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.
17 Chapter 3 summarized the process the research team employed to select the approach used as the basis for devel- oping the estimation methods described in this chapter. The following sections present two proposed methods, both of which are based on the selected approachâbaseline ITE rate adjustment based on empirical data. While both meth- ods use the same computational procedure, they derive adjustment factors in different ways and are applied under different conditions. This chapter: ⢠Presents an overview of the selected approach and the procedures for applying the two estimation methods, ⢠Defines a system of context classifications and provides guid- ance for qualifying a site as infill and for selecting appropriate proxy sites, and ⢠Describes the development of adjustment factors for both methods. 4.1 Overview of Approach Figure 4.1 illustrates the overall approach for estimating infill vehicle-trip generation based on adjusting baseline ITE vehicle-trip data. As shown, person trips are the common denominator allowing the conversion between baseline ITE and infill trip generation. The methods of deriving the adjust- ment factors, described in the subsequent sections of this chapter, are represented by the infill data input boxes (Steps 1 and 3) in Figure 4.1. The approach has five primary steps: 1. Baseline ITE trip generation data are used to estimate the vehicular trip generation of the proposed infill development. 2. Baseline mode share and vehicle occupancy adjustment factors are used to convert baseline vehicle-trip estimates to baseline person trips. 3. An infill mode-share adjustment factor representing the appropriate context is used to convert baseline person trips to infill person trips where the person travels by automobile. 4. An infill vehicle occupancy adjustment factor represent- ing the appropriate context is used to convert infill per- son trips where the person travels by automobile to infill vehicle trips. 5. Infill vehicle trips are used in the evaluation of site traffic impacts. The use of person trips as the common denominator between baseline ITE data and infill data underscores an important assumption in this research study: that land uses in single-use suburban environments (baseline sites) generate approximately the same quantity of person trips as land uses in dense urban environments (infill sites). The assumption that the quantity of person trips generated by a unit of development for a given LUC (e.g., 1,000 gross square feet of floor area, a multifamily dwelling unit, or one seat in a movie theater) is the same regardless of context has been historically supported by land use planners and economists, who often use average employment densities to convert employees to building floor area and vice versa. For example, the amount of floor area per employee for an office building typically ranges from 250 ft2 to 400 ft2. Variations substantially outside of this range are usually associated with the type of land use and not the context in which the land use is located. The research team analyzed per- son trips per household in one metropolitan area and found that, statistically, there were no significant differences in trips between households in different contexts (10). While vari- ability from site to site is expected, on average, ITE data, land use data, and socio-demographic data support the assump- tion that person trips remain constant across the spectrum of contexts. C H A P T E R 4 Development and Application of Methods for Estimating Infill Trip Generation
18 4.2 Proposed Methodology This report proposes an approach of adjusting ITE trip generation data (rates or trips) using a single computational procedure that employs baseline and infill adjustment factors consisting of (a) mode share or percentage of trips made by non-automobile modes, and (b) average vehicle occupancy. The derivation of these adjustment factors under this proposed approach may result from one of two methods, depending on the type of impact analysis or study for which the practitioner is estimating infill trip generation. Table 3.2 summarized the methods, including guidance on which method to use under different conditions. 1. Proxy site method â Uses empirical data collected from a site or sites that serve as a proxy for the proposed project to obtain mode share and vehicle occupancy to adjust base- line ITE trip generation data. There are two variants of the proxy site method: (a) Minimum data collection â Allows the practitioner to derive adjustment factors more quickly and less expen- sively than the other methods identified in this report. A reduction in the time and effort to collect data may be achieved by requiring collection of only the essen- tial minimum data and using basic techniques to sur- vey proxy sites within contexts similar to the proposed project. Not all proxy sites qualify for the minimum data collection variant. (b) Comprehensive data collection â Derives adjustment factors from data collected using multiple techniques to survey proxy sites within contexts similar to the pro- posed project. This variation of the proxy site method is used when the complexity of the site or its surround- ing context precludes the minimum data collection variant, or when more detailed traveler, site, or demo- graphic information is desired. 2. Household travel survey method â Derives mode share and average vehicle occupancy for a particular land use and context by extracting data from the linked-trip data- base of a regional HTS conducted for the metropolitan region within which the practitioner is preparing a study. The household travel survey method has applications and limitations that are different from those of the proxy site method. While it can be used to estimate infill trip gen- eration, the household travel survey method is applicable in broader, more macroscopic circumstances, as described later in this chapter. 4.3 Application of the Proxy Site Method Step 1: Determine the study area context and identify the infill proxy site Qualifying a proposed project as infill development requires examining the development site itself as well as the context within which the site is located. To do this, the practitioner needs to be able to distinguish between a site having travel characteristics consistent with baseline ITE data and a site whose travel characteristics are more varied. Most of this dis- tinction is found in the attributes describing the surrounding context. Because there is not a common definition of infill devel- opment, nor are there widely accepted quantifiable metrics to define different types of context, the practitioner is com- pelled to employ subjective methods to determine if a study site is infill. The application of subjective methods is compli- cated by the fact that human perceptions of the built environ- ment vary greatly as a result of a variety of biases, including local conditions. Considering this and the likelihood that quantifying classes of context at a national scale would be difficult to apply consistently and thus would lack credibility, Diagram representing the approach for estimating vehicle trip generation for infill development based on adjusting ITE baseline trip generation data. The infill data used in the process can be derived from the methods described in this chapter. Note: TIAs = transportation impact analyses. Figure 4.1. Approach for estimating vehicle trip generation.
19 the research team chose to present a flexible and adaptable system for classifying context. The primary requirement of the system is consistency in its application, especially when comparing potential proxy sites with the proposed project being studied by the practitioner. Qualifying a site as infill development starts with the definition selected for this research study in Chapter 2: Infill development or redevelopment is located in fully built areas, often in and around business districts; is walkable; is served by convenient frequent transit; is commonly served by designated bicycle facilities; and generates significant non-automobile mode shares. Using this definition as a qualitative benchmark, the prac- titioner documents the attributes of the proposed develop- ment and the attributes of the ultimate context in which the development will be located. The documented attributes are used subsequently to identify proxy sites, but also may be retained for building a database of project attributes and their associated proxy sites for use by others. The following are examples of attributes for describing con- text that are typically available in local planning and regulatory documents, extracted from existing databases, or observed in the field: ⢠General or comprehensive plan land use and zoning designations. ⢠Residential or employee densities in the surrounding district. ⢠Maximum allowable floor area ratio (FAR) applicable to the site. ⢠Minimum required setback from public right-of-way (a measure of distinguishing between automobile- and pedestrian-oriented development standards). ⢠Off-street parking requirements, allowances for off-site parking and street parking, and so forth. ⢠Public and private parking systems in proximity to site, number of spaces, utilization, and costs. ⢠Existing and planned transit system serving site and vicin- ity (routes, stations, frequencies). ⢠Existing and planned bicycle facilities in vicinity of site and connections to regional system. ⢠Measures of walkability (qualitative or quantitative). The following are examples of attributes for describing a proposed infill development project as well as describing an appropriate proxy site. These attributes are obtained from a combination of the developer and local planning and regula- tory documents: ⢠Project size, number of units, floor area, or expected number of employees. ⢠Project density or FAR. ⢠ITE LUC and general or comprehensive plan and zoning designations. ⢠Preliminary site layout, building orientation, parking spaces, and parking facility orientation. ⢠Walking distance to nearest rail or high-frequency bus transit station or stop. ⢠Proximity to nearest bicycle facilities, and identification of obstacles and barriers to bicycling. ⢠Site pedestrian access. The underlying assumption of the proxy site method is that similar uses in similar contexts will have similar trip-making characteristics. As such, being able to determine that the study area and the proxy site are located in similar contexts is a basic requirement. The combination of area type and type of public transpor- tation designates the context for use in the proposed method- ology and determines whether the method is applicable given the conditions of the study area. In the Institute of Transpor- tation Engineersâ recommended practice, Designing Walkable Urban Thoroughfares: A Context Sensitive Approach (11), the definition of urban areas is based on the concept of âcon- text zones.â Context zones are a discreet set of development- intensityâbased categories on a scale ranging from the most rural or undeveloped area to the most urban or developed area. Although the Designing Walkable Urban Thoroughfares recom- mended practice was developed for the purposes of thor- oughfare design, the context zone system is a useful method for stratifying urban areas whose unique characteristics may affect trip generation. The four zones used to define urban context, listed in increasingly urban conditions, are: ⢠Suburban center (CZ-3), ⢠General urban (CZ-4), ⢠Urban center (CZ-5), and ⢠Urban core (CZ-6). Any context that does not fall into one of these four designa- tions is either a special district, such as a university campus or an airport, or the context is single-use suburban, exurban, or rural, with little or no transit service, and should be analyzed using conventional ITE trip generation rates. The four designations listed previously can be applied to numerous contexts. For example, a suburban town with a population of 40,000 can have the equivalent of an urban core, but the trip generation characteristics of development in this community could be far different from development in the urban core of a large metropolitan city with a popula- tion exceeding 1,000,000. Therefore, the research team fur- ther narrowed the definition by including the type of public
20 transportation serving the subject area. Contexts that qualify for infill development are served by rail or high-frequency bus transit (12). Contexts served only by conventional low- frequency bus transit, even when in close proximity to the proxy site, do not qualify. Figure 4.2 presents diagrams and describes characteristics representing the contexts used in this research study. The diagrams are not intended to visually match a given study area but to illustrate the difference in the key physical site and building attributes between the cat- egories. The characteristics described in Figure 4.2 may assist practitioners in consistent identification of context. Finally, the following qualitative attributes may be used as general descriptors of areas qualifying for infill development: (a) Compactness â A monocentric form of development pat- tern in which the metropolitan area of study has a con- centration of its population within a specified distance of the urban core. Compactness is also represented by a Context Zone Characteristics Suburban General Urban Urban Center Urban Core Land use Low-density, single- use development. Some horizontally mixed-use development, but mostly segregated. Many auto-oriented uses such as big-box retail and drive- through restaurants. Moderate-density mix of single- and mixed-use development. Some auto-oriented uses focused on office parks and urban shopping centers. Moderate- to high- density development. High building coverage of property, with open space between buildings. High- to very high- density development. Very high building coverage of property, often with buildings attached. Orientation of building on site Buildings have large setbacks from street. Buildings oriented toward parking rather than street. Buildings primarily oriented toward street, but some oriented toward parking lots. Buildings integrated into the sidewalk with stoops, arcades, and cafes. Buildings integrated into the sidewalk with stoops, arcades, and cafes. Building height and separation Typically one- to two- story buildings. Buildings do not form a street wall or street enclosure, creating a sense of wide, open space. Mid-rise buildings of one to four stories that partially create a sense of definition. Usually space between buildings reduces the sense of enclosure of the street. Mid-rise to high-rise buildings. Buildings create definition but may have spaces that reduce the sense of enclosure of the street. High diversity of scale and variety of buildings. Tall, high-rise buildings are common. Buildings create definition and a street-wall enclosure. Very high diversity of scale and variety of buildings. Pedestrian access Indirect or nonexistent pedestrian connection to building entries from street. Mix of direct and indirect access to building entries from street. Direct pedestrian connection to building entries from street. Direct pedestrian connection to building entries from street. Parking Primarily in surface lots between buildings and street. Zoning requires all required parking to be contained on site and exclusive to the site. Mixture of surface and structured parking. Off- site and on-street parking may be allowed in lieu of exclusive on- site parking. Predominantly structured parking. Off-site and on-street parking may be allowed in lieu of exclusive on-site parking. Parking may be accessed by alleys. Predominantly structured and underground parking, both private and public. Parking typically accessed by alleys. Notes: Adapted from Designing Walkable Urban Thoroughfares: A Context Sensitive Approach. An ITE Recommended Practice. The Institute of Transportation Engineers. Washington, D.C., 2010. See Appendix A for a more thorough description of each context zone. Figure 4.2. Diagrammatic description of context zones.
21 large proportion of a metropolitan areaâs land being built up (covered with buildings and infrastructure) relative to the total land area. (b) Fully built area â Defined as an area with only a fraction (typically less than 10%) of its land being undeveloped, excluding water bodies and land designated for conserva- tion use, natural preservation, public road rights-of-way, and recreation. (c) A mix of residential and employment â Defined as an area having a generally balanced ratio of jobs to hous- ing units. Contexts where either jobs or housing repre- sent a significant proportion of the land use (e.g., 70% to 100%) have a poor mix and generally will not experience the same level of trip capture as a context with a balanced mix. This attribute may be determined qualitatively by simply observing the mix of uses within the study area, or it may be calculated using available data. (d) A continuous and interconnected pedestrian circula- tion system â Referring to a walkable pattern of blocks and streets having continuous sidewalks and intersection crossings with few or no major obstacles to pedestrian cir- culation within the study area such as freeways, waterways without crossings, or extreme topography. This attribute can be visually ascertained or can be measured through a connectivity index (e.g., a minimum number of inter- sections per square mile). (e) Sites located within walking distance of rail station or a high-frequency bus route â This attribute establishes access to public transit but does not necessarily limit the study sites to TOD. For purposes of assessing proxy sites, the maximum walking distance to a rail station is gener- ally considered one-half of a mile, while the walking dis- tance to a high-frequency bus stop (headways of typically no more than 20 minutes) is considered one-quarter of a mile. Appendix A contains a more thorough description of each of the four context zones used in this report. Step 2: Convert baseline ITE vehicle-trip generation to baseline ITE person-trip generation Start with vehicle-trip generation data from the ITE Trip Generation Manual for the land use classification for which trip generation estimates are desired. The ITE Trip Generation Manual contains guidance on estimating vehicle-trip generation. The directions can be found in Volume 1 of the ITE Trip Generation Manualâ Userâs Guide and Handbook. To convert baseline ITE vehicle trips to baseline ITE person-vehicle trips (person trips employing an automobile mode of travel) requires knowl- edge of two factors: (1) the percentage of trips made by non- vehicle modes of travel represented by the baseline ITE trip generation estimate, and (2) the vehicle occupancy assumed in the baseline ITE trip generation estimate. Referring to the majority of the baseline ITE trip generation data, the most recent version of the Userâs Guide in the Trip Generation Manual (3) states: Data were primarily collected at suburban locations having lit- tle or no transit service, nearby pedestrian amenities, or travel demand management (TDM) programs. However, the term âlittle or noâ implies that there may be some trips, albeit a small fraction of the total trips made by transit, walking, or bicycling inherent in the trip generation rates. Adjustment factors for baseline ITE trip generation data may be derived from other conventional trip generation studies that have the data or may be or collected by the user following the proxy site method. These approaches are described further in the section on deriving infill adjust- ment factors. The conversion of ITE vehicle trips to ITE person trips uses Equation #1: = à â + ï£«ï£ ï£¶ï£¸ï£®ï£°ï£¯  Person-Trips VehTrips VehOcc 100% %Transit %WalkBicycle BASELINE BASELINE BASELINE BASELINE BASELINE Where: ⢠Person-TripsBASELINE = baseline ITE vehicle-trip estimates converted to baseline ITE person trips by all modes of travel; ⢠VehTripsBASELINE = Vehicle-trip generation estimate from the ITE Trip Generation Manual for the subject site; ⢠VehOccBASELINE = Average baseline ITE vehicle occupancy in the baseline ITE trip generation estimate, as input by the user; ⢠%TransitBASELINE = Average transit mode share assumed in ITE trip generation rates; and ⢠%WalkBicycleBASELINE = Average walk and bicycle mode share assumed in ITE trip generation rates. Alternatively, the last two values in the equation may be replaced with the single value: %NonAutomobileBASELINE. Step 3: Convert baseline ITE person-trip generation to infill person- vehicle-trip generation This step converts the baseline ITE person-trips estimate from Step 2 into the equivalent number of person trips using
22 an automobile mode share in infill areas. Step 3 converts the baseline ITE person-trip estimates to infill person-vehicle trips using Equation #2: Person-Vehicle-Trips Person-Trips 100% %Transit %WalkBicycle INFILL BASELINE INFILL INFILL ]) [ ( = à â + Where: ⢠Person-Vehicle-TripsINFILL = Infill person trips using vehic- ular mode of travel; ⢠Person-TripsBASELINE = Baseline ITE vehicle trips converted to baseline ITE person-tripsBASELINE from Step 2; ⢠%TransitINFILL = Average transit mode share applicable for specific infill area based on data collected in the proposed methodology (see section on developing adjustment fac- tors); and ⢠%WalkBicycleINFILL = Average walk and bicycle mode share for specific infill area based on data collected in the pro- posed methodology (see section on developing adjustment factors). Alternatively, the last two values in the equation may be replaced with the single value: %NonAutomobileINFILL. Deter- mination of the applicable transit and walk/bicycle mode shares, or the percentage of all trips by other than an auto- mobile, is the critical part of this step and requires input from one of the proxy site methods. Step 4: Convert infill person-vehicle trips to infill vehicle trips The proposed methodology culminates in the calculation of vehicle-trip generation for the infill site. These adjusted vehicle-trip generation estimates are then used in the con- ventional traffic impact analysis process. The final conversion uses Equation #3: =Vehicle-Trips Person-Vehicle-Trips VehOcc INFILL INFILL INFILL Where: ⢠Vehicle-TripsINFILL = Vehicular trip generation adjusted for urban infill conditions; ⢠Person-Vehicle-TripsINFILL = Infill person trips using vehicle mode of travel resulting from Step 3; and ⢠VehOccINFILL = Persons per vehicle based on local data collection. 4.4 Developing Adjustment Factors for the Proxy Site Method The computational procedure described previously requires obtaining or deriving mode share and vehicle occupancy adjustment factors for baseline and infill equations. Baseline ITE adjustment factors are not context-specific and may be obtained from the literature, extracted from data collection conducted by others, or collected directly by the user. Infill adjustment factors are context-specific and need to be derived from carefully selected proxy sites in order for the estimation approach to produce reasonably accurate and credible results. This section describes data requirements for the variants of the proxy site method (minimum data collection and com- prehensive data collection). In addition, this section: ⢠Provides guidance on selecting proxy sites, ⢠Identifies the conditions that warrant the use of either of the methodâs variants, and ⢠Describes the pre-survey planning that results in the most effective use of limited resources. 4.4.1 Baseline Adjustment Factors Sources of baseline adjustment factors include technical literature, mode-share studies of baseline land uses, and orig- inal data collection documentation. Baseline data may also be newly collected at sites that represent ITE baseline condi- tions. The research team identified several specific sources for obtaining baseline data adjustment factors, including: ⢠Data in the ITE Trip Generation Manual or other original studies to determine if non-automobile mode share and vehicle occupancy data were included in the studies. For example, some sources of trip generation rates, such as the Trip End Progress Reports from the California Depart- ment of Transportation, contain data on transit, walk, and bicycle modes and vehicle occupancies for certain uses. ⢠Non-ITE trip generation data of isolated suburban land uses that contain non-automobile mode share and vehicle occupancy data. These data are frequently collected for traffic impact studies and filed away or retained as propri- etary information. ⢠Estimated values for transit, walk, and bicycle mode share and vehicle occupancy using regional HTS data. ⢠Data collection for mode share and vehicle occupancy at sites similar to the site represented in the ITE trip genera- tion land use categories using the proxy site method. In the event that these sources fail to provide applicable data, a conservative approach assuming zero transit, walk, or
23 bicycle trips, and a low vehicle occupancy (a range of 1.02 to 1.05 persons/vehicle), is recommended for use. 4.4.2 Guidelines for Selecting a Proxy Site Selecting an appropriate proxy site for collecting data is one of the most important aspects of the method. A poorly selected site could result in the significant under- or over- estimation of infill trips. The following guidelines are provided to help the user select proxy sites: ⢠Create a list of the attributes of the proposed project site and its surrounding context representing a time at build out of the project, to the extent this information can be documented (particularly regarding the surrounding context) without speculation. The attributes need to be measurable/observable without undue effort. The context attributes shown in Figure 4.2 can be used as a basis for establishing these attributes. ⢠The selected site should substantially represent the pro- posed project in terms of attributes such as land use, size of development, density or FAR, mix of uses (if applicable), parking supply and proposed parking costs, vacancy rate, and maturity of the development. The user must be able to obtain critical independent variable information about the site, such as gross leasable floor area, or number of dwell- ing units. ⢠The user should attempt to locate a proxy site within prox- imity of the proposed project. If this is not possible, the selected site should be located in a context that substan- tially represents the baseline or infill context (or future infill context) of the project, including network density; type, proximity, and frequency of transit services; level of walkability and bicycle accommodation; density of sur- rounding land uses; and similar amount and availability of off-site parking. See list of context attributes in Step 1 as an example. ⢠GIS mapping may be used to identify physical, regulatory, and demographic attributes as the context of the proposed project. Use queries to map TAZs or census tracts/blocks that might contain the desired attributes for proxy sites. ⢠Use navigational mapping software or websites (e.g., Google Earth) to search and plot the location of businesses or places of similar land use types as the proposed project. Walk Score or similar tools can be used to locate neighbor- hoods or districts with similar walkability traits as a start- ing point for searching for proxy sites. ⢠While data collected from multiple sites are statistically more robust, data collected from a single proxy site may be acceptable as long as the survey planning identifies sites with average to above-average activity, collects data on days and time periods typically representing the peak hours of the land use being studied, and avoids times that may sig- nificantly affect the data, such as holidays, nearby closures of major streets, or days when large special events occur nearby. 4.4.3 Infill Adjustment Factors If mode share and vehicle occupancy data reflecting both the land use and the context of the proposed project are read- ily available, then they may be used directly. If the data are not available, the user may collect the data at a site or sites representative of the subject land use and within the same type of context as the proposed development. Following is an overview of the user data collection requirements for each of the two variants of the proxy method: ⢠For the minimum data collection variant, the site must be configured with its own exclusive parking supply oriented so that vehicles entering or exiting can be observed and counted. The siteâs parking should be sufficient and conve- nient enough so that site users have no need to park off-site and walk to the site, ensuring full capture of the siteâs traf- fic generation. The site must be oriented so that observers can view all entrances, including rear and employee-only entrances. The minimum data collection required for the minimum data collection variant is: â Vehicles entering/exiting site during the a.m. or p.m. peak period of adjacent street traffic. â Number of persons entering and exiting all entries to the subject building(s) on the site, and â The number of persons in vehicles entering or exiting the site. For the comprehensive data collection variant, there are few restrictions on the physical configuration of the siteâs buildings and parking because intercept surveys capture trips that can- not be observed or easily counted. Guidance for collecting data using the comprehensive data collection variant is as follows: ⢠It may be necessary to gain the permission of the proxy site owner/management to conduct intercept surveys of the siteâs employees, visitors, or customers. ⢠The siteâs access points must be oriented so that observers can view all entrances, including rear and employee-only entrances, or observers need to be placed at every building entrance to ensure a thorough count of person trips. ⢠A good resource for planning site-specific comprehensive travel surveys, including strategies for selecting sites and gaining management permission, setting up survey person- nel, innovative tools to improve participation in surveys, and lessons learned, is Trip-Generation Rates for Urban Infill Land Uses in California, Phase 2: Data Collection, Final Report (1).
24 planning horizons. Planning horizons can range from the cur- rent year to 20 years or more in the future, or whatever period of time the proposed project would take to fully build out. The scenarios developed under these timeframes evaluate the effects or impacts of the traffic generated by the pro- posed project when combined with current transportation conditions and when combined with the cumulative traffic forecasted to occur in the future. Traditionally, the analysis of planning horizons identifies near-term, project-specific impacts (impacts caused solely by the project and for which the project is solely responsible for mitigating) and long-term cumulative impacts (impacts caused by the cumulative growth in traffic to which the project contributes and is responsible for mitigating its share of the impact). When selecting a proxy site or sites, the practitioner needs to consider the planning horizons of the impact analysis being prepared and choose a site or sites within contexts that repre- sent the desired planning horizon. This may require select- ing sites in multiple contexts if the proposed project is located in an area expected to undergo substantial change over time. Conversely, if the proposed project is located in a fully built environment with little expected change over time, the practi- tioner can select one context that represents both current and future conditions. 4.5 Application of the Household Travel Survey Method Infill adjustment factors may be derived for sites proposed within metropolitan areas that have current HTS data. This method of deriving mode share and auto occupancy is lim- ited to the land use categories that can be deduced from HTS linked-trip dataâessentially only the general categories (e.g., retail, office, multifamily housing) because the data from the surveys do not always distinguish between land use subcat- egories (i.e., grocery store versus home improvement center). However, HTS data can provide adjustment factors for all context types, and more importantly, they can identify differ- ences in the adjustment factors within each context type due to geographic location and socio-demographic characteristics within a region. This method will result in adjustment factors for general land use categories within any context type, either (a) aver- aged across the metropolitan region, or (b) specific to any TAZ located in the region. Although this method can be used to generate the adjust- ment factors used in traffic impact analyses of infill develop- ment, it can be also be used for broader types of analyses, including: ⢠Creation of a region- or area-wide database of mode share and vehicle occupancies by TAZ (representing the context within the TAZ) for adjusting ITE trip generation rates for ⢠The comprehensive data collection variant does not have the same minimum requirement for data collection as the minimum data collection variant. By definition, the com- prehensive data collection variant is used when more data is desired at a study site. Its intent is to expand data collec- tion and the use of survey instruments when circumstances dictate the need for more information or an alternative approach. The types of survey instruments that may be con- sidered for this method include: â Random sampling intercept surveys to determine mode share, distance traveled to the site, and pass-by trips, and to document traveler demographics for cross-referencing; â Person-trip cordon counts at building entries; â Originâdestination surveys conducted by questionnaire or by observing trips between site and transit, off-site parking, and other land uses; â Vehicle occupancy counts; and â Automatic machine vehicle counts or video data collection. 4.4.4 When to Use Proxy Site Method Variants The minimum data collection variant serves as the default methodology for collecting data from proxy sites to derive adjustment factors. Unless additional data are desired or there are challenges in collecting the necessary data from the proxy site, the minimum data collection variant is sufficient for most applications. The comprehensive data collection variant is employed under the following conditions: ⢠When a detailed breakdown of mode share other than auto- mobile/non-automobile mode of travel is desired, or the practitioner desires traveler data that cannot be obtained from counts or observation; ⢠When the proxy sites do not have exclusive parking facili- ties, or the proxy sites are located where there is nearby but unobservable public or private parking structures, below- ground garages, and street parking where proxy site users are parking and walking onto the site; ⢠When the proxy sites experience a high level of linked trips, where travelers who drive park once and walk to multiple sites, and if the practitioner desires to determine the siteâs demand for primary versus secondary linked trips; or ⢠When the proxy sites have a nearby but unobservable rail station or transit hub, and the siteâs transit mode share by type of transit is desired. 4.4.5 Considerations for Site Impact Analysis Planning Horizons Site impact analyses typically evaluate a proposed project under a range of time-based scenarios often referred to as
25 The minimum required data for estimating infill trip gen- eration adjustments are listed in Table 4.1, which describes the required variables and how they are used in deriving the adjustment factors. These variables are generally standard in travel surveys and should be available, in one form or another, from all the major metropolitan areas. The linked trip data contain numerous other variables and helpful infor- mation for cross-referencing household, person, vehicle, and activity data. Typically, HTSs record individual segments of each trip separately every time the traveler stops for a specific purpose on the way to an ultimate destination, including when chang- ing modes. For example, driving from home to the train sta- tion, taking the train, and walking to the workplace are three segments of a single trip, each using a different mode of travel. These are called unlinked trips. However, the travel described here is actually one home-based work trip using rail transit as the primary mode of travel. Driving to the train station and walking to the workplace are secondary. The consolidation to a single trip purpose by a single mode describes a linked trip. Household travel survey records of unlinked trips are manip- ulated to produce linked trips. The linked trip data contain multiple variables from the four categories of data: household, person, vehicle, and travel/activity. Linked trip data are made up of individual trip records, each of which represents one personâs travel for an activity by the primary mode of travel. Each trip record is identified by a general trip purpose [e.g., home-based work, home-based shopping, nonâhome-based trips, start and end time of travel, mode of travel, passengers (if by auto), mode of access to primary travel mode, origin and destination activi- ties and place, and numerous other data]. sites in different locations in the region to ensure consis- tency in infill development traffic impact studies within the region or area; ⢠Scenario analyses comparing the transportation benefits or impacts of shifting growth in development between urban infill and suburban or greenfield locations; ⢠Studies of large-scale activity centers requiring an under- standing of how the centerâs mode share is influenced by its location within the region, proximity to transit, and other built-environment characteristics; and ⢠Development of local or regional trip generation rates and mode shares covering a range of contexts for inclusion in agency traffic impact analyses guidelines. 4.5.1 Required Data for the Household Travel Survey Method The data normally available from an HTS to use for the household travel survey method can be divided into four categories: ⢠Household data â Characteristics of the household and its location. ⢠Person data â Demographic, socioeconomic, and employ- ment information for one or more members of the household. ⢠Vehicle data â Type, ownership, and usage of private vehi- cles available to household members. ⢠Travel and activity data â Detailed travel, activities, and origins/destinations of the daily trips by one or more household members. HTS Variable Definition Origin activity purpose/ destination activity purpose or origin land use/destination land use Provides the activity purpose or land use of the origin and the destination of the trip â used to associate the trip with a particular land use. General purpose Provides home-based and nonâhome-based trip information â used to cross-check data and to populate adjustment factors when using travel demand forecasting model data. Primary mode of travel Provides the primary mode of travel for individual trip records (ignoring mode of access) â used to develop adjustment factor mode split. Origin TAZ/destination TAZ Provides the zone of the origin and the destination of the trip â first used to identify trip records within TAZs designated as general urban/urban center, then used to classify the trip as inbound or outbound in the extraction of peak hour records. If available, an address can be used to determine the origin or destination zone. Day of trip Identifies the day the trip occurred â used to classify trips as weekday or weekend. Start time/end time Starting and ending time of the trip â used to classify trips in either a.m. or p.m. peak period. Number in vehicle Provides the number of people in a vehicle â used to determine vehicle occupancy. Table 4.1. Linked trip data variables in deriving adjustment factors.
26 Residential trip records do not record delivery, services, or guest trips unless recorded by another participant. This should not significantly affect the derivation of mode split or vehicle occupancy of residential uses. 4.6.1.2 Restaurant Land Uses Restaurant-related trips were selected from the trip records with an origin purpose or a destination purpose classified as âmealsâ or an origin activity or a destination activity classi- fied as âeat a meal outside of home or work.â The trip records only include patron trips. Restaurant trip records do not capture employee, delivery, or service trips unless recorded by another participant. This may affect calculation of the mode split or vehicle occupancy. Employee-related trips are captured in the âworkâ trip purpose. 4.6.1.3 Retail or Shopping Center Land Uses Trips related to retail or shopping centers were selected from the trip records with an origin purpose or a destination pur- pose classified as âshopping away from the home,â âpersonal services/bank/government,â or âshop.â The âpersonal services/ bank/governmentâ purpose includes barber, beauty shop, dry cleaning, banking, and government services. The trip records include only patron trips. Retail trip records do not capture employee, delivery, or service trips unless recorded by another participant. This may affect calculation of the mode split or vehicle occupancy. Employee-related trips are captured in the âworkâ trip purpose. 4.6.1.4 Office Land Uses Office-related trips were selected from the trip records with an origin purpose or a destination purpose classified as âwork or work related.â These trip records include all work trips and do not classify the origin or destination as âoffice.â At least under the MTC variables, there is no reasonable way to separate work trips from those specific to an office build- ing except by reviewing the participantâs comments. How- ever, this was not deemed a practical or accurate way to isolate office-buildingârelated trips. While this is a potential source of error and inaccuracy, its use was appropriate because the resulting mode split and vehicle occupancy represents work- related trips for all land uses, including office buildings. Fur- ther, a comparison of the resulting mode splits and vehicle occupancies looks reasonable compared to generalized work mode splits for the entire region or within ½ mile of rail sta- tions as published by MTC. Therefore, while use of the work or work-related trip purpose captures trip records to non- office locations, the data appear to reasonably represent work trips to office land uses. Based on a review of the data variables contained in the various categories of data, the research team determined that the linked trip data contained the appropriate information for deriving mode split and vehicle occupancy for various land uses and time periods. Linked trip data were selected as the best source of data because their trip records were cross-referenced to the variables needed to calculate the adjustment factors. 4.6 Data from the San Francisco Bay Area The research team initially assessed the travel survey data from one selected metropolitan area to determine if adequate information could be extracted from the available records to generate the adjustments factors by LUC. The research team selected the HTS data from the Metropolitan Transportation Commissionâs (MTC) 2000 Bay Area Travel Survey (BATS). This activity-based survey was conducted with over 15,000 households in the nine-county Bay Area, which had a year- 2000 population of 6,800,000 in nearly 2,500,000 households. The survey data were readily available, well documented, and summarized by MTC for cross-checking against the result- ing adjustment factors. In addition, GIS data were readily available that provided information about TAZs, proximity to transit (rail and bus), and available transit headway data. 4.6.1 Variables in the Household Travel Survey Method The research team selected four land use categories for the development of adjustment factors using the household travel survey method. They are residential, restaurant, retail, and office. Because the available data are activity-based and not place-based, the type of land use at the origin or destina- tion of the trip needs to be inferred from the trip records. The linked trip data contain a variable for the activity purpose at both ends of the linked trip. From the available choices, the investigators selected those activities that best represented an activity at a specific land use. Missing from the HTS data are trips made by delivery or service people, with the exception of survey participants who are employed in these fields. This is a potential source of error in estimating actual trips but is not a significant error in calculating mode split or vehicle occupancy. The activities used to determine land uses associ- ated with trips are described in the following. 4.6.1.1 Residential Land Uses Residential-related trips were selected from the trip records with an origin purpose or a destination purpose classified as âhome.â The majority of the records for residential-related activities are from the residents themselves.
27 (used in travel demand forecasting) was used to determine employment density. The upper limits of suburban-center densities were used to determine the lower limit of general urban/urban-center context densities. Suburban densities range greatly, so the research team considered an array of criteria for suburban areas but selected the criterion documented in note 6 (see Notes and Citations section) to establish the lower limit of the GU/UC contexts. These density criteria and a combination of the criteria were used to isolate the Bay Area TAZs that meet individual housing or employee density criteria or both. This resulted in a relatively small number of TAZs when compared to the entire nine-county Bay Area, but the TAZs produced an ade- quate number of trip records to determine the viability of the extraction process. These zones were the initial pool for a series of steps that incrementally reduced the pool and the eventual size of the sample trip records. 4.7.2 Transit Type, Frequency, and Proximity Adjustment factors were developed for GU/UC land uses within ½ mile of a rail (or ferry) station and within ¼ mile of a high-frequency bus corridor (defined as a line or combina- tion of bus or bus rapid transit lines with a maximum 15-min headway for 6 or more hours of the day). For the San Fran- cisco Bay Area, GIS was used to map rail and ferry stations and to identify a ½-mile walk buffer around each station. A similar analysis was prepared for high-frequency bus lines (13). 4.7.3 Other Criteria The remaining two criteria for selecting TAZs that repre- sent the specified type of urban context are âcompactnessâ and âcontiguous and continuous urban street system.â These two criteria cannot be distilled from the household travel sur- vey method data, and for all practical purposes are qualita- tive rather than quantitative. The intent of these criteria is to enable the analyst to confirm that the site is located within a walkable areaâdefined as an area with pedestrian facilities throughout a study area that is compact enough and diverse enough for most daily needs to be met by walking. 4.7.4 Selecting the Geographic Units of Data The unit of geography used for tabulating HTS data is typi- cally a form of known zonal system. Most HTS databases con- tain cross-references for multiple zone systems such as census tracts, census block groups, census blocks, public-use micro- data areas, and TAZs. Most of the geographic units are too large and relatively unknown to those who prepare traffic impact analyses to be considered for the level of data manipulation 4.6.1.5 Limitations of the Office Data MTCâs activity at origin or destination does not distinguish work or work-related trips by type of workplace except through participant comments. 4.7 Procedure for Applying the Household Travel Survey Method 4.7.1 Determining Context Context, as used in development of adjustment factors using HTS data, is made up of: ⢠Density, intensity, and mix of the surrounding land uses; ⢠The type of, frequency of, and proximity to transit; ⢠Compactness of the surrounding land uses; and ⢠Access to a contiguous and interconnected urban street system. In the proxy method, context is determined by comparing the actual built environment with attributes delineating four context categories. In the household travel survey method, con- text is determined at the scale of the TAZ using GIS and mea- sures of urban intensity such as population and employment densities. This process is explained in the following sections. The household travel survey method was developed and tested under two context categories within the nine-county metropolitan San Francisco Bay Area. The research team chose to develop and test the household travel survey method within general urban and urban-center contexts and used GIS to fil- ter out the TAZs in the metropolitan Bay Area representing these two contexts. The criteria used to identify TAZs that were predominantly composed of these two contexts were from research to develop trip generation rates for infill development in California (1). The California study used the density of residential units, the density of employees, or the density of both to define the intensity of urban TAZs. The combination of residential and employment densities identified TAZs composed of mixed- use development, and the balance or imbalance of the ratio of jobs to housing indicated the degree to which trip inter- nalization would occur. The criteria from the California research were used to iden- tify TAZs in the San Francisco Bay Area that met the definition of general urban/urban center (GU/UC)âa range of housing and employee densities from suburban to urban core. To set the upper limit of these two contextsâ densities, the lower limit of density in known urban core contexts was used. In the ini- tial example of the San Francisco Bay Area, the known urban cores of San Francisco and Oaklandâs central business districts were used. Year-2000 census data were used to determine resi- dential density, and MTCâs year-2000 employment database
28 4.7.5 Extracting Mode-Share Adjustment Factors Extracting the adjustment factors from a linked-trip data- base involves the following major activities: ⢠Begin with a linked-trip database of all daily travel in which each record contains a trip that has at least one trip end orig- inating in or destined to a TAZ that meets context criteria. ⢠Use the activity purpose at trip origins and destinations to extract records with activities that represent the four pro- posed study land uses (residential, restaurant, retail, and office). ⢠Identify TAZs within a ½-mile walk buffer of rail/ferry stations or within a ¼-mile walk buffer of bus routes or needed in the proposed methodology. Therefore, the geo- graphic unit of TAZ was chosen for the following reasons: ⢠TAZs are often the smallest scale of geographic zones in urban areas, with the exception of census block groups or blocks, providing a reasonable resolution for focusing on trip records meeting transit proximity criteria. ⢠Trip origins and destinations in the linked-trip database are identified primarily by TAZ. ⢠Census data, as summarized by MTC in the 2000 BATS, can be aggregated at the TAZ level. ⢠Existing and future land use and demographic informa- tion (for use in travel demand models) are aggregated by TAZ and are useful for developing adjustment factors for future site conditions. GU/UC Subdivision of Records Number of Linked Trip Records Residential Restaurant Retail Office Entire Bay Area 176,083 36,827 67,295 72,275 GU/UC areas 23,763 6,123 9,086 11,154 Weekday (all GU/UC records) 20,983 5,143 7,605 10,755 Weekday â HF bus 20,372 5,044 7,354 10,519 Weekday â rail 12,787 3,303 4,501 7,007 Weekday â HF bus â a.m. peak hour 4,802 442 411 2,800 Weekday â HF bus â a.m. peak hour â inbound 256 379 374 2,751 Inbound percentage 5% 86% 88% 97% Weekday â HF bus â a.m. peak hour â outbound 4,625 63 53 88 Outbound percentage 95% 14% 12% 3% Weekday â HF bus â p.m. peak hour 3,975 616 1,318 2,708 Weekday â HF bus â p.m. peak hour â inbound 2,955 455 901 205 Inbound percentage 73% 72% 64% 7% Weekday â HF bus â p.m. peak hour â outbound 1,100 173 516 2,591 Outbound percentage 27% 28% 36% 93% Weekday â rail â a.m. peak 2,962 257 260 1,806 Weekday â rail â a.m. peak â inbound 156 215 237 1,774 Inbound percentage 5% 84% 88% 97% Weekday â rail â a.m. peak â outbound 2,855 42 33 62 Outbound percentage 95% 16% 12% 3% Weekday â rail â p.m. peak 2,506 394 803 1,793 Weekday â rail â p.m. peak â inbound 1,844 300 557 153 Inbound percentage 72% 74% 65% 8% Weekday â rail â p.m. peak â outbound 711 105 306 1,710 Outbound percentage 28% 26% 35% 92% Notes: Peak hour trips based on trip records, not trip ends; therefore, total peak hour does not equal the sum of inbound and outbound. HF bus = Proximity to high-frequency (HF) bus route stop (¼-mile walk buffer). Rail = Proximity to rail station (½- mile walk buffer). Table 4.2. Summary of linked trip records at each level of extraction.
29 corridors that meet transit proximity criteria, and extract records that have at least one trip end within these zones. ⢠Separate weekday from weekend trip records. ⢠Extract records in which at least one trip end begins or ends during a defined peak hour of adjacent street traffic (a.m. and p.m.) from the daily set of trip records. ⢠Using the appropriate variables as a filter, separate the records representing inbound and outbound trips. ⢠Separate the resulting hourly records for each LUC by pri- mary mode of access. Starting with a linked-trip database of a finite number of trips, each step in the sequence extracts records and reduces the number of records available for the next step. The ana- lyst needs to review the number of records resulting from the extraction process for each land use category and mode of travel and determine whether the number of records provides a reasonable base from which to calculate relative mode shares. In the development of this process, the research team did not attempt to ensure that the results contained a statistically sig- nificant number of records. Table 4.2 summarizes the number of records resulting after each step of the extraction process. 4.7.6 Estimated Mode Share by Land Use The trip records for each LUC were aggregated by mode to determine mode-share percentages for transit (rail and bus) and walk/bicycle. The values in Table 4.3 are the p.m. peak hour adjustment factors derived for GU/UC contexts in proximity to rail stations and high-frequency bus route stops. 4.7.7 Estimated Vehicle Occupancy Adjustment Factors In addition to mode share, Table 4.3 presents the vehicle occupancy adjustment factors derived from the 2000 BATS data, for use in converting between infill person trips by auto- mobile and infill vehicle trips. Infill Adjustment Factors for GU/UC Contexts Within Walking Distance Of: High- Frequency Bus Stop Rail Station a.m. p.m. a.m. p.m. Multifamily Residential (ITE LUC 223) Transit 20.2% 17.5% 19.3% 16.2% Walk/bicycle 13.4% 13.3% 13.2% 13.7% Vehicle occupancy 1.61 1.60 1.58 1.61 General Office (ITE LUC 710) Transit 23.6% 22.4% 20.6% 20.6% Walk/bicycle 8.4% 8.7% 9.1% 9.4% Vehicle occupancy 1.36 1.27 1.35 1.27 Retail/Shopping Center (ITE LUC 820) Transit 12.7% 10.7% 13.1% 11.7% Walk/bicycle 11.4% 15.3% 12.3% 16.3% Vehicle occupancy 1.50 1.49 1.55 1.53 Quality (Sit-Down) Restaurant (ITE LUC 932) Transit 26.7% 14.3% 25.3% 15.5% Walk/bicycle 20.8% 16.6% 20.6% 19.8% Vehicle occupancy 1.37 2.07 1.44 2.13 Source: Mode share and vehicle occupancy adjustment factors were extracted from linked-trip data records developed from the 2000 Bay Area Travel Survey, Metropolitan Transportation Commission. Table 4.3. Example mode share and vehicle occupancy adjustment factors for the San Francisco Bay Area.