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.
2.0 PLANNING CONTEXT FOR INTEGRATED URBAN MODELING 2.! Introduction This chapter discusses the context or motivations for integrated urban modeling. Two main motivators exist. The first discussed in Section 2.2, is We transportation plarming context, in terms of the proscribed legal (policy) framework. The most important elements of this are the Intermodal Surface Transportation E~czency Act of 1991 O'SHEA), its successor, the 1998 Transportation L;quityActfor the 21st Century (TEA-21) and the Clean AzrActAmendments of 1990 (CAAA). The second, discussed In Section 2.3' is the modeling contend which is one of the means that is required to implement the transportation planning framework (de facto if not always de jure). In this context, the Travel Model Improvement Program (TMIP) is key. To furler define the operational planning context for integrated urban models, a survey of "key players" was ur~dertaken as part ofthis project. The findings Tom this survey are summarized in Section 2.4. Section 2.: then concludes this chapter with a definition of integrated modeling needs as denved from the foregoing discussion. 2.2 Policy Framework for Integrated Transportation - Land-Use Planning 2.2.1 ISTEA When it became law in December 1991, ISTEA fundamentally changed transportation planning in the United States. As Marshall [1997] comments, ISTEA changed the federal rules that govern state and metropolitan transportation planning. The Act decentralized the former "top-down" transportation planning approach, in which federal mandates largely influenced the timing and location of new state highway construction, in favor of strengthening local and regional authority in transportation plaruiing and decision-makina. ISTEA's key requirement with respect to integrated transportation - land-use planning was that transportation planning consider "the likely effect of transportation policy decisions on land use and development and the consistency of transportation plans and programs with provisions of all applicable short- and long-term land use and development piers." Cited in Shuck et al. ~ 995, pp. v-vi] Marshall notes that state and metropolitan planning agencies also were required to consider the overall social, economic, energy, and environmental effects of their transportation decisions.
TCRP H-12 Final Report From the point of view of transit, the Act also required "Marshall, ~ 9973: . alternatives to new highway construction to be given serious consideration as means of addressing congestion and capacity problems; · all modes (including transit) to be considered together as an integrated urban transportation system (multi-modal planning), and . long-range transportation plans (LRTPs) to be financially constrained and balanced against future revenue sources. Therefore, only those projects for which funding was expected to be reasonably available dunug the planning period could be included in the ERTP and in the 3-5 year transportation improvement program (TIP). Marshall also cites other impacts that ISTEA encouraged or broadened, all of which were favorable to transit and other alternatives to new highway construction: . the mandate for MPOs to have jurisdiction over the project selection process, in areas with a population of 200,000 or more (Transportation Management Areas); with jurisdiction retained by the state in smaller areas (but with cooperation from the MPO). Prior to ISTEA, state DOTs had significant influence over programs and · · · - project pr~ontles, · greater public consultation in transportation decision-making, . . . flexibility for state and metropolitan planning agencies to program non-traditional (i.e.. non-highway focussed) improvements intennodal connectivity, i.e., promoting improved coordination among the services and facilities provided by transportation and transit agencies, and Improved coordination of an MPO's transportation plans with those of adjacent urban areas. Although (improved) integration and consistency between transportation and 1-use plans was required by ISTEA, the Act did not specify the extent to which it was required. Nor did it specie the means for achieving improved integration and consistency. Local, state and federal goverrunents could have jurisdiction in transportation improvements but the mandate for land-use planning, zoning. etc., generally lay with local governments. Some local governments have used mp-reduction ordinances arid zoning regulations to better manage congestion Weiner and Ducca, 19963; however, in our experience these cannot be considered as generating the same level of impacts as capital-intensive transportation works. Nor is the impact ofthe charges in travel behavior
TCRP H-19 Final Report . . on lar~-use easily identified. Marshall notes that the requirement for greater consistency and coordination resulted in transportation and land-use plans that conform closely with each other. She cites a Florida MPO, which developed its LRTP as the transportation element of the county's comprehensive plan. Each was consistent with the other. The impacts of the coordinated transportation - land-use planning requirement also revere apparent on specific transit projects. For example. local land-use plans in the vicinity of new transit stations were to be conducive to, or promote use of, the transit facility as one pre-condition of obtaining federal funding for the transit facility. Measures could include high-density:. mixed-use zoning, integrated development around or near the transit station, and strong pedestrian and transit linkages to the station. Marshall notes that the authority given to larger MPOs in selecting project priorities has forced them to become better consensus builders among all levels of government. The relevance to integrated transportation - land-use planning is that this consensus can allow land development (which is within the mandate of local governments) to be better matched with project priorities (which are driven by the MPO). 2.2.2 TEA-21 ISTEA's authorization of surface transportation programs expired in September 1997 (although funding allocations were extended to the spring of 1998). TEA-21, the Transportation Equip Act for the 21st Century -- ISTEA's replacement -- provides authorization for surface transportation programs for another six years. TEA-21 builds upon ISTEA's initiatives; although in some cases with different emphases. Of relevance to integrated modeling, TEA-21 retains the general structure of metropolitan and statewide transportation plans described above. Representatives 199S, pp. H3810, H3812~: . _ Of particular importance [House of TEA-21 has "streamlined" the planning process by replacing He aforementioned list of fifteen metropolitan planning "factors" (and twenty statewide factors) with a shorter, broader list of seven issues which transportation plans must "consider''. No specific reference Is made to the integration of land-use and transportation plans under this provision. Issues that could be interpreted as including land-use planning (i.e., which must be supported by the transportation plans) are: Economic vitality of the metropolitan area (or, for statewide plans, the State and the nation as a whole), "especially by enabling global competitiveness, - 13
TCRP H-12 Final Report productivity arid efficiency.'' (factor A) Protect and enhance the environment, promote energy conservation and improve quality of life. (factor D) These intelpretahons are illustrative only. Clearly a definitive interpretation is not yet available, given that TEA-21 is relatively never-. Failure to consider any ofthe seven s~eamI~ned factors is not "reviewable" In court, in any matter affecting a "~arlsportation plan, a transportation improvement plarl, a project or strategy, or the certification of a planning process." Taken together, these two changes could be interpreted as suggesting a "softening" of the requirement for integrated transportation - land-use planning. However a more appropriate interpretation may be that integrated transportation - land-use planning is nova- considered under the broader rubrics of economic growth and development, the environment and energy considerations. The Conference Report on TEA-21 tHouse of Representatives, ~ 998, p. H3908] does establish the link, but softens the requirement ("encourage"), with the following wording: "In considering the relationship between transportation and quality of life, metropolitan plar rung organizations are encouraged to consider the interaction between transportation decisions and local land use decisions appropriate to each area. The language ti.e. of the seven s~eamI~ned factors] clarifies that the failure to consider any specific factor is not reviewable in court." Somewhat stronger wording appears as part of the project justification requirements for capital grants or loans. Here, the Report notes [p. H3857] that the justification for a "new fixed guideway [transit] system or extension of an existing fixed guideway system" must (among other things): . . "...recogn~ze reductions in local infrastructure costs achieved through compact land use development," "identify and consider mass transportation supportive existing land use policies and fixture patterns, and the cost of urban sprawl;" arid · "consider the degree to which the project promotes economic development. - 14
TCRP H-12 Final Report . - 2.2.3 CAAA As with ISTEA arid TEA-21, the Clean Air Act Amendments of 1990 (CAAA) require improved integration of transportation and land-use plans. Of note, ISTEA also established the Congestion Mitigation and Air Quality (CMAQ) program. In order to fund projects that enhanced . . . ~. . - 1 .1 air quality In ozone arid carbon monoxide nonatta~nment areas. Thus. MPOs that had not attained CAAA standards could use CMAQ funds for projects that would help control or reduce these ~- ~funds for projects that worsened existing emissions. At the same times MPOs could not use federal air quality problems. "Marshall, 1997] 2.3 Modeling Framework for Integrated Transportation Land-Use Planning - The Tra~vel Model Improvement Program (TMIP) was established in 1993 by the U.S. Department of Transportation and the Environmental Protection Agency. Important influences behind the program were the ISTEA and CAAA requirements for improved abilities to forecast travel demand by all modes the impacts of demand at a variety of levels (rangin , Dom policy to air quality and energy impacts), and the relationship between travel and land-use. However, even before the enactment of ISTEA and CAAA, the transportation modeling community had begun to question He internal consistency of the traditional four-step travel demand forecasting process. The basic process, developed in the 1950s and 1960s, focused upon facility expansion especially roads. Since then, the process has not changed significantly, although there have been major advancements over time in travel demand theory, algorithms and data in each of the four steps. In addition to inconsistencies among the steps (for example, the treatment of different steps as related traveler decision-processes, rather Man sequential decisions, and the movement towards activin-~-based modeling and to disaggregate modeling), the overall process is not viewed as responsive to today's needs -- notably, system optimization rather than expansion, congestion management, air quality management arid more emphasis upon the integrated movement of persons and goods. rather than vehicles. Other impetuses were the growing interest in privatization, which required travel demand forecasts at a "bankable statement" quality the development of (or desire to develop) up-to-date Ravel data sets and the popular availability of powerful desktop computers and tools (such as GIS). Weiner arid Ducca tI996] summarize TMIP's evolution and progress, from concept to six, related activity "tracks": 2 Originally five tracks. Track F. the sixth track' u as subsequently added to the original program. - 15
TCRP H-19 Final Report Track A: Outreach the purpose of which is to receive input Tom Me user community and to disseminate TMIP results. This is or wall be achieved Trough newsletters, reports, an Internet information clearinghouse, TMIP conferences and training arid technical assistance centers. Track B: Near-Term Improvements aims at incorporating the best new techniques arid approaches into the current practice of travel demand forecasting. Track C: Long-Term Improvements aims. over the On-term. to develop new waYs of forecasting travel via the TRANSOMS effort. ~ , This is a new procedure for m~cros~mutahon of Gavel demand, behavior and Impacts on a region-wide basis. It is a complete redesign of the traditional four-step process. Among its components is a population and activity simulator, in which activities are estimated for individuals as input to the simulation of daily travel plans. The population module ~ ~ ~ ~ ~ ~ ,. ~ .q ~ r,1 ~ uses household ~ntormat~on trom the Bureau ot the census. Some components of TRANSIMS are being field tested, however, the overall availability of the prototype is still some years away. . Track D: Data examines the needs for and methods of collecting validating and maintaining data that describe travel behavior. This track also included an assessment of the different types of ong~n-destination surveys, as well as other types of data, arid sponsored the development of new manuals for designing. administering and validating Ravel surveys. · Track E: Land-Use is Me track of prime relevance to this project. It is charged with improving the treatment of data collection, analytical techniques and the integration of land-use arid transportation planning. · Track F: Freight deals with the traditionally neglected area of freight Remarry modeling. The inclusion of Track E In TMIP recolonizes both the importance of upgrading the existing capabilities for integrated land-use - transportation modeling -- current American models are almost 20 years old -- and the relative lack of understanding of the relationship between land-use and transportation (notably, in the relationship of individual and household activity patterns to travel demand). Track E explicitly recognizes the importance of feedback loops between Arouse and travel demand forecasting models, and the need to provide improved land-use inputs to the travel demand forecasting process Chunk et al., 19953. Weiner and Ducca tI996] observe that the "level of effort in this track is relatively modest compared with the necessity to substantially upgrade land-use forecasting procedures and will need - 16
TCRP [I-12 Final Report to be increased in the future." An important Track E activity was the February 1995 Lanai Terse Modeling Conference tsee Shunk et al., 1995]. This conference provided both a status report on culTent capabilities arid a blueprint for furler Track E actions. It resulted in lists of suggested improvements to existing models and for the development of new models. Issues and concerns that were identified at the conference include Shuck et al., 199~: · Most existing integrated models are not sufficiently sensitive to policy issues. . - . . Most models are not easily understood by over participants and decision-makers in the planning process (politicians, senior government officials, non-modelers, developers, the public, etch. Existing land-use models are not sufficiently- well linked to travel demand or environmental models to allow a valid assessment of Me interaction among land-use, transportation and environmental impacts. There is little agreement on the theoretical underpinnings of existing integrated models and on their application. · Generally, Were is insufficient behavioral content In existing Arouse models. . Land-use models generally have an overly strong dependency upon travel demand model inputs and assumptions, with insufficient interaction between the two. · The role that models play in the decision-making process should be reviewed. . . . Any new or enhanced integrated models should have a clear. graphical orientation, in order to provide decision-makers and other participants more meaningful information. Linkages to a broader set of planning issues (police health care, services schools, etc.) also would be useful. ~ open space, Data limitations represent considerable problems to the use of existing models. Problems were cited in the availability of reliable disaggregate data by household type. as well as the need to provide appropriate and reliable information as input to the TRANSIMS microsimulation efforts. - 17
TARP H-19 Final Report Transit is not represented adequately in land-use or travel demand forecasting models. The last point -- regarding the indequate representation of transit -- requires further examination here. Specifically. Were are two concerns which should be addressed w-i~in the overall travel demand modeling improvement process: Accuracy of ridershipiorecasts. A 1989 UMTA report questioned the veracity of rail transit ridership forecasts tPickrell, 19893. It compared the initially forecasted ridership of ten raid systems with actual observed volumes. The report fourld, generally, that the actual readership was significantly lower than the forecasts. It recommended that forecasting procedures be improved in several ways: bringing the forecasting horizon closer to the present (i.e.. shorter-tenn rather tears lon~-term); . . . . . developing procedures to isolate and examine cause-effect determinants (of ridership); conducting sensitivity analyses for validating forecasting models and for assessing the effects of different assumptions, and, comparing forecasts with observed conditions elsewhere. The report generated considerable comment in the transportation community. It is noted here, in part because the TMIP aims to address some of the stated concerns (e.g., Prove the accuracy of forecasts). More Important to this pro ject, however, ~ ~ , ~ the accuracy ot transit ndershlp forecasts Is s~gmt~cant to tne simulation of land-use impacts that may result from a new transit line (and vice-versa). Therefore, it is an Important consideration for short- and long-term directions for integrated modeling. Simulation of transit ridership. Many commercially-ava~lable travel demand forecasting models allow transit demand to be simulated in considerable detail. It can be said that the state-of-the-art in modal split techniques, which is utility-based and uses disaggregate choice models, is fairIv well advanced. This concerns the demand side. However, Me supply side -- i.e., the assignment of transit Lips, and the representation of the traveler's choice (components of travel times, respective weights, reluctance to transfer out-of-pocket costs, etc.) -- can be somewhat problematic in several ways: · the difficulty of simulating a complex urban transit network; . the need to account for such difficult to measure attributes as travel time weights and value-of-time, · the scale of calibration that is required to simulate major routes, compared with the simulation of major roads, - 18
TCRP H-12 Final Resort the fact that Me transit share In most cities is smaller than that of We auto and, therefore there are relatively fewer observations on which to base the forecasts; the fact that marry transit investments consider modes that are new to a neighborhood or a region the travel behavior for which cannot be considered simply as an extension of existing transit ridership patterns. arid the fact that some models assign transit ridership on the basis of the rider's shortest path, with relatively little consideration of Me impact of crowding or other capacity restraints on the rider's choice (compared with the auto assignment, in which demand and capacity are more strongly related). Therefore, there is a need for improved means of simulating transit ridership on. at least, major routes. In Me short-term, this would involve the use of better methods of simulating the supply side of transit in existing models. In the long-term, this simulation must be incorporated in, or be integrated with, the next generation of integrated urban models. 2.4 Survey of Users Much of the literature focuses upon the "supply" side of integrated land-use models -- i.e., theory, al~ori~rns and technical aspects. Very little ofthe literature considers the "user" side -- i.e., how these models are perceived by the actual or potential user community. The ~ 995 TMIP Land Use Modeling Conference is an obvious arid important example of the user side s perspectives. However, our focus here is upon "softer" issues and concerns that were identified by the different perspectives, on the basis of their individual experience and knowledge, conditions in their · · . ~ sc actions, etc. To obtain the state-of-the-practice user-side perspective, interviews were conducted with ~ ~ selected "players," or interests, In integrated transportation - arouse modeling and planning. These included academics. consultants. MPOs Federal (DOT) officials members of the Project Pane] and other interests in the transportation - land-use planrung process. The sample was not intended to be random or scientific. The respondents were chosen in order to provide a broad perspective, Tom a detailed technical ur~derstar~ding to ~ broad appreciation for the issues. MPOs were also differentiated qualitatively by size, economic structure (i.e., fast-growth Sunbelt Is. siower-growth Northeastern city). transit- or auto-onentation, integration of land-use and transportation planning arid the role of moclels in planning decisions (land-use and/or travel demand models). In total ~ ~ respondents were successfully contacted and interviewed. The ~ ~ respondents represented the following: - ?~9
TIP H-17 Final Report 4 MPOs or regional land-use/~ar~sportation planning agencies. These respondents were users of travel demand forecasting models. Expenences with integrated lo. transportation - land-use models varied In both level of detail arid satisfaction. · 2 Federal of finials actively Involved in ISTEA / TEA-2 ~ and TMIP. 2 consultants involved in Untempted transportation - land-use planning. Neither is a modeler per se. · 2 researchers/academics in transportation - land-use planning. with experience in or an understanding of modeling. · ~ member of a development ~ndus~y research institute. Four of the respondents were members ofthe Project Panel. A series of fourteen linked questions was asked. The questions were deliberately open- ended. They were designed to gain a perspective of We relationship between transit accessibility and land-use decision-making, and how (or ill integrated -- or other -- models could be used to develop and explain this relationship. Table C. ~ in Appendix C lists the fourteen questions. The table shows that some of these had related questions, which were used to explain or detail the main question. It also lists the reason for the question (what it was intended to uncover), and summarizes the main themes and ideas that were provided in response. As expected, there was a wide range of responses to the questions. However some common themes emerged, as follows: . Some respondents felt that ISTEA's intentions3 (in requinug greater consistency between transportation and land-use plans) were deliberately vague, in order to promote the need for integration without being seen as interfering with local jurisdictions In larld-use planning. Others saw the requirement as a reactive approach (i.e., identifying what transportation services were required to serve a given land-use scenano), while still others saw this a more open process (i.e.- which transportation services best suited a given land-use scenario, or which land-use best suited a given transportation scenario). Finally, others noted the linkage among land-use (which generates travel), transportation arid air quality. 3 The survey was conducted prior to the enactment of TEA-2 1. Hence, all references are to ISTEA. However, given the basis of TEA-2 1 in ISTEA, the comments still stared. - 20
TCRP H-12 Final Report Respondents implemented the ISTEA requirement In different ways. Some used integrated models, although these could not mode} land markets. Another did not use integrated models, but used improved data and surveys to upgrade existing travel demand models, and developed re~ion-~de transportation - Iand-use plans. . . . Most respondents cited a Parke of policy issues that must be addressed in an urbar~ area. ranging from long-term plans, to corridor- arid site-specific planning arid development. Also noted was the need to account for. or provide information to, political decision-making and the decisions of developers and buyers. Similarly, a range of transit issues also must be addressed. These included infrastructure needs (such as a new- ART lined, but also improved service on existing facilities (such as higher frequency bus service). Another noted We need to examine ridership in the face of decTirung populations. Virtually all respondents cited the need to be able to mode] the behavior of Ian markets and land rents at a microscopic level -- i.e. at the level of individual households and individual land owners arid developers. A related need concerned the ability to simulate decision-making charactenstics and relationships at these levels (householders, tared owners and developers). Some respondents suggested that political decision-makina also should be simulated, although all noted the difficulty in doing so. Others suggested that Me models should not go to this stage; instead, serving to provide more "informed" input to political decisions. The latter respondents noted that the political decision-making process was inherently less "rational" and, therefore, less easy to simulate. Some respondents focused upon the locational aspects of iar~d-use decisions, whereas some others noted the need to model the economic context of Iand-use decisions. The importance of transit accessibility in Arouse decisions varied, depending upon the commitment of the local gove~nent~s) to transit investment, arid their desire (or ability) to promote Me return on the investment through special transit districts near stations. high density zoning near stations and integrated transportation links to the station. Most respondents noted that transit accessibility, however, was not a high-ranking factor in a developer s decision to build at a particular location (or what to build). It was one of maIly factors, tangible to a varying degree. One respondent noted that public policy could direct development by promoting certain corridors as part of an
TCRP H-17 Final Report overall Packager of favorable attributes which would include good transit accessibility. . . Satisfaction with existing integrated land-use models vaned. The most common theme -- among users only -- was that these models were too coarse to simulate the transportation - land-use interaction at anything more detailed than a regional or policy level. Other comments were: some models lacked an economic developmental perspective (i.e.. explicit incorporation of land rents. land markets, etc.~; o the models were not sufficiently sensitive to simulating the impacts on land use of changes in the transposition network and vice-versa (i.e., at a sub regional or corridor level); o there was a need to capture the dynamics of land-use over time; o o the complexity of the land-use and development process was difficult to model (but must be addressed); and current land-use data are poor (with respect to quality. level of disaggregation and/or availability), especially employment data. Satisfaction with current methods of simulating transit demand and accessibility (i.e., with Ravel demand models) also varied. Some respondents noted that Gavel demand forecasting models were not responsive to current issues (such as air quality, demand managements etc.) nor could they simulate non-~aditional modes or the overall trip chain well. When probed about the specific representation of transit demand and accessibility, some respondents noted that they had made deliberate and detailed efforts to incorporate and simulate explicitly the various components of transit accessibility (e.g., out-of-vehicle and in-vehicle times). as well as other alternatives to the auto (walking. bicycling, etch. However, this effort depended upon the specific urban area and its resources. When asked whether they had the necessary stools to properly assess the transit - land-use interaction, respondents identified several needs, notably: o improved simulation of accessibility (walk v. drive) to transit; o boarding/aTightin~ dynamics at stations; - 22
TCRP H-12 Final Report o ability to measure impacts of transit corridor investments (lack of sensitivity); arid need to Improve algorithms and strengthen the accuracy of land-use modeling (i.e., as opposed to travel demand modeling. . Only a few respondents were familiar wed TMIP. These respondents noted that TMIP had raised the transportation cornmuni~7's awareness of several issues (for example. data quality and availability). However, the role of TMIP activities in fulfilling the integrated transportation - land-use requirement depended upon the outcome and products of the program, which were still under development. (On the other hand, this suggests Mat it is not too late to incorporate integrated models with the other four tracks.) When asked if Integrated land-use models could serve as one of these tools, one respondent noted that the ability to simulate household behavior and attributes at a disaggregate level, and the need for appropriate data, were of more use to him than integrated models. The difficulty In acquinug or maintaining t~me-senes data on land rents was cited as a major obstacle to mode! application -- i.e., there was a need to simulate land rents in the models, and also to develop the necessary data. Other obstacles to the use of integrated models were: 0 data quality, accuracy and management. o credibility and acceptability of the mode} results by decision-m~ers, developers. etc.; 0 the importance of"non-technical" factors (i.e., politics) in planning decisions; o level of complexity of models, as well as the need for familiarity with how they work; and 0 level of effort (cost) required to develop and apply the models. 2.5 Synthesis and Problem Definition The need for improved integration of transportation and land-use planning is established in ISTEA, TEA-2 1 and CAAA. These acts govern the practice of urban transportation planning in the United States. TMIP recognizes the importance of developing analytical tools which are capable of supporting this integration. However, the existing integrated models do not meet the requirements ~,~
below: TCRP H-12 final Report that have been identified in the modeling, academic and practitioner communities. The main requirements for integrated transportation - land-use models are summarized The economics of land markets and rents must be incorporated in the models. This reflects Me need for a broader Reagent of the decision-making on the demand side (i.e.. mbividu~s and households) as well as the supply side (i.e., developers and land owners). A. 4. A. 6. A more aynamic feedback between the gavel Remarry modeling component and the land-use modeling component is required. This reflects technical modeling requirements' but also may require different approaches to transportation - land-use planning -- i.e., that transportation and land-use plarls should be developed ~ tandem with each, rather than one reacting to the other. Ultimately. there is a fisher need to link larld-use, transportation arid air quaTib. Inte.~=ratea7 models must be capable of addressing macro issues (for example, region- w~de policy analysis) and micro issues (for example, the impacts of and on individual transportation facilities or developments). Integrated models must be capable of providing graphicaZ/y-oriented information to decision-makers and stakeholders at all levels: politicians, developers. the public, etc. InteOoratea' models should provide improved information for political decision- mak~ng, rather than simulating the political decision-making process itself: This allows the models to address the technical issues at hand, thereby allowing more informed debate and the ability to address a broader range of concerns. With regards to forecasting transit Impacts on land-use (and vice-versa), there is a need for the improved! simulation ant! assignment of transit demand on specific facilities. These reflect the need for Unproved measurement of transit demand. These may be considered most appropriately in travel demand forecasting models (existing and TRANSIMS), which is beyond the scope of this project. How-ever, there is a need to incorporate these improved capabilities in integrated models, and to better replicate both the changes in land-use that may result and readership that is induced as a result of land-use changes. 7. Disaggregate household information shoul~form the basis of the integrated model. This is consistent with the TMIP direction of microsimulation. but also with the need to simulate better the dynamics of household decision-making. - 24
TCRP H-12 Final Report 8. Data must be compatible -- In quality, scale, availability-- and replicability -- with any improved" modleling capabilities. In particular. land market data are problematic. since relative few- data exist at the requiem level of detail and in sufficient time points In time to support Mends analysis. model development. etc. The opportunity exists to exploit We capabilities of GIS (geographic ir~ormation systems), to manage arid present data at the required level of detail. 9. The travel demand component of the integrated models should be consistent with current dlelJelopments, including activity based modeling, replication of individual traveler's choices and time-of-day variations In Ravel pattems; in addition to the TRANSIMS microsimulation approach. . These issues define the directions that should be followed in the development of the Tong- term integrated model. However. it is appropriate also to consider the role of integrated models in TMTP, for two reasons: If the ~nte=~,rated model is to serve as a too} for policy analysis in a public forum, an important role for it wall be in scenario development in addition to scenario analysis. The term "scenario" refers both to transportation network or service scenarios and to land-use scenarios. ~- The TRANSIMS activity-based travel demand module estimates individuals' activities, their characteristics and locations (where, ~ hen and why people want to travel). These estimates are based upon the characteristics of individuals, their households and vehicles, which are determined by a synthetic population generator. However. there is (or will be) a need for a more fundamental and broader ability to examine He allocation of human activities over time, before these can be input to the synthetic population generator. As it stands now, Track E (land-use) stands separately from the other four (original) TMIP Racks. Its reJahonship is not fully defined. which perhaps reflects the state of our understanding of the transponahon - iand-use interaction (let alone how to simulate it). Likely for the same reason, TMIP's emphasis, quite naturally has been on the other tracks focused on the simulation of travel Remarry. We note Hat the current project contains procedural similarities. and in many cases similarities in content, with each of the other four (original) tracks: outreach, fatal near-term improvements and long-term improvements. These tracks -- A through D, respectively -- clefine a logical sequence of considerations. - 23
TCRP H-17 Anal Report Therefore, this project was considered in terms of its individual components. Strictly speaking, TCRP Project H-12 was not conducted under the auspices of TMIP. However, it represents a significant contribution to Track E and, we believe, wall help move that ~ack's activities forward. Therefore, we propose that Track E also adopt the approach descnbed herein. This would describe a four-track integrated mode! process that is parallel to the four existing travel demand tracks. It also allows direct integration =d coordination, at the appropriate time, of similar tracks (for examples near-term improvements in integrated models with near-term improvements in travel demand forecasting procedures, etc.~. - 26
