety, interrupted vehicle flow and uncertain travel times.
Transportation economists argue for implementing road pricing as a first-best solution to efficiently
alleviate congestion externalities. However, this is controversial and insight is lacking in key domains
which could lead to different outcomes than those predicted by economic theory. First, optimal
pricing requires that tolls are designed to be variable making them quite complex for drivers
comprehension. Second, it raises questions regarding social equity and public acceptability
(perceptions of fairness play a key role in public acceptability of pricing schemes). Third, situational
constraints such as household obligations (e.g. childcare), work organization and availability of
information may also affect individuals’ responses to pricing schemes and limit their effectiveness.
Second-best schemes have been suggested to circumvent the difficulties in implementing first-best
solutions it has been suggested that an incentive for avoiding peak-hour travel can achieve a
similar behavioral response to that of pricing. By rewarding those commuters who are willing to shift
their commuting times or switch to alternative travel modes, overall penalization of drivers through
tolling is avoided and overall welfare could well be improved by reducing peak demand (examples
rewards: free public transport, toll/road pricing discounts). People respond more favorably and are
more motivated when rewarded rather than punished. It still remains questionable if rewards can
,sustain behavior changes in the long run. Studies reveal that commuters’ habitual behavior of rush-
hour driving and constraints related to household and work schedules are important factors which
limit the positive impact of rewarding on car travel. Moreover, following Prospect Theory,
perceptions of loss aversion suggest an asymmetry in the valuation of the reward as a prospective
gain and the losses incurred by the time disruptions in changing one’s habitual schedule
importance of time; asymmetrical values of time which are higher when perceiving travel time losses.
Thus, if a reward is to change commuter behavior, the time losses should be perceived as relatively
small.
Netherlands: using rewards to change commuters’ behavior due to the Spitsmijden program (peak
avoidance). Initial results provided evidence of substantial behavior changes in response to the
rewards, with commuters shifting to earlier and later off-peak times and more use of public transport
as well as working from home. However, household and work constraints, flexible work
arrangements and support measures such as travel information could well influence the response
Although the reward is the main motivation in potentially choosing to participate in a reward-based
scheme, lack of flexibility in daily schedules was the main reason to reject it.
Results
Social support was stated by all participants as facilitating behavior change. This included mainly
arrangements with the employer, and with family members.
Regarding mediating factors, gender has a significant effect on rush hour driving, suggesting men
tend to change behavior more often than women. Additionally, factors relating to usual or habitual
behavior have significant results. Participants associated with classes A and B (2.5–5 rush-hour trips
at pre-test) were more likely to continue driving during the rush-hour compared to classes C and D
(0–2.5 trips). The use of other modes for commuting has a positive effect on not driving. This shows
that positive experience with travel by other modes also may encourage their use to gain a reward.
Scheduling constraints (e.g., child chauffeuring) is effective in discouraging a change of behavior.
Support measures have a positive effect on behavior change. Participants who stated they discussed
flexible working times with their employers were less likely to drive during the rush-hour.
Participants with ability to start working later are more likely also to drive later.
Participants in the money group who reported practicing with avoidance behavior during the pre-test
were more likely to avoid the rush-hour. Participants in the money group, who reported in retrospect
a greater difficulty (in terms of effort) in changing behavior, were also less likely to avoid the rush-
hour this indicates that positive or negative perceptions regarding experiences can have an
influence on the likelihood to change behavior. The attitudes towards driving alternatives iis also
important in influencing change of behavior (e.g., participants with a positive attitude to cycling were
more likely to change behavior by not driving).
Participants with frequenter use of traffic information were more likely to drive later. Participants
with frequenter use of public transport information and/or stating they had searched for public
transport alternatives to support their behavior change were more likely to change behavior by not
driving. Even without an extrinsic reward, the Yeti has instrumental value by allowing easier access to
travel information which in turn encourages also the use of nondriving alternatives to commute.
Discussion
In terms of the factors influencing commuter travel behavior, the results indicate that the reward is
the primary factor affecting their choices and the likely trigger that stimulates commuters to consider
changing their behavior. However, this effect is mediated by other factors. These include socio-
, demographic characteristics, situational factors (home and work related), habitual behavior and
experience, attitudes, travel information and even weather.
The rewards
Both the monetary reward and the Smartphone reward are effective, at least in the short run, in
reducing rush-hour car commutes. This is the most prominent factor influencing behavior. The 3€
level already has the largest influence on behavior change while the 7€ level and the mixed level
have relatively only a marginal effect. Thus in terms of cost-effectiveness it seems that most of the
benefits from a change in commuters’ behavior can be accomplished with a low level or similar. The
Yeti credits seem to be equally effective to the 3€ level.
Yeti users were more likely to choose to drive later compared to participants in the money group
Yeti users had 24 h access to travel information. This leads us to suggest that the change of behavior
is also influenced by travel information availability.
Despite the effectiveness of the rewards, it is difficult to conclude from a relatively short longitudinal
study about the impacts of rewards in the long run. Motivation theories suggest that if intrinsic
motivation kicks in, the change of behavior is more likely to be sustained. Post-test: once rewards
ceased that avoidance shares dropped.
Socio-demographics
Gender and higher education have significant effects on commuters’ behavior. Men are more likely
to avoid the rush-hour compared to women. Women’s lower motivation to avoid the rush-hour,
despite the possibility of gaining a reward, can be associated with many issues (e.g., women have
more time constraints compared to men for various reasons, mainly household tasks and child raising
obligations; leaving work early in the afternoon to pick up children from nurseries, limits ability to
start work later).
Education is a known proxy for latent income effects. Income is regarded as a key issue determining
willingness to pay for travel purposes as well as the value of travel time savings. Participants with
higher real income are likely to be less sensitive to a marginal monetary gain compared to
participants with lower incomes.
Situational factors
Scheduling constraints such as household obligations (e.g. childcare, children chauffeuring) and work
organization have been found by others to influence individuals’ responses to pricing schemes and
limit their perceived effectiveness. Conversely, factors related to flexible working times appear to be
important in encouraging behavior change (e.g., participants that could start working later were
more likely to drive later) flexibility, especially at the work place, is a key issue in promoting
changes in travel behavior. Home-related support measures such as household arrangements did not
have any significant effect on behavior-change. Also relates to loss aversion in travel time:
participants who perceived the time disruption losses as minor (because of flexible schedules)
compared to the gained reward, were more likely to change their behavior in response to the
reward.
Habitual behavior and experience
In the long run habitual travel behavior, is quite relevant for promoting or discouraging a behavior
change different from one’s usual travel behavior. Habitual behavior is less intentional more
automated and script based.
The effect of habitual behaviour is evident. Specifically, participants with higher rush-hour commute
frequencies during the pre-test (reward classes A and B) were relatively less likely to avoid the rush-
hour compared to participants with lower rush-hour frequencies (class C, D). e. In addition, real-time
