Reverse-Engineered Land Use Patterns to Minimize ...

Bose and Fricker

1

“Reverse-Engineered” Land Use Patterns to Minimize Congestion Amica Bose PB Transit & Rail Systems Inc. (A Parsons Brinckerhoff Company) 230 W. Monroe Street, Suite 1575 Chicago, IL 60606-4701 Phone: (312) 803-6641 FAX: (312) 803-0035 Email: [email protected] Corresponding Author: Jon D. Fricker School of Civil Engineering Purdue University West Lafayette, IN 47907-1284 Phone: (765) 494-2205 FAX: (765) 496-1105 Email: [email protected]

Submittal Date: 29 July 2002 Resubmitted: 13 November 2002

Abstract + Text + References = 4234 words (plus 7 tables and 5 figures, equivalent to 7234 words) Abstract A neighborhood land use pattern designed to accommodate the most frequently taken non-work trips within an acceptable distance from home was developed. Rather than start from a specified set of land uses and study the travel characteristics, mixed land use patterns that fit the observed tripmaking behavior of people were formulated. The result was called a Reverse Engineered Neighborhood, or REN. The REN was tested against a “Euclidean” development that had separated land uses. Results show a substantial reduction in non-work trip lengths (in terms of both travel time and distance) in the REN when compared to the Euclidean development. The efficiency of the REN is the result of having more trip destination choices available to residents at acceptable distances. This paper describes and demonstrates the procedures in the analysis, presents the results of the analysis, and suggests directions for further study.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

2

INTRODUCTION As traffic congestion continues to worsen in many US urban areas, the connection between land use patterns and travel patterns is being looked at with greater interest. Arguments for “smart growth” abound. Mixed land use is gaining advocates, but what form should mixed land use take? New Urbanists seem to begin with a concept based on the neighborhood our (grand)parents grew up in, then offer a community that we “should” live in, if we want to reduce dependence on motor vehicles and create a more livable environment. Transit-Oriented Design is, as the name implies, based on the presumption that residents will rely on transit as their preferred mode. These, and other, alternatives to the common suburban design with separated land uses and cul de sacs may have merit. However, in each case, the starting point seems to be a proposed land use pattern that is intended to engender a desired result. The expectation is that travelers will change their tripmaking habits in response to the new neighborhood design. In the research reported in this paper (1), the starting point for neighborhood design is the current tripmaking behavior of individuals, as indicated by several national databases. Instead of hoping that persons will change their tripmaking behavior because of a particular design concept, why not design a neighborhood around typical travel behavior? The steps leading to the design of such a neighborhood are described below. The transportation attributes of this neighborhood are compared to those of a conventional suburban development. Conclusions are drawn regarding how well these two neighborhoods perform with respect to their different land use patterns.

MODEL DEVELOPMENT The central concept is, rather than start from a defined set of land uses and study the resulting travel characteristics, instead shape land use patterns to fit the travel patterns of people. The methodology followed is what one could term “reverse engineering”. The intent is to start with existing trip making behavior and identify land use configurations that would satisfy the most common trip purposes. The resulting “Reverse Engineered Neighborhood” (REN) is designed to accommodate the most frequent non-work trips within an acceptable travel distance from home. It is presumed that persons will seek employment anywhere in the urban area, and not be limited to jobs that may be available within the REN. Likewise, REN residents will not be restricted to the REN when making trips for any purpose. The idea is to provide opportunities for destinations that satisfy most frequent trip purposes within the REN. The REN design is in contrast to a “Euclidean” neighborhood design. The Euclidean neighborhood is similar to a conventional suburban development, with residential and non-residential uses separated, and featuring low densities rather than compactness. The analysis of any given neighborhood design is performed using an appropriately modified version of the standard four-step travel demand modeling procedure. In this study, TransCAD software was used. The neighborhood design being analyzed will not exist in isolation; it will be embedded in an urban area of about 45 square miles. TransCAD’s Tutorial directory includes a representation of UTOWN. Because UTOWN has often been used to demonstrate and test software, it will be the “study area” referred to in this study. The TransCAD version of UTOWN has 5 zones. (See Figure 1.) Any neighborhood to be analyzed will be inserted in a corner (shaded in Figure 1) of the south-central zone. The following subsections describe the procedures for development of a “module”, one or more of which would constitute a REN. Trip Frequency and Land Use Type To invoke the Reverse Engineering methodology, two aspects of trip making behavior have been considered: 1) trip frequency and 2) trip rates. The frequency of trip making by trip purpose for a typical household gives an idea of the type of land uses that should be present in the REN. The trip rates to specified land use types can be used to estimate the area that will be needed by the chosen land uses. In order to address the trip frequency issue, a list of the most commonly visited land uses was made from the Yellow Pages categories in the telephone directory. A subjective estimation of the frequency of trip making by average households to these land uses was made. The levels of trip frequency considered were: • 1 trip / week • 1 – 4 trips / month • < 1 trip / month

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

3

Each land use type was placed into one of these trip frequency categories. Table 1 lists the land uses that were considered. The idea was to generate a comprehensive list of land uses that would fulfill most of the non-work trip needs of the people living in the REN. The land uses were further grouped into the broad major categories compatible with the Nationwide Personal Transportation Survey (2) trip making categories for which data are available. Trip Rates and Land Use Areas While frequency of trip making to/from households was used to obtain the type of land uses, trip rates at the nonhome end were used to determine the number and size of each of those chosen land use types. In turn, the total area to be devoted to each land use in the module could be estimated. Trip rates for the 11 major categories used in the NPTS were used as a starting point. The annual trip making data were converted to weekly trip rates, which were then used in conjunction with the trip rates as obtained from ITE Trip Generation Report (3). The following steps explain the procedure used. Step 1 Trip Rates from ITE Trip rates for the 11 major NPTS categories were found in the ITE report, using those pages where Gross Floor Area (GFA) was the independent variable. Step 2 Conversion to Weekly Trip Rates Annual vehicle trip frequencies per household from NPTS (2) data were converted to weekly trip frequencies. Weekly frequencies account for the variation in trip making behavior between weekends and weekdays. The ITE trip rates were converted to trip rates per 1000 sq. ft. GFA per week to be compatible with the NPTS trip data. Where trip rates for both weekdays and weekends were available, the conversion to weekly trips used a weighted sum of the weekday and weekend (Saturday and Sunday) trips. Step 3 Sample Area Calculation The area requirement for each land use was determined by considering the relationship between the number of weekly trips per household for each category (from NPTS) and the trip rates per 1000 sq. ft. GFA from ITE. In the sample calculation below, the land use category “shopping center” (code #820, ITE Report (3)) is used as one example of the trip purpose “shopping”. From Table 2: Trip rate for shopping center = 289.82 trips/week/1000 sq. ft. GFA (The weekly trip rate is a weighted sum of the rates for weekdays, Saturday, and Sunday.) NPTS trip frequency for shopping = 9.63 trips/week/HH Area required for shopping center = (1000/289.82)* 9.63 = 33.24 sq. ft./HH Area required for a shopping center to support 100 households = 33.24 * 100 = 3324 sq. ft./100 HH The calculation shown above is repeated for all the NPTS categories of trip making. Note in Table 1 that “shopping center” is just one example of the general category “shopping”. A calculation similar to the one above must be carried out for the other kinds of shopping land uses listed in Table 1. Step 4 Distinguishing Work and Non-Work Trips Steps 1 through 3 were repeated for all the other trip purpose categories. In many cases, as in “Other social and recreational” and “Other family business”, more than one land use type was chosen (see Table 1), and a single weighted ITE trip rate for that trip purpose was arrived at. The NPTS trip purpose “To and from work” has not been included in these area calculations, because workplaces such as manufacturing plants and office parks were not thought to be an appropriate use of land in a neighborhood intended to reduce trip lengths for non-work trip purposes. Furthermore, there is no guarantee that such employment centers would have a significant number of their employees living in the REN. Most employment opportunities will remain outside the neighborhood elsewhere in the study area. The non-residential uses in Table 1 will generate some employment that will be captured in trip generation attraction equations, but the focus in the

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

4

REN design is the tripmaking behavior of the neighborhood’s households. Consequently, small offices compatible with a residential neighborhood, like banks and law offices, have been included. Step 5 Estimate Non-Residential Area per 100 Households The sq. ft. area requirements associated with 100 households in the module (see Table 2) were added up to obtain the total non-residential floor area requirement. The total area requirement for 100 households can be used as a basis for estimating the number of households (and hence the population) required to support a specified amount of non-residential land uses. Determination of the Number of Land Uses In Table 2, the floor area of non-residential land uses required to support 100 HH units has been calculated. The term “non-residential” for this analysis means commercial and retail land uses, educational institutions (schools), churches, and public service buildings (post office, public library). It also refers to the small-scale workplaces that fit in a neighborhood setting, like real estate and insurance agencies. In this section, the methodology for obtaining the number of units for each land use type will be presented. Step 1 Area Required for Each Land Use Type The types of land use were selected from Table 1, based on the 11 major categories as formulated by NPTS. The average value of the ‘X’ variable (as obtained from the ITE report (3)) gave the floor area requirement for one unit of each land use. However, for most land use types, the average floor area in the ITE database was found to be much too high for a neighborhood like the REN, which seeks to have compact land uses. In such cases, the floor areas of the non-residential land uses were based on examples appropriate to the module drawn from the local community. In addition, land uses like copying center, florist, barber shop, for which trip making areas were not available in the ITE Report (3), were estimated from local examples. Step 2 Number of Units of Each Land Use Type Two variables are of importance here: the number of units of non-residential land uses and the number of household (HH) units. There is interdependency between the number of HH units and a particular non-residential land use type. If the number of units of each land use type is held constant, then the number of HH units required to support that land use can be determined. For example, from Table 3: Number of supermarkets = 2 Number of clothing stores = 1 Adding the other land use categories under “shopping”, the total floor area required = 32056 sq. ft. From Table 2: Total floor area required for “shopping” per 100 HH per week = 3324.46 sq. ft. # of HH required to support a total shopping floor area of 32056 sq. ft. = (32056/3324.46)*100 = 964 HH units. The number of units for each category (e.g., 2 supermarkets and 1 clothing store above) was established based on the analyst’s judgment, based on the “floor area per 100 HH” value calculated earlier. In this step’s example, the total floor area requirement for shopping indicated the number of HH units that would be required to support the land use category “shopping”. Computing the number of HH units also makes it possible to estimate the population and the area requirement for residences in the module. Step 3 Adjustments for Multi-Story Buildings Adjustments were made to account for the space requirements of multi-storied buildings. Floor areas for multi-story buildings were converted to land areas using a Floor Area Ratio (floor area/land area) of 1.0. For example, a building with 5000 sq. ft. of floor space on two or more stories would be placed on a 5000 sq. ft. plot of land. Step 4 Number of Households Required in a Module The floor area required to support 100 HHs (excluding workplaces) has already been calculated in Table 2. In the “reverse engineering” methodology, the number of HHs required to support the total floor area (for non-residential uses, excluding open spaces) of a module can be obtained as shown below: Total floor area for trips, excluding open spaces (based on Table 3) = 32056 + 38649 + 69000 + 44320 + 3036 + 11000*4 (# of mixed-use buildings) = 231061 sq. ft.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

5

Floor area per 100 HH for non-residential uses, excluding workplaces (see Table 2) = 32380.76 – 24175.07 = 8205.69 sq. ft. # of HH required to support a total non-residential floor area of 231061 sq. ft. = (231061 /8205.69)* 100 = 2816 HH units. Therefore, the number of HHs required in one module is 2816. Step 5 Population in a Module The population required to support the non-residential land uses is obtained by assuming an average of 2.5 persons per HH (2). Population = 2816 * 2.5 = 7040 Step 6 Area Required for Housing Units A density of seven HHs per residential acre was used. This means that the average lot size for a housing unit would be 1/7 acre. To the extent that apartment houses are used, the average lot size for the remaining HHs would be larger than 1/7 acre. Using 7 HH units per acre, housing land area required for 2816 HH = 2816*1/7 = 402.27 acres = 0.6275 sq. mi. This figure was found to be in conformity with the area requirements as prescribed in the New Urbanist Lexicon (4). Step 7 Streets Twenty percent of the total land area for residential and non-residential purposes was set aside for the street network (5). Step 8 Total Area of One Module Residential area + Non-residential area + Streets = Area of 1 module Total plan area for non-residential (except mixed use buildings) = 180860 sq. ft. This is the sum of the “Total plan area” values in Table 3. Assuming 50% lot coverage (to account for setbacks and green space, if any), the land area for non-residential land uses = 180860*2 = 361720 sq. ft. For mixed-use buildings, the land area needed is computed using a floor-area ratio (FAR) of 1.00: At the bottom of Table 3, the plan area of a 4-story mixed use building is set at 11000 sq ft. Therefore, the land area is 4 stories * 11000 sq ft/story * 1.00 FAR = 44000 sq. ft. Total land area for all non-residential land uses = 361720 + 44000 = 405720 sq. ft. Space for parking: Consider 1 space per 300 sq. ft. of GFA, for non-residential uses (Zoning Digest (6)) 1 space = 18.5 feet x 9 feet (INDOT (7), Highway Research Board (8)) Parking area required for all non-residential land uses = (231061/300)*18.5*9 = 128205 sq. ft. Total land area for non-residential uses + parking space = 405720 + 128205 = 533925 sq. ft. = 0.0190 sq. mi. Total land area for residential and non-residential land uses = 0.6275 + 0.0190 = 0.6465 sq. mi. Adding 20 percent of this area for streets, total land area = 1.2*0.6465 = 0.7758 sq. mi. = 0.88 * 0.88 mile square The land area for one module was calculated to be roughly 0.88 x 0.88 mile square. In order to provide ample open space, the total size of the module was rounded up to 1.0 x 1.0 mile square. Of the total land area, more than 20 percent (128 acres) was available for open space, which could be in the form of parks, ponds, playgrounds, sports fields, neighborhood commons, etc. This exceeds the Urban Land Institute (5) guideline of 15 acres per 1000 population. Step 9 Area of REN The REN neighborhood prototype was formed out of four similar modules, each 1.0 by 1.0 mile square. Although one square mile is a good size for a neighborhood (New Urban News (9), Kunstler (10)), it may not be large enough to allow this analysis to detect the transportation impacts of a particular mixed land use pattern within the 45 sq. mi. UTOWN study area. In addition, a four-module REN permits inclusion within the REN land use types that are visited with moderate frequency by a typical household. Instead of being excluded from a one-module REN, these land use types can be included in one or two of the four modules in the REN. Moreover, a four-module REN provides the opportunity to explore how the edges (Alexander (11), Kunstler (10)) of each module can interface with

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

6

(a) the adjacent modules and (b) the major road that cuts through the REN. Thus, the total size of the REN is 2.0 by 2.0 miles square. ANALYSIS In preparation for the standard four-step modeling process, the REN was converted to the appropriate input file format. A Euclidean neighborhood, called “EUCLID”, was also prepared. EUCLID has the same size and zone structure as the REN. Like the REN, EUCLID has 352 zones to allow for a detailed portrayal of land use patterns. However, EUCLID is predominantly a residential neighborhood, with no retail and non-retail establishments. The same production and attraction equations (12) and the same friction factor functions were used for both neighborhoods. Both the REN and EUCLID were tested using two scenarios: 1. The entire study area network, consisting of REN (or EUCLID) and the five adjacent zones of UTOWN. 2. The REN (or EUCLID) neighborhood as a subarea.

RESULTS The UTOWN study area and the REN (or EUCLID) subarea were analyzed using suitable measures of effectiveness for home-based work, home-based non-work, and non-home based trip purposes. Measures of Effectiveness (MOEs) A comparison of the average trip lengths by trip purpose was made. Although the UTOWN study area population remains the same for the two cases, the type and number of land uses vary within the REN and the EUCLID neighborhoods. This has a bearing on the trips produced and hence on the travel times and distances. Table 4 gives a comparison of the average trip lengths using loaded travel times (in minutes) for REN and EUCLID, considering all 357 zones in the UTOWN study area. Table 4 shows the percent decrease in the trip lengths for the UTOWN study area with REN when compared to UTOWN study area with EUCLID. The decrease is minimal in case of the HBW trips, because most work trips would still go outside the REN. The EUCLID neighborhood has no HBW attractions, because it has no workplaces, while the REN has a few. The decrease in trip lengths for the UTOWN study area with REN is greatest in the case of HBNW trips. This is because many frequently-visited non-residential land uses are located within the REN, while the EUCLID neighborhood has few non-residential land uses. Because most of the trips in UTOWN do not have either end in the REN/EUCLID subarea, the impacts of the subarea neighborhood design are muted in the UTOWN-level results shown in Table 4. To focus on the changes in tripmaking by REN/EUCLID residents, a subarea analysis was performed. Table 5 gives a comparison of the trip lengths (in minutes) for trips having at least one end within the REN or EUCLID. In this case, the decrease is minimal in the case of HBW trips, because most work trips would go outside the REN. Comparing Tables 4 and 5, it can be observed that the percentage reductions in the average trip lengths are much greater when only the trips that originate or terminate within the REN or EUCLID are considered. Tables 6 and 7 give a comparison between the average trip lengths (in miles) for the regional models and the subareas respectively. Trip lengths are substantially shorter for HBNW and NHB trips for the REN subarea. There are also modest reductions for HBW trips in the REN case. Figures 2 and 3 show the Trip Length Distributions (TLDs) in terms of distance (miles) for HBW, HBNW and NHB trips. More than 25 percent of HBNW trips are less than a mile long for the REN case, compared to around 18 percent for EUCLID. That EUCLID has any HBNW trips at all is because (a) the HBNW attraction equation that was used included dwelling units as an independent variable and (b) it is possible to find attractions in the zones close to EUCLID. The TLDs also show that highest percent of trips for the HBNW trip purpose is found in the range 0-1 mile for the REN (Figure 2) compared to 3-4 miles for EUCLID (Figure 3). Figures 4 and 5 give the TLD plot for trips that have at least one end inside the REN and EUCLID neighborhoods, respectively. In Figure 4, more than 70 percent of HBNW and NHB trips are less than two-thirds of

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

7

a mile long. Because the trips in this analysis include only trips by motor vehicle, there is potential for a shift to non-motorized modes (like walking or biking) for these very short trips. For trips in and out of EUCLID (see Figure 5), it is again true that a substantial percentage of HBNW and NHB trips (approximately 50-60%) are less than two-thirds of a mile long. However, there are no trips less than two-thirds of a mile for the HBW trip purpose. This plot suggests, as does Figure 4, that a shift from auto to bike and walk modes for these shorter trips is possible. CONCLUSIONS The comparative study of the REN and EUCLID neighborhoods indicate that, in terms of average trip length, the REN is superior. This suggests that allocation of land uses and their type have a definite bearing on trip lengths and hence on the congestion in a network. Residents of the REN are able to choose from a wide range of non-residential land uses that are within the neighborhood, thereby reducing their trip lengths both in terms of distance and time. In this study, the characteristic of primary concern is the land use in the REN and EUCLID. Possible differences in the street design characteristics, like lane widths, parking, and characteristics that relate to the network have not yet been incorporated in the travel demand model. The results of this study show that there is a large percentage of short trips (above 75% for REN and 5060% for EUCLID) in the range of 0-0.67 miles. These findings can be used to investigate the possibility of converting these fairly short auto trips to non-motorized modes, such as walk and bike. The study described in this paper will extend its analysis to faithful representations of New Urbanist neighborhoods and Transit-Oriented Design, while refining the REN design to reflect more detailed data and alternate ways to represent land use patterns that are “reverse engineered.” ACKNOWLEDGMENTS This work was supported by the Joint Transportation Research Program administered by the Indiana Department of Transportation and Purdue University. The contents of this paper reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the Federal Highway Administration and the Indiana Department of Transportation, nor do the contents constitute a standard, specification, or regulation. The authors are particularly grateful to Paul Ricotta and Jim Lam of Caliper Corporation for their assistance in using the TransCAD software. REFERENCES 1.

Bose, Amica. Alternative Land Use Patterns to Minimize Congestion, Masters Thesis, School of Civil Engineering, Purdue University, West Lafayette, Indiana. August 2002. 2. Bureau of Transportation Statistics. Summary of Travel Trends: Nationwide Personal Transportation Survey Report. Washington D.C.1995. http://www.bts.gov/ntda/npts. Accessed January 2002. 3. Institute of Transportation Engineers (ITE). Trip Generation. 6th Edition, Volumes 1 of 3 and 2 of 3, Washington D.C. 1997. 4. McLaughlin, Richard. A New Urbanist Lexicon: Part 5. Planning Minnesota Online. 1996. http://www.mnapa.com/urbanlex.html. Accessed June 2001. 5. Urban Land Institute. The Community Builders Handbook. Anniversary Edition, Washington D.C., 1977. 6. City of Alexandria, Virginia. Zoning Digest: Parking and Loading Requirements.2002. http://ci.alexandria.va.us/city/zoningdigest/zd_plreqs.html. Accessed May 2002. 7. Indiana Department of Transportation (INDOT). Standard Drawings. 1999. 8. Highway Research Board. Parking Principles: Special Report 125, Washington D.C. 1971. 9. New Urban News. New Urbanism and Traditional Neighborhood Development. Comprehensive Report and Best Practices Guide. Ithaca, NY. 2000. 10. Kunstler, J. H. Home from Nowhere. Atlantic Monthly, pp. 43-66, September 1996. 11. Alexander, C. S., Ishikawa, M., Silverstein. A Pattern Language. Oxford University Press. 1977.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

8

12. Sosslau, Arthur B., Amin B. Hassam, Maurice M. Carter, and George V. Wickstrom. Quick-Response Urban Travel Estimation Techniques and Transferable Parameters – User’s Guide. NCHRP Report 187, Comsis Corporation. 1978.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

9

LIST OF TABLES TABLE 1 Land Use Types by their Frequencies TABLE 2 Representation of Area Calculations Based on Major Trip Purposes TABLE 3 Detailed Calculations for Non-Residential Area Requirement TABLE 4 Comparison of Average Trip Lengths for the UTOWN Study Area with REN/EUCLID TABLE 5 Comparison of Average Trip Lengths for Trips In and Out of REN/EUCLID TABLE 6 Comparison of Average Trip Lengths by Distance for UTOWN Study Area TABLE 7 Comparison of Average Trip Lengths by Distance for Trips In and Out of REN/EUCLID LIST OF FIGURES FIGURE 1 UTOWN Study area with REN FIGURE 2 UTOWN Study Area TLD (by Distance) with REN FIGURE 3 UTOWN Study Area TLD (by Distance) with EUCLID FIGURE 4 TLD (by Distance) for Trips In and Out of REN FIGURE 5 TLD (by distance) for Trips In and Out of EUCLID

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

10

TABLE 1 Land Use Types by their Frequencies NPTS Category

Land Use

Trip Rate (per HH)

Shopping

Grocery (Supermarket) Convenience Store Clothing Hardware store Shopping center and Mall Specialty retail center Bakery Furniture Store Electronics Store Sporting Goods- retail

> 1 trip/week > 1 trip /week < 1 trip/month < 1 trip/month 1-4 trips/month 1-4 trips/month > 1 trip/week < 1 trip/month < 1trip/month < 1 trip/month

Doctor/Dentist

Doctor Clinic Hospital Vet Clinic

< 1 trip/month < 1 trip/month < 1 trip/month < 1 trip/month

Other Family Business

Post Office Banks Credit Union Gas Station Car Wash Auto repair center Drugstore Copying services Books & Magazine Center Florists Barber Shop Beauty Salon Attorney Accountant Engineering Consultant Insurance Office Real Estate Dry Cleaners Commercial Washers, Dryers Plumbing repair and service

1-4 trips/month 1-4 trips/month 1-4 trips/month > 1 trip/week 1 trip/week > 1 trip/week 1-4 trips/month 1-4 trips/month 1-4 trips/month < 1 trip/month 1-4 trips/month 1-4 trips/month 1-4 trips/month

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

11

TABLE 2 Representation of Area Calculations Based on Major Trip Purposes Trip Purpose in NPTS

Trip/year/HH1 Trip/day/HH

Trip/week ITE trip /HH rate/week

Sq. ft./100 HH/week

Earning a Living 1. To or From Work 2. Work related business

553.00 80.00

2.13 0.31

10.63 1.54

44.00 -

24175.07 -

Family and Personal Business 3. Shopping 4. Doctor/Dentist 5. Other family business

501.00 33.00 626.00

1.37 0.13 1.72

9.63 0.63 12.04

289.82 157.25 831.62

3324.46 403.57 1447.60

6. School and Church

98.00

0.38

1.88

77.97

2417.26

7. Vacation 8. Visiting friends and family 9. Other social and recreational

2.00 155.00 269.00

0.01 0.42 0.74

0.04 2.98 5.17

844.18

612.79

2.00 0.00

-

0.04 -

-

-

Miscellaneous 10. Other 11. Purpose not reported

Total 2321.00 1 Summary of Travel Trends, 1995 Nationwide Personal Transportation Survey

TRB 2003 Annual Meeting CD-ROM

32380.76

Paper revised from original submittal.

Bose and Fricker

12

TABLE 3 Detailed Calculations for Non-Residential Area Requirement Trip Purpose/ Land Use Type Shopping Supermarket Convenience Store Clothing Hardware store Total Other Family Business Post Office Banks Gas station (w/ convenience mkt.) Drugstore Copying Center/Office Supply Books and Magazine Center Florist Barber Shop Beauty Salon Plumbing Repair and Service Commercial Washers and Dryers Dry Cleaners Total Schools and Church Elementary school Day Care Center Church Total Other Social and Recreational Sit-down restaurant Fast-food restaurant Drive-in restaurant Coffee Shop Ice-cream and confectionary Pizza Place Fitness center (2-story building) Public Libraries Movie theater Video Rental Community Center (2-story building) Total Doctor/Clinic (4-story building) Total plan area Area for mixed-use buildings

TRB 2003 Annual Meeting CD-ROM

Unit Area (sq. ft.)

# Units

Plan area (sq. ft.)

12864 1675 1749 1155 17443

2 1 2 1

25728 1675 3498 1155 32056

4524 3610 1000 925 700 700 754 700 800 500 1200 1100 16413

1 2 3 2 1 0 0 1 1 1 2 2

4524 7220 3000 1850 700 0 0 700 800 500 2400 2200 38649

47000 4000 18000 69000

1 1 1

47000 4000 18000 69000

2500 2500 1500 800 500 500 7402 6363 7055 2700 5000 36820 3036

3 2 1 1 1 1 1 1 1 1 1

7500 5000 1500 800 500 500 3701 6363 7055 2700 2500 38119 3036 180860 11000

1

Paper revised from original submittal.

Bose and Fricker

13

TABLE 4 Comparison of Average Trip Lengths for the UTOWN Study Area with REN/EUCLID Average trip length (mins)

REN + 5 zones

EUCLID + 5 zones

% Change w.r.t. EUCLID

HBW

12.403

12.677

- 2.2

HBNW

7.818

8.573

- 8.8

NHB

10.179

10.567

- 3.7

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

14

TABLE 5 Comparison of Average Trip Lengths for Trips In and Out of REN/EUCLID Trips in and out of REN

Trips in and out of EUCLID

% Change w.r.t. EUCLID

HBW

6.322

7.455

-15.2

HBNW

2.401

3.047

-21.2

NHB

2.273

3.862

-41.1

Average trip length (mins)

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

15

TABLE 6 Comparison of Average Trip Lengths by Distance for UTOWN Study Area Average trip length (miles)

REN + 5 zones

EUCLID + 5 zones

% Change w.r.t. EUCLID

HBW

4.601

4.712

-2.4

HBNW

2.933

3.233

-9.3

NHB

3.790

3.949

-4.0

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

16

TABLE 7 Comparison of Average Trip Lengths by Distance for Trips In and Out of REN/EUCLID Trips in and out of REN

Trips in and out of EUCLID

% Change w.r.t. EUCLID

HBW

2.292

2.407

-4.8

HBNW

0.909

1.012

-10.2

NHB

0.763

1.278

-40.3

Average trip length (miles)

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

17

FIGURE 1 UTOWN Study area with REN (shaded block) Inserted in a Corner of a Zone

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

18

Study Area TLD (by Distance) with REN 30 25

Percent

20 15 10 5

8.41-

7.48-8.41

6.56-7.48

5.63-6.56

4.70-5.63

3.78-4.70

2.85-3.78

1.92-2.85

1.00-1.92

0-1.00

0

Distance (miles)

HBW HBNW NHB

FIGURE 2 UTOWN Study Area TLD (by Distance) with REN

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

19

Study Area TLD (by Distance) with EUCLID 25

Percent

20 15 10 5

8.41-

7.48-8.41

6.56-7.48

5.63-6.56

4.70-5.63

3.78-4.70

2.85-3.78

1.92-2.85

1.00-1.92

0-1.00

0

HBW HBNW

Distance (miles)

NHB

FIGURE 3 UTOWN Study Area TLD (by Distance) with EUCLID

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

20

TLD (by distance) for Trips In and Out of REN 80 70

Percent

60 50 40 30 20 10 6.03-

5.36-6.03

4.69-5.36

4.02-4.69

3.35-4.02

2.68-3.35

2.01-2.68

1.34-2.01

0.67-1.34

0-0.67

0

HBW HBNW

Distance (miles)

NHB

FIGURE 4 TLD (by Distance) for Trips In and Out of REN

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Bose and Fricker

21

TLD (by distance) for Trips In and Out of EUCLID 70 60

Percent

50 40 30 20 10

Distance (miles)

6.03-

5.36-6.03

4.69-5.36

4.02-4.69

3.35-4.02

2.68-3.35

2.01-2.68

1.34-2.01

0.67-1.34

0-0.67

0

HBW HBNW NHB

FIGURE 5 TLD (by distance) for Trips In and Out of EUCLID

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.