National Academies Press: OpenBook

Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand (2012)

Chapter: Chapter 1 - Research Objectives and Main Methodology

« Previous: Executive Summary
Page 24
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 24
Page 25
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 25
Page 26
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 26
Page 27
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 27
Page 28
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 28
Page 29
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 29
Page 30
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 30
Page 31
Suggested Citation:"Chapter 1 - Research Objectives and Main Methodology." National Academies of Sciences, Engineering, and Medicine. 2012. Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand. Washington, DC: The National Academies Press. doi: 10.17226/22689.
×
Page 31

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

24 C h a p t e r 1 three Levels of Specification The research agenda for the SHRP 2 C04 project topic required attention to both theoretical and applied perspectives. Suc- cinctly stated, the primary objectives are as follows: • Theory and Research. Develop mathematical descriptions of the full range of highway users’ behavioral responses to congestion, travel time reliability, and pricing; and • Application in Practice. Provide guidance for incorporating these mathematical specifications into various travel demand models currently in use (and under development), recogniz- ing the complex nature of supply- (network-) side feedbacks (via traffic assignment techniques). The research was conceptualized in three interconnected levels of behavioral rigor and practical application, with vary- ing levels of sophistication and associated inputs: • Level 1. Behavior foundations; • Level 2. Advanced operational modeling (activity or tour based); and • Level 3. Opportunities for prevailing practice (aggregate trip based). Because supply–demand interactions are critical for con- gestion and pricing solutions (including network equilib- rium), these interactions offer a second dimension, as reflected in Figure 1.1. Level 1: Behavior Foundations The first level, as shown in Figure 1.1, corresponds to behav- ioral models intended for a deep understanding and quanti- tative exploration of travel behavior. These models include many kinds of variables, often explicitly controlled under stated-preference (SP) settings (e.g., preferred arrival time and schedule flexibility) and not all of which can be produced by most network or supply-side models (e.g., travel time reli- ability, particularly in the event of nonrecurring incidents). These models seek to address the full range of possible short- and long-term responses, but they may also focus on select choice dimensions (e.g., route and departure time choices or usual workplace location choice). Supply-side variables for such models can be based on observed or generated measures (or both) of congestion, reli- ability, and price (via, for example, an SP survey design). Mul- tiple, repeated observations can be used for direct derivation of reliability measures. Typically, there is no consideration of equilibrium at this stage, and the linkage between the demand and supply sides is essentially one-directional. Research associated with the widest possible range of behav- ioral responses is important for the construction of an “ideal” behavioral model. Such a model is free of implementation con- straints, but with some simplifying assumptions, it is able to serve as the starting point for operational models. In particular, this exploratory level considers dynamics—adjustments of within-day, as well as day-to-day, time frames: short term (in which case the effect of information could be included), medium term, and long term—as well as the correspondence of the time scale to different choice dimensions. For example, in certain situations for short-term analysis, route choice might be the only relevant dimension, but departure time choice may be equally important for day-to-day and medium- and long- term responses. Level 2: Advanced Operational Modeling The second level relates to the emerging set of relatively advanced, yet operational, activity-based models (ABMs) that are integrated with state-of-the-art dynamic traffic assignment (DTA) models for network simulation. These models allow for the incorporation of a wide range of possi- ble short- and long-term responses that are embedded within Research Objectives and Main Methodology

25 the choice hierarchy. For example, a traveler’s acquisition of an E-ZPass or transponder may be linked to his or her sub- sequent choice of payment type (at the lower level of the behavioral hierarchy). The integrity of operational models requires that each choice dimension should be allocated a proper “slot” in the hierarchy, with upward and downward linkages to related choices. Operational and computing time requirements often limit the total number of choice dimensions and alternatives, but this restriction is lessening with time. Another relevant constraint in model application is that all measures of congestion, reliability, and price must be compatible with the demand model’s specification and can be generated by the network simulation. Moreover, the demand and supply sides should be integrated in an equi- librium setting, which imposes certain limitations on how variables like travel time variability are generated, as direct methods based on multiple observations of the same trip typically are infeasible. Level 3: Opportunities for Prevailing Practice The third level relates to the larger number of existing model systems used by most metropolitan planning organizations (MPOs) and state departments of transportation (DOTs) in the form of aggregate trip-based models (frequently referred to as “four-step” models). Although rather restrictive in design, such models are prevalent and offer opportunities for meaningful and immediate contributions to the state of travel demand modeling practice. Although the four-step framework emphasizes short-term responses to highway congestion and pricing policies (including changes in route, mode, and in some cases, departure time, for each trip seg- ment), it allows for the incorporation of some trip distribu- tion and even trip generation effects through generalized cost impedances (mode choice logsums) and accessibility measures (destination choice logsums). The four-step model framework also allows for some indirect reflection of pricing on long-term choices, including workplace location and car ownership. A serious restriction of four-step models, also common to most ABMs in current implementation, is that they rely on static assignment procedures. Static assignments generate only crude average travel time and cost variables, and reliability can be incorporated only through certain proxies. In this respect, the C04 research project has aimed to push the boundary of the network models to achieve greater behavioral sensitivity within the demand models, along with natural integration of all system components. Although several advanced models and methods exist, they require special data sets and longer run times, along with other use restrictions, many of which are purely technical. For exam- ple, DTA at a full regional scale is not yet realistic, although with computational advances and parallel processing oppor- tunities, a dramatic breakthrough may be anticipated in the next 5 to 10 years, or possibly sooner. These constraints on practical applications also relate to the current limitations of demand models in terms of possible number of choice Figure 1.1. Levels of analytic sophistication.

26 dimensions and numerical realizations in the microsimula- tion process. Incorporation of results in applied travel Models It is important to note that each level is not independent and disconnected from the others. The team aims to establish a consistent and holistic conceptual framework in which sim- plified and pragmatic models can be derived from more advanced models, rather than being reinvented (which is probably the current state of the relationship between travel modeling theory and practice). This way, it is believed that the current project could be successful in a very important respect: by bridging the gaps between theory and practice to the great- est extent possible. The major framework for the discussion of the proposed models primarily considers the full regional model frame- work, although facility- and corridor-level models are also considered. This focus also has a consequence for the analy- sis of the existing data sets and their possible use in the current research. It is based on the recognition that for a deep understanding and proper modeling of congestion and pricing impacts, a full framework, with chosen and nonchosen alternatives, should be available to both users and nonusers, for which a full regional travel data set and model is needed. It is essential to know, both at the model estimation stage and the application stage, level-of-service (LOS) variables such as travel time, cost, and reliability for nonchoices routes, modes, time-of-day (TOD) periods, and destinations. This holistic framework is generally missing in simplified models and surveys, which limits their utility for this research. According to the adopted levels of sophistication, the research results of this C04 topic are grounded in one or more of three applied modeling contexts: • Aggregate (Four-Step) Models and Static Assignment Tools. In general, these models offer a limited framework for the incorporation of congestion and pricing effects. In particu- lar, it is problematic to incorporate travel time reliability measures in these frameworks. However, the team has for- mulated several simplified approaches that can be imple- mented within these models as they are still in use by many MPOs and DOTs. For example, the perceived highway time concept can be readily incorporated on both the demand and network simulation sides. • SP-Based Models. Most advanced behavioral models are primarily intended to provide insights into individual travel behavior. These models (especially in an SP setting) can include additional behavioral variables related to the dynamics of decision making (e.g., previously used route or option), referencing mechanism (e.g., travel time sav- ings versus the actual trip time), flexibility of the work schedules, and preferred arrival time. Usually, however, these network or supply-side variables are not easily simu- lated in model application; thus, the corresponding model cannot be directly used in the framework of an applied regional travel model. • ABMs. These advanced applied travel models and network simulation tools are characterized by a fully disaggregate structure and rely on individual microsimulation. They take full advantage of a detailed level of segmentation by household and personal characteristics and can include complicated decision-making chains and behavioral mech- anisms. They are, however, limited to only forecastable input variables. State of the art and practice in Modeling Congestion and pricing Addressing Impacts of Congestion Many of the modeling aspects of concern for this research are generally associated with travel behavior and would be rele- vant for any travel model improvement. In any travel model, there are certain time–cost trade-offs that are not necessarily related to highway pricing and congestion. And practically, any travel model would benefit from a fuller set of behavioral responses and associated choices. The C04 research, how- ever, is specially designed to substantially extend our under- standing and ability to model the impacts of transportation pricing and congestion. For this reason and due to this focus, the research was not intended to result in a general travel model improvement guide, although all recommen- dations and developed approaches were arrayed as to their relevance for the general state of the art and the practice of the profession. The most interesting and unique aspect of the research is the focus on impacts of congestion and associated pricing on traveler response and transportation system perfor- mance. Almost all existing models are already sensitive to congestion through (average) travel time variables, at least in their assignment and mode choice components. So, what is special in congestion that requires a special consideration beyond the framework of conventional model structures and approaches? Travel time reliability is one of most important aspects investigated in this research that has been generally recognized as a critical missing component in the previous generation of

27 travel demand models. Congestion is associated not only with longer average travel times, but also with higher levels of unreliability (unpredictability) of travel times, which is what makes it so onerous to highway users. A great deal of the proj- ect effort was related to the measurement and incorporation of reliability in model structures. But important as it is, reli- ability is not the only additional issue or variable that needs to be added to the existing travel models in order to have a better accounting for congestion. The team believes that a deeper understanding of the effects of congestion on travel behavior should include several addi- tional considerations: • Perceived Highway Travel Time. The practice of using dif- ferential weights for different travel time components has been universally accepted for transit modeling. Transit in- vehicle time, walk time, and wait time are perceived by the riders differently. The corresponding estimated utility coefficients normally range between 1.0 and 4.0, with the highest weights associated with waiting time under uncer- tain conditions. But there has been no parallel effort to estimate perceived highway time as a function of highway LOS, which has always been assumed to be a totally generic variable in both route choice and mode choice contexts, as well as in any subsequent use of mode choice logsums or generalized cost in trip distribution and upper-level mod- els. However, a behavioral analogue between an uncertain waiting time for an unreliable transit service and for being stuck in a car in a traffic jam is appealing. The team believes that the idea of a perceived time structure (e.g., by travel speed categories) might be beneficial from both a theo- retical and a practical modeling perspective. The introduc- tion of weighted highway time would improve mode choice models and allow for the elimination of some mys- terious constants (such as “rail bias for trips to the central business district”). • Different Pattern of Highway User Behavior in Presence of Unpredictable Travel Times. Another assumption underly- ing conventional modeling approaches that becomes unreal- istic under congestion conditions is that travelers (specifically highway users) possess full information about all possible routes and modes and make rational decisions. In behavioral terms, congestion and the associated unpredictability of travel times lead to travelers making many irrational deci- sions based on intuition and past experience that might not be relevant for the current situation. In practical modeling terms, it might be expected that the associated choice models would have relatively smaller coefficients for travel time and cost (more random behavior and regardless of value of time [VOT]) compared with models estimated for uncongested areas where travel time is predictable. As a result, in a route choice framework large deviations from the shortest path might be expected. This general pattern might be affected by the travel information system, and more so as congestion cre- ates demand for real-time information. The impact of intel- ligent transportation systems in this context represents an interesting research challenge (though not in the current project’s scope). • Disequilibrium (Lagged Equilibrium) Between Travel Demand and Supply. Another interesting and less- investigated consideration relates to the equilibrium for- mulation. It is generally recognized that travel models should reach a perfect (simultaneous) equilibrium between the demand and supply sides. A corresponding theory and effective algorithms have been well established for aggre- gate four-step models. Although it is more empirical with microsimulation ABMs, the intention, however, is still to reach a perfect equilibrium. It is interesting to note that integrated land use and transportation models have never used a concept of static equilibrium because the land use and transportation responses belong to different time scales. Most integrated land use and transportation models incor- porate the concept of lagged equilibrium. In reality, there are numerous and very different time scales within a travel demand model itself. In the presence of congestion that makes travel time unstable, the process of traveler learn- ing and adaptation associated with reaching equilibrium becomes longer and fuzzier. Research has shown that it might be beneficial to revise the formulation of transporta- tion system equilibrium accordingly. Modeling Toll Roads and Managed Lanes In the same way that the experience of congestion is not only about longer travel times, priced highway facilities themselves are more than just roads with better travel times available at additional cost. Important qualities of a tolled facility, as well as the traveler’s perception of them, include many other aspects that make priced highway facilities qualitatively dif- ferent from free highways, to such an extent that they can be better modeled as a different mode, rather than merely a dif- ferent route in the network (Spear 2005; Erhardt et al. 2003). Since the difference between priced and free highway facilities from the traveler’s perspective is probably not as great a differ- ence as between highway and transit modes, in the mode choice structure, the choice between toll and nontoll routes (preroute choice) is normally placed in the lowest level of nested struc- ture. However, it is important to keep the toll–nontoll choice as a discrete choice in order to allow for the inclusion of various utility components and biases. The assignment framework is more limited in this respect.

28 The following specific preroute factors (beyond travel time and cost) should be considered and estimated for all types of priced facilities: • Reliability. Toll roads and managed lanes, especially those with variable and real-time pricing, are perceived as reliable modes of transportation; congested free roads are inher- ently unpredictable (Bates et al. 2001; Brownstone and Small 2005; Small et al. 2005); • Safety. Some drivers may perceive priced facilities like mainline toll (high-occupancy toll [HOT]) lanes to be safer than the mainline because of the separation from the other lanes of travel, and specifically from trucks (Brownstone et al. 2003); and • Carpooling Opportunity. Several additional factors come into play if pricing or traffic restrictions are differentiated by car occupancy. Different forms of high-occupancy vehi- cle (HOV)–HOT lanes have been recently applied in many states. Valuable experience with these lanes has already been accumulated, and a significant body of model estimation work and research has been published. The team plans to cooperate closely with the ongoing NCHRP Project 8-36B, Task 52, Changes in Travel Behavior/Demand Associated with Managed Lanes. highway Utility Forms in Different Demand Choice Frameworks Highway Utility Components Highway travel utility is the basic expression of combining various LOS attributes and costs as perceived by the highway user. It is directly used in the highway trip route choice; for example, it is used between the managed lanes and general- purpose lanes on the same facility. It also constitutes an essen- tial component in mode and TOD choice utilities. The form of highway utility function is also important for modeling other (upper-level) travel choices because it serves as the basis for accessibility measures. Thus, it is essential to explore the formulation of highway travel utility and its components first, having in mind a simplified framework of route choice in the highway network, because the complexity builds when addi- tional choice dimensions are considered. In most travel demand models, including those developed for practical and research purposes, the highway utility (U) takes the following simple form: (1.1)U a T b C= × + × where T = travel time; C = travel cost; a < 0 = coefficient for travel time; b < 0 = coefficient for travel cost; and a/b = VOT. Coefficients for travel time and cost normally take negative values, reflecting the fact that travel in itself is an onerous function necessary only for visiting the activity location. Thus, the travel utility is frequently referred to as the “disutil- ity” of travel. Some research has questioned the negative character of travel utility in some contexts. In particular, a positive travel utility was associated with long recreational trips on weekends (Stefan et al. 2007). Also, an interesting effect was observed for commuting trips for which com- muters seem to prefer a certain minimum time and are not interested in reducing it below this threshold (Redmond and Mokhtarian 2001). The standard representation of highway travel utility as a linear function of two variables with constant coefficients is extremely simplified. A great deal of the present research effort has been devoted to the elaboration of this basic form in the following ways: • Investigation of the highway user perception of travel time by congestion levels. This means that a simple generic coef- ficient for travel time could be replaced with coefficients differentiated by congestion levels; • Inclusion and estimation of additional components, of which travel time reliability has been identified as the most important one. With respect to average travel time and cost, reliability is seen as an additional and nonduplicating term; and • Testing more complicated functional forms that are non- linear in time and cost, as well as account for randomly distributed coefficients or VOT (in addition to any explicit segmentation accounting for the observed user heteroge- neity). With these enhancements, VOT is not assumed as a constant, but becomes a varying parameter depending on the absolute values of time and cost, as well as reliability. As a working model, the team has adopted the follow- ing general expression for the highway travel utility, and explored it component-by-component over the course of the research: (1.2) 1 3 1 3 1 5 U a T b C c Rk k k m m m m n n nk ∑ ∑∑[ ] [ ]( ) ( )= × ϕ + × φ + = == where k = 1 represents the uncongested highway travel time component; k = 2 represents the congested highway travel time component (extra delay); k = 3 represents parking search time;

29 k = 4 represents walk access or egress time (e.g., from the parking lot to the trip destination); k = 5 represents extra time associated with carpooling (i.e., picking up and drop- ping off passengers); Tk = (average) travel time by component; m = 1 represents highway toll value; m = 2 represents parking cost; m = 3 represents vehicle maintenance and operating cost; Cm = travel cost value by component; n = 1 represents disutility of time variation (first measure of reliability); n = 2 represents schedule delay cost (second measure of reliability); n = 3 represents utility of (lost) activity participation (third measure of relia - bility); Rn = reliability measures by component; ak, bm, cn = coefficients to be estimated; and jk(. . .), fm(. . .) = functions for nonlinear transforma- tion of time and cost variables. This formulation makes it more difficult to calculate VOT, although the calculation is still possible, in the same way that value of reliability (VOR) can be calculated for the first type of reliability measure (assuming that this reli- ability measure is in minutes). VOR essentially represents travelers’ willingness to pay for reduction in travel time vari- ability in the same way as VOT represents their willingness to pay for (average) travel time savings. More specifically, VOT (in the context of willingness to pay tolls for saving time in congestion conditions) can be calculated by the fol- lowing general formula: )( )) ( (= ∂ ∂ ∂ ∂ = ′ϕ ′φ, (1.3)2 1 2 1 2 2 2 1 1 1 VOT T C U T U C a T b C A similar calculation can be implemented for VOR. With nonlinear transformation functions, VOT and VOR are no longer constant values. They now depend on the absolute val- ues of time and cost variables at which the derivatives of the transformation functions are taken. The innovative components that relate to perceived high- way time, travel time reliability, and nonlinear transforma- tions are discussed in the subsequent sections. It should be noted that some components, specifically perceived travel time and the three reliability measures, might be correlated statistically (and also conceptually duplicative to some extent). Thus, it is highly improbable that the entire formula would ever be applied. Instead, it serves as a conceptual framework for which particular structures can be derived and tested statistically against each other. Perceived Highway Time Perceived transit time has been recognized and routinely used in travel models for some time. For example, in most mode choice models and transit assignment algorithms, out-of- vehicle transit time components like wait and walk are weighted compared with in-vehicle travel time. It is not unusual to apply weights in the range of 2.0 to 4.0, reflecting the fact that travel- ers’ perception of out-of-vehicle time is perceived as more onerous than in-vehicle time. Contrary to transit modeling practice, practically all travel models include only a generic highway time term; that is, the same coefficient is applied for each minute of highway time regardless of the travel conditions. However, there is some compelling statistical evidence that highway users perceive travel time differently by congestion levels. For example, it is intuitive and behaviorally appealing that highway users driving in congested conditions might per- ceive the longer travel time as an additional delay or penalty on top of the anticipated free-flow (or some expected rea- sonable) time. Thus, the research has explored a segmentation of travel time coefficients by congestion levels, expecting that the time spent in congestion conditions has a larger disutility. A larger disutility associated with congestion would have at least two behavioral interpretations: • Negative psychological perception (similar to the weight for walking to or waiting for transit service); and • Simplified operational proxy for reliability (that should be explored in combination with the explicit reliability measures). Several related research works report statistical evidence of high perceptional weights that highway users put on travel time in congested conditions (Small et al. 1999; Axhausen et al. 2007; Levinson et al. 2004; McCormick Rankin Corp. and Parsons Brinckerhoff 2008). Multiple indications in recent analyses of travel surveys suggest that the perception of the time saved by respondents in revealed preference (RP) surveys is about double the actual measured time saved (Small et al. 2005; Sullivan 2000). This might well be a mani- festation that in the RP framework travelers operate with per- ceived travel times, in which time spent traveling through congested segments is psychologically doubled. Major Focus for Improvement for Demand analysis The C04 research into opportunities for extended research and improvements in travel demand analysis with respect to pricing and congestion has focused on improving the

30 following key structural dimensions or components of demand models: • Primary Choice Dimensions. Models are grouped by pri- mary choice dimensions that relate to congestion and pric- ing (e.g., route, TOD, and mode choices). It is shown how improved specifications of models can be effectively used in models that relate to the upper-level choices in the indi- vidual travel decision-making hierarchy, including destina- tion choice, tour and trip generation, and household car ownership. Joint choice formulations for route and mode, route and TOD, and mode and TOD, as well as route mode and TOD, have also been investigated; • Specification of Highway Utilities. For each choice model, the basic specification that includes average time and cost have been investigated, first including possible linear and nonlinear specifications. The main points of improve- ments include possible nonlinear effects and trip length scaling; • Segmentation Options. For each model various segmen- tation strategies are tested, including a full segmentation of the choice model by travel purpose and partial seg- mentation of travel time by congestion levels or travel cost (by income group and occupancy), or both. Main points of improvement studied include substantiation of the concept of perceived highway time with significantly different VOT by congestion levels as a proxy for travel time reliability; • Income Effects. Special focus is on the impact of house- hold income and corresponding functional specifications of the highway utility. This includes segmentation of some coefficients by income group, using income-specific con- stants and scaling of the cost by income (as a continuous variable). Alternative approaches are compared in a sys- tematic way and recommendations for best functional forms are made; • Car Occupancy. Special focus is on the car occupancy effects. Travel forecasting models commonly assume that travel costs should be divided by vehicle occupancy, with the implicit assumption that those costs are shared among those traveling together. This hypothesis is questioned, and alter- native formulations are explored, including segmentation of cost or time coefficients (or both) by occupancy, occupancy- specific bias constants, nonlinear scaling occupancy effects, and spate analysis of interhousehold and intrahousehold carpools. Alternative approaches are compared in a system- atic way, and recommendations for best functional forms are made; and • Incorporation of Travel Time Reliability. Significant focused effort has been made to test different travel time reliability measures and incorporate them in the route, mode, and TOD choice utility expressions. The results represent cut- ting edge research and provide valuable insights into travel- ers’ decision-making process and preferences. The estimated models provide VOR estimates that along with the VOT estimates portray travelers’ willingness to pay for different types of highway improvements. Major Focus for Improvement for Network Simulations The C04 research project has addressed recent advances in traffic microsimulation tools, dynamic equilibrium algo- rithms and implementation techniques for large-scale net- work applications, richer behavioral representation in network models, and ways to generate travel time distributions and reliability measures. Salient points of the research include the following: • Need for Microsimulation. Capturing user responses to pricing and reliability is best accomplished through micro- simulation of individual traveler decisions in a network platform. These responses must be considered in a net- work setting, not at the facility level, and the time dimen- sion is essential to evaluating the impact of congestion pricing and related measures; hence, a time-dependent analysis tool is required. Microsimulation of individual traveler choices provides the most general and scalable approach to evaluate the measures of interest in this study; • More Robust DTA Required. Simulation-based DTA mod- els have gained considerable acceptance in the past few years, yet adoption in practice remains in its infancy. The current generation of available models only considers fixed, albeit time-varying, origin–destination trip patterns. Greater use and utility will result from consideration of a more complete set of travel choice dimensions and incor- poration of user attributes, including systematic and ran- dom heterogeneity of user preferences; • Improved Algorithms for Regional Scale Modeling. Such algorithms for finding equilibrium time-varying flows have been based on the relatively inefficient method of suc- cessive averages, and its implementation in a flow-based procedure did not scale particularly well for application to large metropolitan networks. New implementations of the method of successive averages and other algorithms that exploit the vehicle-based approach of simulation-based DTA have been proposed and demonstrated on large actual networks; • Traveler Heterogeneity. Incorporating heterogeneity of user preferences is an essential requirement for modeling

31 user responses to pricing in a network setting. New algo- rithms that exploit nonparametric, multicriteria shortest- path procedures allow VOT (which determines users’ choice of path and mode in response to prices) to be continuously distributed across users. Efficient implementations of these algorithms have been demonstrated for large network application as part of this study; and • Network Reliability Measures. Most simulation models do not produce reliability estimates of travel time along network links and paths. Two practical approaches were formulated and explored as part of this work to estimate variability measures of travel time in the context of net- work assignment tools. The first exploits trajectory infor- mation in micro- and mesosimulation tools; the second employs a robust relation established between the first and second moments of the travel time per unit distance. These are illustrated for application in conjunction with network evaluation tools.

Next: Chapter 2 - Review and Selection of Data Sources »
Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand Get This Book
×
 Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s second Strategic Highway Research Program (SHRP 2) Report S2-C04-RW-1: Improving Our Understanding of How Highway Congestion and Pricing Affect Travel Demand includes mathematical descriptions of the full range of highway user behavioral responses to congestion, travel time reliability, and pricing. The descriptions included in the report were achieved by mining existing data sets. The report estimates a series of nine utility equations, progressively adding variables of interest.

The report explores the effect on demand and route choice of demographic characteristics, car occupancy, value of travel time, value of travel time reliability, situational variability, and an observed toll aversion bias.

An unabridged, unedited version of Chapter 3: Demand Model Specifications and Estimation Results is available electronically.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!