|
Nationwide Personal Transportation Survey - Indianapolis Metro Area Special Add-on Study
Click HERE for graphic. NATIONWIDE PERSONAL TRANSPORTATION SURVEY INDIANAPOLIS METROPOLITAN AREA SPECIAL ADD-ON STUDY EXECUTIVE SUMMARY What is the Nationwide Personal Transportation Survey (NPTS)? The NPTS is a nationwide survey of trips and travel conducted at approximately 7-year intervals. The 1990 NPTS survey used a computer-assisted telephone interviewing (CATI) technique based on a random digit dialing (RDD) sample to reach residential phone numbers. It was conducted by a private contractor Research Triangle Institute (M, located in Research Triangle Park m the Raleigh-Durham area of North Carolina. Why Special Add-On Interviews in Indianapolis? The City of Indianapolis contracted with the RTI in 1990 to provide sample interviews of 989 selected households. The following factors contributed to the City's decision to purchase the special add-on interviews in Indianapolis to coincide with the 1990 decennial Census: In 1964, a home-interview survey of 10,564 households was conducted by the Indianapolis Regional Transportation and Development Study (IRTADS). This is the only comprehensive travel survey in Indianapolis. The last major update of the Indianapolis travel demand forecasting model was completed in 1987. The Indianapolis Urbanized Area has grown from a population of 743,155 in 1970 to 914,761 in 1990 - an increase of 23%. The current forecast year of 2005 needs to be advanced to Year 2010 or 2020 to meet federal requirements of the 20-year planning horizon period for long range planning. Recent passage of the Intermodal Surface Transportation Efficiency Act (ISTEA) of 1991 legislation put additional emphasis on the metropolitan planning for intermodal transportation flexibility and integration with other land use and air quality objectives. What is the Purpose of this Report? The purpose of this report is to summarize the 1990 Indianapolis Special Add-on NPTS results. It provides the survey methodology and the background information about the survey respondents. It also presents national trends in personal travel compared with local travel characteristics in 1990. It concludes with limitations and uses of the data. Definition of Trips in the NPTS The 1990 Indianapolis Special Add-on survey collected the general socioeconomic household information in addition to information on all trips, by all modes for all household members age 5 and older. Information was collected for both "travel day" and "travel period" trips. The travel day is a designated 24-hour period, usually the previous day which can be any day of the week, for which the respondents' movements throughout the day were reported. A travel day trip is defined as any one-way travel from one address (place) to another by any S - 1 means of transportation (e.g., private motor vehicle, public transportation, bicycle, or walking) taken during the designated travel day. A travel period is the 13 days immediately preceding the travel day and the designated travel day for the sampled household, for a total of 14 days. Travel period trips are trips 75 miles are longer (each way) that occurred during the 14 day travel period. Comparison with the Census Data The 1990 NPTS is statistically representative of Indianapolis in every major aspects of demographic characteristics but underestimates the total Marion County households by 2% and population by 8% (Figure S-1). Comparison with National Travel Statistics The local NPTS compares reasonably well with the nationwide NPTS for a wide array of travel characteristics. Refer to Tables III-1 through HI-8 in the report for more detail. There were some differences that merit highlighting. The average vehicle trip length is 1.34 miles (or 15 %) shorter in the Indianapolis area than the nation's average. This shorter trip length is also reflected in a lower vehicle miles of travel per household. Locally, we travel 7.24 miles less per day per household than the nation as an average. Considering that the number of trips per household are comparable with the national average, we can conclude that the differences in daily vehicle miles of travel and average trip length are due to local residents living closer to where they work, shop and other trip purposes. The Indianapolis vehicle trips per household rate of 4.53 is three percent (3 %) below the national NPTS average trip rate of 4.68 trips per household. This and the shorter trip length then produce the shorter local Vehicle Miles of Travel (VMT) than the national average (Figure S-2). The local NPTS data compares favorably with the default values from the National Cooperative Highway Research Program (NCBRP) Report # 187: Quick-Response Urban Travel Estimation Techniques and Transferable Parameters that is a commonly used document for transferable default values for the traditional four-step transportation planning process. Figures S-3A through S-3C show the comparison with the NCBRP # 187 default values for the urban area similar to Indianapolis with population between 750,000 to 1,000,000. The person trip rates of 6.4 and 6.7 for the 1990 NPTS and the NCBRP # 187 respectively are not significantly different. Indianapolis is two percent (2%) smaller in term of the share of home-based work purpose trips. But, the auto occupancy rate of 1. 1 persons per vehicle is significantly lower than the NCBRP # 187 default value of 1.33 for homebased work trips. S - 2 Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Travel Demand Technical Attributes Indianapolis long range transportation planning uses the traditional four-step process for forecasting future travel demands. The following technical data from the local NPTS provides major inputs to this planning process: TRIP PURPOSES: Person or vehicle trips are stratified into different trip purposes to provide more reliable models. The Indianapolis travel demand model uses five major trip purposes: Home-Based Work, Shopping, School, Other and Non- Home-Based Other. The comparison of these five trip purposes between the 1990 NPTS and the Indianapolis Travel Survey of 1964 shows that the percent of home-based work trips has declined by 4%, from 27% in 1964 to 23% in 1990. This suggests that other facets of urban life such as family visits and other social or recreational activities are becoming more important relative to working for a living. TRIP GENERATION: A cross classification table of trip production rates was developed by COMSIS Corporation in 1987 to replace the multiple regression equations in the original 1964 model. The detailed comparison of trip rates by different household size and family income shows that there is a close correlation between 1990 NPTS data and the 1987 COMSIS trip rates for Home-based Work and Shopping trip purposes. There are minor differences of two (2) percent for Home-based Work Trips and one (1) percent for Homebased Shopping Trips. Differences for other three trip purposes are higher. MODAL SPLIT: Between 1964 IRTADS and the 1990 NPTS, Auto Driver Trips are up by 13 %, Auto Passenger trips are down by 19% and the Truck Driver trips are up by 5 %. In 1990, Ninety Five percent (95 %) of Indianapolis residents used private vehicles such as autos and trucks to commute to work. This is approximately four percent (4%) higher than the national average. Conversely, only 1.5 percent of Indianapolis residents used public transit for their daily work trips. The national average was 5.5 percent. Since there are only 45 transit trips in the 1990 NPTS data, any interpretation of public transit usage will not be reliable. AUTO OCCUPANCY: Auto Occupancy rates from the 1990 NPTS has been consistent with other recent studies of the Indianapolis rates. Further analyses reveal there is a strong correlation between the auto occupancy and the household size. Transferability of Technical Application The NPTS Indianapolis Special Add-on Survey provides the most comprehensive data on local travel characteristics for the Indianapolis area since the 1964 IRTADS. The local NPTS will be most useful in updating the Long Range Plan for the Indianapolis Metropolitan Planning Area as required by the Intermodal Surface Transportation Efficiency Act. S-6 With proper precautions, trip generation rates, trip length distribution for work trips and auto occupancy ratios for both home-based work trips and shopping trips can be transferred to other cities of comparable size and characteristics if primary data sources are not cost effective. In general, key local and national NPTS data are consistent and stable over years. It is advisable to use these transferred data in other cities with caution. The percent of homebased work trips and the auto occupancy rate, especially for the work trip purpose, in the NCHRP #187 needs to be revised immediately for the application of the Quick Response System. S - 7 Click HERE for graphic. TABLE OF CONTENTS EXECUTIVE SUMMARY......................................S - 1 I. INTRODUCTION...........................................1 II. INDIANAPOLIS HOUSEHOLD CHARACTERISTICS........................................3 III. COMPARISON WITH NATIONAL TRAVEL TRENDS..........................................12 IV. TRAVEL DEMAND FORECASTING MODEL TECHNICAL ATTRIBUTES...................................22 V. SUMMARY................................................37 I. INTRODUCTION What is the Nationwide Personal Transportation Survey (NPTS)? The NPTS is a nationwide survey of trips and travel conducted at approximately 7-year intervals. It is a telephone survey of 22,300 households throughout the United States. Previous surveys were conducted by the United States Department of Transportation in 1969, 1977, and 1983. In 1990, the NPTS was conducted from March 1990 to March 1991 to coincide with the decennial Census for collecting additional individual and household travel information to supplement the regular 1990 Census data. The Indianapolis add- on interviews began in June of 1990 and ended in March of 1991. The 1990 NPTS differs from earlier NPTS' in two ways. Previous surveys were home interviews conducted in person while the 1990 survey used a computer-assisted telephone interviewing (CATI) technique based on a random digit dialing (RDD) sample to reach residential phone numbers. Previous surveys were conducted by the Bureau of the Census while the 1990 NPTS was conducted by a private contractor, Research Triangle Institute (RTI), located in Research Triangle Park in the Raleigh-Durham area of North Carolina. Why Special Add-On Interviews in Indianapolis? Without the special add-on interviews, the national sample of the 1990 NPTS would have yielded a target sample size of 45 households in the Indianapolis Metropolitan area. This would not be statistically valid at the metropolitan area level to draw any conclusion on the validity of transportation planning models which were developed based on the 1964 survey of area residents. The City of Indianapolis, after reviewing other available alternatives, contracted with the RTI to add 944 interviews for a total sample size of 989 households. In the fall of 1964, a home-interview survey of 10,564 households was conducted by the Indianapolis Regional Transportation and Development Study (IRTADS). This is the only comprehensive travel survey in the Indianapolis Metropolitan area. With the exception of journey-to-work travel data collected during the 1970 and 1980 decennial Census, the Indianapolis travel demand models have had to rely on old information. While data collection activities have diminished, the demands placed on travel models in the long-range planning have gained in importance. For example, the models are used to determine traffic volumes for major highways, to negotiate with developers for transportation improvements associated with proposed developments, and to evaluate transportation control measures for meeting the City's air quality objectives. The last major update of the Indianapolis travel demand forecasting model was completed in 1987. The Indianapolis Urbanized area has grown from population of 743,155 in an area of 381 square miles to 914,761 in an area of 476 square miles from 1970 to 1990. The current forecast year of 2005 needs to be advanced to Year 2010 or 2020 to meet federal requirements of the 20-year planning horizon period for long range planning. Recent passage of the Intermodal Surface Transportation Efficiency Act USTEA) of 1991 legislation put additional emphasis on the metropolitan planning for intermodal transportation flexibility and integration with other land use and air quality coordination. This is the background and environment from which the City of Indianapolis decided to seize the economy of scale opportunity to supplement work trip data from the Census and the Census Transportation Planning Package (CTPP) with additional data for all trip purposes through the 1990 NPTS. The City is also planning to conduct external station surveys to acquire more recent data on trips having an origin or destination outside the study area. What is the Purpose of this Report? The purpose of this report is to summarize the 1990 Indianapolis Special Add-on NPTS results. It provides a general overview of the survey methodology and the general household, person and vehicle background information about the survey respondents. It also presents national trends in personal travel comparing with local travel characteristics in 1990. It concludes with limitations associated with the data and appropriate uses of the data. This report is structured into the following sections: I. INTRODUCTION II. INDIANAPOLIS HOUSEHOLD CHARACTERISTICS This section describes the survey instrument, the type of questions asked, the type of information obtained from the survey and defines key terms. It maps the number of respondents by zip code, compares population and number of household data with the 1990 Census, and identifies the limitations associated with the data. III. COMPARISON WITH NATIONAL TRAVEL TRENDS This section consists of a series of tables which present key information from the Indianapolis Special Add-on NPTS data files. National trends are shown and local statistics are compared with the national data. IV. TRAVEL DEMAND FORECASTING MODEL TECHNICAL ATTRIBUTES This section discusses the detailed analyses of the NPTS data in the context of the Indianapolis travel simulation model technical attributes for forecasting travel demand in the metropolitan area. V. SUMMARY This section summarizes conclusions which can be drawn from the survey results and identifies appropriate uses of the data for the Indianapolis transportation planning process and their transferability to other medium size (e.g. range of population 750,000 - 1,250,000) urbanized areas. 2 II. INDIANAPOLIS HOUSEHOLD CHARACTERISTICS The 1990 Indianapolis Add-on NPTS was a sample telephone survey conducted over a period of 9 months between June 1990 to March 1991 to insure that seasonal variations in travel were represented. The sample included a total of 989 households throughout the Indianapolis Urbanized Area. As shown on Map II-1 the vast majority (961) of the households interviewed were located in Marion County. This was a function of the need to randomly select the households based on their telephone prefix and the need to stay within the Indianapolis Urbanized Area. Definition of Trips The survey collected the general socio-economic household information in addition to information on afl trips, by all modes, for all household members age 5 and older. Information was collected for both "travel day" and "travel period" trips. The travel day is a designated 24-hour period, usually the previous day, for which the respondents' movements throughout the day were reported. A travel day trip is defined as any one-way travel from one address (place) to another by any means of transportation ( e.g., private motor vehicle, public transportation, bicycle, or walking) taken during the designated travel day. A travel period is the 13 days immediately preceding the travel day and the designated travel day for the sampled household, for a total of 14 days. Travel period trips are trips 75 miles or longer (each way) that occurred during the 14 day travel period. The sample was a random digit dialing sample which allowed for inclusion of unlisted telephone numbers. Attempts were made to contact and interview each household member age 14 and older. For persons age 5 through 13, an adult household member reports their trips. The person answering the questions about the whole household (e.g., household composition, income,vehicles available) as well as his/her own travel, was usually on the phone about 20 minutes. The interviews for other household members were about 10- 12 minutes each. Data Files The data from the survey was organized into six files showing the number of records in the Indianapolis Add-on survey: Household (989) Person (2,005) Vehicle (1,682) Travel Day trips (6,332) Travel Period Trips (430), and Segmented Trips (45). 3 The raw survey data then underwent a 10-step weighting procedure to reflect the sample design and selection probabilities, as well as adjustments to compensate for survey non-response and non-coverage. These weighting factors were necessary in order to expand the sample data to estimates of the total population. Indianapolis Household and Personal Characteristics Figures II-1 through II-5 show the profile of Indianapolis 1990 NPTS respondents for the following key socioeconomic variables: Household Annual Income Number of Vehicles by Household Distribution of Sex and Race, and Licensed Drivers by Sex and Age Group. Representiveness of Sample Data In order to determine how representative the 1990 NPTS data are, selected distributions were compared with external information from the U.S. Census Bureau. The nationwide data appears to underestimate the total population from the 1990 Census by 8% while underestimating the household total by 2%. Step number six of the weighting procedures adjusted the sample data to sum the March 1990 Current Population Survey (CPS) estimates of the number of U.S. households for: Census Region Household size MSA status Race (black, non-black) Ethnicity (Hispanic, Non-hispanic). The March 1990 CPS estimates were used because the Decennial Census data were not yet available at the time the final weights were assigned. At the local level the weighting procedures were not as effective. The differences between the local 1990 NPTS and the 1990 Census in terms of total population and total number of households are in the 20% to 30% range. For this reason, absolute unadjusted values in Table IEI-L from the 1990 Indianapolis Add-on NPTS, such as total number of households should not be used without further adjustment. One possible adjustment is suggested by the RTI staff and recalculated in Table 111-2. Basically, the Indianapolis Add-on sample involved sampling only nine months of the national survey year. To obtain better estimates for the area, the aggregated population and household estimates need to multiply the survey weights by a factor of 4/3, adjusting the totals to better estimate for the year. Estimates of ratios and percentages would not be affected, only the gross estimated totals. 4 Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. The adjusted NPTS data results in a better match with the available Marion County Census information. The adjusted differences are in the range of 2 to 8%. Based on how well the local NPTS compares with the nationwide NPTS it can be concluded that the local survey sample is representative of the local area when analyzing the data in relative terms such as trips per household or distribution frequencies. In general, the distribution of household size and age groups is statistically similar to that of the known regular Census data and the survey sample seems to reflect the population and household characteristics of the Indianapolis metropolitan area. Figures H-6 and H-7 show graphically the comparison of household size and age group. 10 Click HERE for graphic. Click HERE for graphic. III. COMPARISON WITH NATIONAL TRAVEL STATISTICS This section consists of a series of tables that present some of the more important travel characteristics information. each table contains a summary statement, national trends and a comparison of local and national NPTS data. Highlights of these tables follow: Indianapolis' average household size of 2.15 persons per household is lower than the national average of 2.45 by 19% (Table III-1). The average number of vehicles available per household (1.61) in Indianapolis is lower than that of the nation (1.77). In 1990, ten percent (10%) of Indianapolis households had no available vehicles (Table III-2). Indianapolis' average vehicle age of 7.4 years is slightly newer than the average vehicle age of the nation as a whole (Table III-7). Indianapolis' vehicle occupancy rate of 1.53 persons per vehicle is lower than the national average of 1.60 persons per vehicle. Indianapolis' current work trip vehicle occupancy rate of 1.1 persons per vehicle is identical to the national average (Table III-4). Indianapolis' annual vehicle trips per household of 1653 is slightly less than the national average 1702 by 2.9% (Table III-3). Indianapolis' journey to work commuting trip distance of 9.28 miles is shorter than the national average of 11 miles by 18.5%. For all purposes, the average vehicle trip length in Indianapolis is 7.53 miles which is 19.5% shorter comparing with the national average of 9.0 miles (Table III-3). Indianapolis' average annual Vehicle Miles of Travel (VMT) per household of 12,460 is less than the national annual average VMT of 15,100 by 21.2% (Table III-3). The primary mode of transportation to work in Indianapolis and the nation is the use of private vehicles (autos and trucks). The 1.5% public transit usage for work trips when respondents are asked about travel to work during the previous week in Indianapolis is significantly lower than the national average of 5.5%. Walking and bicycling which are included in the other mode of transportation category have higher percent of usage than public transit in Indianapolis (Table III-5). 12 13 Click HERE for graphic. 14 Click HERE for graphic. 15 Click HERE for graphic. 16 Click HERE for graphic. 17 Click HERE for graphic. 18 Click HERE for graphic. 19 Click HERE for graphic. 20 Click HERE for graphic. 21 IV. TRAVEL MODEL TECHNICAL ATTRIBUTES Indianapolis long range transportation planning uses the traditional four-step process for forecasting future travel demands. Figure IV-1 shows the generalized flowchart of this technical process. The Indianapolis special add-on study is intended to provide validity checks on the following key technical attributes which are major inputs and assumptions to this planning process: A. Trip Purposes B. Trip Generation Rates C. Average Trip Length D. Modal Split E. Auto Occupancy F. Other pertinent travel behavior statistics. TRIP PURPOSES Personal trips are stratified into different trip purposes to provide more reliable models since not all trips are homogeneous in character. The 1990 NPTS shows the mean trip miles for the following major trip purposes with their corresponding statistics (Figure IV-2): MEAN TRIP MILES BY PURPOSE Trip Purpose (Code) Mean Standard Deviation # of Cases Work(1) 9.08 9.62 1371 Shopping(3) 4.37 5.92 1121 Oth Family/Pers Bus(4) 7.09 29.08 1397 School/Church(5) 5.43 8.11 727 Visit Friends or Rel(8) 10.98 43.16 634 Oth Social & Rec(10) 9.15 19.60 831 Total 7.80 24.53 6332 It is apparent from these statistics that Work, Shopping and School are three major trip purposes with less deviation from the mean and hence more statistically stable for trip purpose stratification. nationally, the average annual vehicle trips per household for the trip purpose of going from home to work has declined from 31.9% of all trip purposes in 1969 to 26.3% in 1990 (Figure IV-3). The home-based work trips have become less dominant in 1990. The Indianapolis Regional Transportation and Development Study (IRTADS) travel demand forecasting model aggregates trips into the following five major trip purposes: 22 Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Home-based Work Home-based Shopping Home-based School Home-based Other Non-Home-based Other Figure IV-4 shows the comparison of these five trip purposes between the 1990 NPTS and the 1964 IRTADS. It is noted that the percent of home-based work trips has declined by 4%, from 27% in 1964 to 23% in 1990. The home-based school trips have grown approximately 7% between these two survey years probably due to the difference in the aggregation of trip purposes. TRIP GENERATION Trip Generation is the first analytical step that provides travel estimates based on land use and socioeconomic characteristics. Indianapolis initially used the multiple regression equations in the original 1964 IRTADS. A simplified trip production and attraction process was developed by COMSIS Corporation in 1986. The new trip production rates are based on the cross classification of household size (3) and family income (3) for all five (5) trip purposes. The trip production trip rates cross classification of these 3 by 3 data cells and their notations is presented as follows: Trip Production Cross Classification Household Size Household Income Low Medium High (1 person) (2-3 persons) (4-10 persons) Low (<25K) A B C Medium (25-55K) D E F High (>55K) G H I Work trips were calibrated with the 1980 Census Urban Transportation Planning Package information and other trip purposes were developed with synthesized data from other cities. Figures IV-5 through IV-9 show the close correlation between 1990 NPTS data and 1986 COMSIS for all 9 data cells of Home-based Work trips and Home-based Shopping trips. Two (2) percent difference for Home- based Work Trips and one (1) percent for Home-based Shopping Trips. Differences for other three trip purposes are higher: -41% for Home-based School, +22% for Non-home-based Others. Overall, the 26 Click HERE for graphic. Click HERE for graphic. AVERAGE TRIP LENGTH Average trip length (in minutes) in the Indianapolis Special Add-on Survey is 13.98 minutes. This compares favorably with the mean trip length of 14.07 minutes in the 1986 COMSIS data. Figure IV-10 shows the comparison of the mean auto travel time by trip purpose in graphic form. With the exception of the home-based Social trip purpose, the mean travel time between these two studies are very similar. For a closer look at the trip length frequency distribution curves, Figure IV- 11 through IV- 14 examine both the individual frequency percentage and the cumulative percentage of trip length frequency for all trip purposes as well as for the Home-based Work trip purpose. It is apparent that the 1990 NPTS data tend to group around the 5-minute intervals possibly because of the survey questionnaire and the telephone interview techniques employed by the RTI survey team. Looking at the cumulative curves, both work trip and all purpose are much similar to the 1964 IRTADS Origin and Destination data. The only discernible difference seems to indicate that NPITS has captured more short trips than IRTADS. MODAL SPLIT Modal Split is covered by the 1990 NPTS as the main means of transportation used for the previous day trip. Twenty one (21) categories were included in the survey ranging from auto, truck, motorcycle to bus, Amtrak, bicycle, walk and school bus. Since there are only 45 transit trips in the 1990 NPTS data, any interpretation of public transit usage will not be reliable. Figure IV-15 shows the comparison of general modal split information between the 1964 IRTADS and the 1990 NPTS. In general, Auto Driver Trips are up by 13 %, Auto Passenger trips are down by 19 % and the Truck Driver trips are up by 5 AUTO OCCUPANCY Auto Occupancy rates from the 1990 NPTS has been consistent with the Indianapolis earlier rates as demonstrated in Section III. Further analyses of the NPTS data reveal strong correlation between the auto occupancy and the household size. However, household family income has a lesser influence on the auto occupancy rate. The following is the average auto occupancy rates for these subgroups of household: AUTO OCCUPANCY BY HOUSEHOLD SIZE AND INCOME Low Household Size (1 person) 1.37 Medium Household Size (2-3 persons) 1.65 High Household Size (4-10 persons) 2.45 Low Family Incom (<25K) 2.06 Medium Family Income (25-55K) 1.96 High Family Income (>55K) 1.99 29 Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. OTHER TRANSPORTATION STATISTICS 1990 NPTS Indianapolis Special Add-on interviews produce many other valuable statistics on personal travel behaviors not available elsewhere. The following miscellaneous information captures some of the often used data for further reference. Carpool: Fifty-two percent (50%) of respondents reported to be driving alone, twenty-seven percent (27%) in two-person carpool, 10% in 3-person carpool, 8% in 4-person carpool and the remaining 3 % in a larger carpool. Parking Fees: Only fourth-tenth of a percent (.4%) reported to pay some kind of parking charges in Indianapolis. Peak Trip: Fifty-seven percent (57%) of Indianapolis person trips were reported to be taken during morning or afternoon peak rush hours. Accidents: Twenty-six (26%) of people interviewed by the RTI research firm reported having one accident in the last five years while four percent reported two accidents in the same period. Person-Miles-of-Travel (PMT) and Vehicle-Miles of Travel (VMT): The mean household PMT in Indianapolis is reported to be around 50 while the VMT is approximately 32. Both tend to vary with the household size as shown in the following table. PMT and VMT Household Size Mean PMT Mean VMT 1 22 16 2 46 34 3 54 37 4 82 45 5 71 42 Total 50 32 Highway Usage: The 1990 NPTS also provides sample of highway usage on different facility types similar to the traditional functional and facility classification. In Indianapolis, 1947 trips for a total of 16,937 miles were sampled. The average vehicle trip length of 8.70 miles is slight higher than the reported total for all trips. For this reason, only the percentage of trips and the percentage of the total miles are reported in the following table and presented in Figure IV-16: 33 Click HERE for graphic. Highway Usage Facility Type % of Trips % of Miles Av. Trip Length (miles) 2-3 Lane Highway 55 37 5.89 Undivided 4+ Lane Highway 14 8 5.16 Divided 4+ Lane Highway 17 22 11.19 Interstate & Freeway 14 33 20.48 Total 100 100 8.70 Make of Vehicles: The 1990 NPTS provides the interesting profile of the make of vehicles on the streets and highways of the Indianapolis Metropolitan Area as shown in the following table and Figure IV-17. Make of Vehicles Make % Total Dodge 6.3 Plymouth 2.6 Ford 17.2 Buick 6.1 Cadillac 2.4 Oldsmobile 10.2 GMC 1.8 Nissan 2.3 Chevrolet 20.8 Honda 3.7 Toyota 4.9 Pontiac 4.6 Travel Period Trips: There are 430 trips of 75 miles or longer made by the survey respondents in the last 14 days. Auto (71 %), Passenger Van (1 2 %) and Air Plane (9 %) are major modes of travel. However, this data is not statistically significant. Transit Segmented Trips: There are only 45 segmented trips for 14 person transit trips: 7 (50%) Home-based Work, 5 (36%) Home-based Other and 2 (14%) Non-home-based Other trips. 35 Click HERE for graphic. V. SUMMARY The NPTS Indianapolis Metropolitan Area Special Add-on Survey Data provides the most comprehensive data on local travel characteristics for the Indianapolis area since the 1964 IRTADS 1964 origin-destination study. The local NPTS will be most useful in updating the Long Range Plan for the Indianapolis Metropolitan Planning Area as required by the Intermodal Surface Transportation Efficiency Act of 1991. The development of the Long Range Transportation Plan is the responsibility of the Department of Metropolitan Development, Division of Planning as the designated Metropolitan Planning Organization (MPO) for the Indianapolis Urbanized Area which includes all of Marion County and portions of the surrounding Counties. Development of the Plan is a prerequisite to the area receiving some $21 million in federal transportation funds annually. The lack of local travel data has been problematic in the last two updates of the Indianapolis Long Range Transportation Plan. The local NPTS will be used to refine the travel simulation models used to estimate future travel and thus the recommended roadway improvements resulting from the Long Range Transportation Plan. The recommended improvements become the Official Thoroughfare Plan and are programmed as they near implementation in the Indianapolis Regional Transportation Improvement Program. In the City of Indianapolis-Marion County, the Official Thoroughfare Plan is adopted by the Metropolitan Development Commission as a segment of the Comprehensive Plan for Marion County. As such it is used by the City to ensure the orderly development of its transportation system. The NPTS was conducted to coincide with the 1990 Census. As part of the Census, the City of Indianapolis has also arranged to purchase the "Census Transportation Planning Package'(CT?P). This package will provide additional journey to work information for the Indianapolis Metropolitan Planning Area and will be used in conjunction with the local NPTS and other Census information to refine the travel simulation models. The Census Transportation Planning Package will not be available until some time in 1993 while the Intermodal Surface Transportation Efficiency Act requires that the Long Range Transportation Plan must be updated to reflect the requirements of the Act by October 1993. The Indianapolis NPTS will also be used as a comparison with the nationwide data. This comparative data can be used to infer how well our local transportation system is performing, i.e. average travel time can be an indication of congestion or how we compare with other travel characteristics, i.e. vehicle miles of travel per household. If the local data is found to be significantly different from the national average we would need to ask the question why and apply any unique character in establishing trends and developing projections. Representative Travel Characteristics The average Indianapolis household, according to the 1990 NPTS, has 2.15 persons, 1.23 workers, 1.60 licensed drivers and 1.61 vehicles. On an average day, it makes 4.53 vehicle trips averaging 7.53 miles for a total of 34.13 vehicle miles of travel (VMT). The 1990 37 NPTS underestimates the Marion County households by 2% and population by 8% (Figure V-1). Comparison with National Travel Statistics Generally the local NPTS compares reasonably well with the nationwide NPTS. Tables III-1 through III-8 compared the local NPTS with the nationwide NPTS for a wide array of travel characteristics. Refer to them for more detail. There were some difference that merit highlighting. The average vehicle trip length is 1.34 miles (or 15%) shorter in the Indianapolis area than the nation's average. This shorter trip length is also reflected in a lower vehicle miles of travel per household. Locally per household we travel 7.24 miles less per day than the nation as an average. Considering that the number of trips per household are comparable with the national average, we can conclude that the differences in daily vehicle miles of travel and average trip length are due to local residents living closer to where they work, shop and other trip purposes. Locally, the vehicle trips per household rate of 4.53 is 3 % below the national NPTS average trip rate of 4.68 trips per household. This and the shorter trip length then produce the shorter local VMT than the national average (Figure V-2). National Cooperative Highway Research Program (NCHRP) Report # 187: Quick-Response Urban Travel Estimation Techniques and Transferable Parameters is a commonly used document for transferable default values for the traditional four-step transportation planning process. Figure V-3 shows the comparison with the NCHRP # 187 default values for the urban area similar to Indianapolis with population between 750,000 to 1,000,000. The person trip rates are very close. Indianapolis is 2% smaller in term of the share of home-based work purpose trips. The auto occupancy rate of 1.1 persons per vehicle is significantly lower than the NC@ # 187 default value of 1.33 for home-based work trips. The following discussions summarize the technical comparison of the 1990 NPTS statistics with the major parameters used in the last major update in 1987. Trip Generation Rates The trip generation information (number of trips per household by household size and income category) will be useful in adjusting the rates used in the trip generation models. The specific adjustments will be determined in 1993 when the long range plan is being updated and additional Census Transportation Planning Package journey to work data is available. The comparisons of the trip rates derived from the local NPTS with the trip production rates currently being used by Division of Planning staff in forecasting future travel indicate that overall the rates developed by the COMSIS Corporation are on target for the Home-Based Work Trips and the Home-Based Shopping Trips. Differences in the 22 % to 41 % range were found in the Home-Based School, Home-Based Other, and Non-Home Based Other trip purposes (as shown in Figures IV-5 through IV-9). These three trip purposes account for approximately 58 percent of the internal trips forecasted by the Year 2005 Minutp travel simulation models. Based on the local NPTS, overall the current trip rates for these three trip purposes exceed the rates derived from the local NIPTS. This would imply that the Year 2005 38 Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. Click HERE for graphic. traffic forecast is over stated. Actual experience with the Year 2005 traffic assignment however has generally found the forecast to be under assigned. In light of the local NPTS data this under assignment may be due to an under assignment of external trips. The Division is attempting to conduct an external station survey to verify the number of trips coming from outside the study area. The distribution of trips among the five trip purposes is also a function of the trip rates used. As can be seen in the following table there are differences between the NPTS results and the COMSIS models. These differences could suggest that the COMSIS models might be over stating the number of Home-based Work trips while understating the number of Homebased School trips. However, another explanation could be an increase in the proportion of work trips by the Year 2005 and a decrease in the proportion of school trips during the same time frame. With the continued increase in the number of women in the work force and the continued decrease in household size this explanation might account for all or part of the differences identified. The important point is that the NPTS data indicates that a close reevaluation of the Year 2005 trip rates should be made as part of the 1993 update of the Indianapolis Long Range Transportation Plan. Trip Purposes (% of Total Trips) NPTS COMSIS Home-based Work 22.8% 28.3% Home-based Shopping 11.8 13.4 Home-based School 17.3 9.6 Home-based Other 26.3 26.4 Non-Home based Other 21.8 22.3 Average Trip Length (In Minutes) The average trip length from the COMSIS modeling process compares very well with the local NPTS data for all five trip purposes as shown below. Average Trip Length Comparison by Purpose (in minutes) Purpose NPTS COMSIS Home-based Work 18.09 18.19 Home-based Shopping 11.75 12.29 Home-based School 9.30 9.52 Home-based Other 13.56 13.78 Non-Home-based Other 12.61 12.23 All Purposes 13.98 14.07 42 The average trip length is a function of the Trip distribution model, which is loosely based on the law of gravity. The trip distribution model links up the trip ends estimated in the trip generation phase. Based on the above comparison it appears that the distribution model is doing a good job and would require little or no adjustments. Modal Split No conclusion can be drawn from the 1990 NPTS regarding the modal split models because of the small number (45) of transit trips included in the local survey data. Auto Occupancy In travel simulation modeling auto occupancy is used to convert auto person trips (persons making trips by auto) into auto vehicle trips before they can be assigned to the highway network. If the auto occupancy rates used for this purpose are too high the number of auto trips assigned to the network will be too low, if the auto occupancy rates are too low the number of auto trips assigned to the network will be too high. The Year 2005 COMSIS models derived average auto occupancies from the 1980 Urban Transportation Planning Package for the Home-based Work trips and from the 1964 IRTADS survey and synthesized data from other cities for the other trip purposes. The following table compares the auto occupancies used by COMSIS with the local NPTS data. Average Auto Occupancy Purpose NPTS COMSIS Home-based Work 1.1 1.14 Home-based Shopping 1.64 1.60 Home-based School 2.67 1.36 Combined Other 2.20 1.79 As the comparisons show the average auto occupancies used by COMSIS compare well with the local NPTS data for Home-based Work and Shopping trips indicating that little or no adjustments to the auto occupancy rates are needed. For other trip purposes, adjustments may be needed depending on the probable impacts on the model accuracy. Transferability Given multiple checks and comparisons, there are many consistent travel behaviors from the national NPTS and the Indianapolis Special Add-on data. With proper precautions, trip generation rates, trip length distribution for work trips and auto occupancy ratios for both home-based work trips and shopping trips can be transferred to other cities of comparable size and characteristics for their model calibration and verification process, if primary data sources are not cost effective. 43 In general, key local and national NPTS data as shown in Figure V-4 are consistent and stable over years. It is advisable to use these transferred data in other cities with caution. The percent of home-based work trips and the auto occupancy rate, especially for the work trip purpose, in the NCHRP # 187 need to be revised immediately for the application of the Quick Response System. 44
