Home

Use of Geographic Information Systems for

Contents

1. 20 a f o Ridership Cost NO YES NO YES NO YES Farebox Recovery Rate NO YES On time Performance E Small BMedium W Large NO YES Customer Satisfaction NO YES Accidents TABLE 22 AGENCY SIZE AND PERFORMANCE MEASURES ASSOCIATION Performance Measure Ridership Cost Farebox Recovery Rate On time Performance Customer Satisfaction Accidents Measure of Association Relating to Agency Size 3 169 2 204 0 217 6 373 2 117 5 315 38 0 205 0 332 0 897 0 041 0 347 0 070 0 397 0 165 0 076 0 446 0 274 0 054 0 179 0 108 0 051 0 307 0 175 0 037 Chi square Significance Gamma Kendall s tau c Significance 0 070 0 384 0 682 0 009 0 140 0 772 SUMMARY AND CONCLUSIONS The goal of this project was to determine the state of practice of transit agencies in the use of performance measures and advanced technologies to enhance transit management and performance measurement TCRP Report 88 asserted that agencies measure performance for three main reasons The first is that it is required to a certain extent for every agency Second it is useful for a transit agency to assess its performance in order to maintain and improve their service And third accurate information is needed to present to decision making bodies to ensure proper service as well as to the public so that people know if their transit system is valuable The resul
2. TABLE 8 SOURCES OF DATA SOURCE COUNT PERCENT RANK Fare box data or other accounting reports 87 88 78 1 Complaints 85 86 73 2 Data collected in house or from other field surveys 83 84 69 3 Accident Reports 82 83 67 4 Automatic Passenger Counter APC Systems 36 36 73 5 Automatic Vehicle Location AVL Systems 31 31 63 6 Others Other Technologies 7 7 14 7 Outside Groups Sources 4 4 08 8 Question What are your sources of data Check all that apply FIGURE 6 SOURCES OF DATA 100 5 90 80 70 60 50 40 30 20 Fare box dataor Complaints Data collected Accident Reports Automatic Automatic Other Outside other accounting in house or from Passenger Vehicle Location Technologies Groups Sources reports other field Counter APC AVL Systems surveys Systems To identify the changes of data collection methods the survey asked if the data was collected differently in the past Roughly 2 3 of respondents used the same way of collecting data as in the past Of those who did collect data differently we asked them to provide the methods previously used Table 9 and Figure 7 illustrate the previously used methods and show that it was most common for agencies to use manual methods and less advanced technologies 23 TABLE 9 PREVIOUS COLLECTION METHODS METHOD COUNT PERCENT RANK MANUAL REPORTS SURVEYS 23 67 65 1 LESS ADVANCED TECHNOLOGIES 9 26 47 2 ESTIMATE
3. gt f so what method s did you use Please list all past methods 6 Have you incurred any problems trying to collect data _ YES CINO gt If so what problems have occurred Please check all that apply _ A Data inconsistencies or measuring error L B High costs C C Inadequate equipment C D Limited data measures available L E Time consuming L F Unwillingness of participants to cooperate _ G Others please specify 7 Which of the following data sources are used for analysis Please check all that apply LC A Local planning agencies L B Local public works department L C National Transit Database NTD _ D State Department of Transportation DOT _ E U S Census Bureau L F Others please specify PART Il USE OF GEOGRAPHIC INFORMATION SYSTEMS GIS 8 Does your transit agency have a GIS software program that it can use C YES C NO gt Ifno why not Please check all that apply L A Lack of personnel to implement maintain software program L B No budget for software programs L C Not necessary in a small transit agency L D L E Software not available Software not user friendly 64 _ F Too costly _ G Others please specify 9 Do you use GIS and or GPS to collect data C YES CINO gt If no why not Please check all that apply A Lack of personnel to implement maintain software program B No budget for software programs C Not necess
4. Enter or click anywhere outside the text box You can edit the title by double clicking on it You can also click Change Symbol to edit the font type size etc Once your title is acceptable go to Insert and choose Legend to bring up the Legend Wizard The first box allows you to choose which layers you want included in your legend Use the multiple arrow buttons to move all layers back and forth and the single arrow buttons to move a single layer back and forth After clicking Next you are allowed to change the title of the legend After clicking Next again you are given options on the Border Background and Drop Shadow of your legend Click Next to be able to change the symbols that will appear for each layer in the legend Clicking Next one final time you are given a choice of spacing options Once you are satisfied with your selections click Finish Your legend will appear The next item is the north arrow which can be added using the Insert menu and clicking on North Arrow Choose your north arrow and click OK The last item that every map should have is the scale For this you can use a scale bar scale text or both For the scale bar go to Insert and choose Scale Bar Choose your scale bar For the scale text go to Insert and choose Scale Text Choose your scale text Once you have added all necessary map items move t
5. OTHER TECHNOLOGIES D Fare box technology Scheduling software Video cameras Announcement information Wireless handheld devices display technology for data collection Summary of Part Ill The third part of the survey focused on AVL and APC technologies AVL and APC technologies were not extensively used by the respondents Only one third of transit agencies have APC and 43 3 have AVL technologies For those that have AVL nearly all of them reported that AVL had improved on time performance While nearly all agencies felt that AVL could be used to measure performance only 27 of those 45 who have AVL technology currently use AVL data to measure transit performance The results for the APC questions were quite similar to those of the AVL questions Most agencies do not have APC but of those who do nearly all feel it has provided useful information to measure transit performance The agencies indicated that APC could definitely be used to measure performance but once again most agencies do not currently use APC for measuring performance This indicates a possible area of improvement for transit agencies if more agencies could find a way to take advantage of the two technologies 33 The survey question did not ask why the agencies that have AVL and APC do not use these technologies to evaluate transit performance So a follow up telephone interview was conducted We found that the most impo
6. Research Center for Urban Transportation Studies University of Wisconsin Milwaukee Milwaukee Wisconsin 53201 0784 9 SPONSORING MONITORING AGENCY NAME S AND ADDRESS ES 10 SPONSORING MONITORING AGENCY REPORT NUMBER Great Cities Universities Transportation Consortium University of Alabama Birmingham 11 SUPPLEMENTARY NOTES 12a DISTRIBUTION AVAILABILITY STATEMENT 12b DISTRIBUTION CODE 13 ABSTRACT Maximum 200 words This study addresses the use of Geographic Information Systems GIS for transit performance measurement Transit agencies that are members of the American Public Transportation Association APTA were contacted via survey to determine the use of performance measures GIS and other technologies such as Automatic Vehicle Locators AVL and Automatic Passenger Counters APC The results of the survey revealed a distinct gap between agencies believing the technologies are useful and agencies actually implementing them The National Transit Database NTD provided agency size and revenue data which matched with the surveyed agencies demonstrated a sharp contrast in regards to an agency s technology use and its size or revenue generated One reason that may discourage agencies from implementing GIS and other technologies is a lack of procedures and methods for implementing GIS for performance evaluation This study also includes a step by step guide to using GIS for performance assessment at the system and route level
7. Route Segment which was not included in calculating the average score of the Route Route Segment Summary of Part Il Part II of the survey indicated that GIS has been generally recognized as an efficient and effective way to integrate analyze and report transit performance particularly at the route or route segment level It is particularly useful for evaluating social accessibility measures The survey revealed that about two thirds of the respondents have GIS available in their agencies but only a little over half 56 of transit agencies have used GIS in analyzing and reporting transit performance It signifies that there are some transit agencies that have GIS software but do not use it for data analysis and reporting The major reasons for not having GIS or not using GIS software are the lack of technical personnel and the lack of budget Except for a few small transit agencies most agencies seem to recognize the importance of using GIS and GPS technologies for collecting analyzing and reporting transit performance data But many of them lack the technical personnel and budget to effectively use the technology for these purposes Use of Other Technologies for Transit Performance Measurement The use of other technologies was also explored with the survey This was done since other technologies besides GIS can be used to evaluate transit performance The two major technologies that were asked about in Part III were Automatic Vehicle Locato
8. and choose Data then Export Data You want to Export Selected Features and Use the same coordinate system as this layer s data source Click the Browse button 2 to choose your folder to save in choose Save as type Shapefile and rename it Route_1_ Stops Click Save and OK Click Yes when asked if you want to add the new exported data to the map as a layer Clear the selected features and uncheck the box next to the Stops layer you may also want to change the width and color of the symbol You can now see the stops for Route 1 Step 10 Buffering Route 1 Bus Stops To buffer the Route_1_ Stops layer follow the same procedure as step 5 You will locate the transit service catchment area by buffering all locations within a quarter mile of each route 1 bus stop To buffer open ArcToolbox Click on Analysis Tools then Proximity and choose Buffer This will launch the Buffer dialog box Choose the Route_1_ Stops layer as the Input Features layer then click the Browse button and navigating to the folder where you are saving your files and layers Name the layer Route_1_Stops_ Buffer Under Distance type in 25 for Linear unit and choose Miles from the drop down menu For Dissolve Type choose ALL then click OK You can see the transit service catchment area shown in Figure 30 represented by the polygons
9. for each technology GIS AVL and APC Table 20 shows measures of association along with their respective significance levels The measures of association tested shown in Table 20 indicate that the use of technology is closely related with the size of transit agencies The general pattern seems to be the larger the transit agency size the more likely they are to use technologies to measure transit performance and for other purposes For example larger agencies are more likely to have GIS AVL and APC technology and the more likely it is to use GIS or GPS in performance measurement But in terms of believing AVL and APC can improve useful information and can be used to measure transit performance there is no difference in terms of agency size all believe they can 34 TABLE 19 AGENCY SIZE AND TECHNOLOGY Agency Size Question topic Response Small Medium Large GIS accessible NO 11 9 2 YES 18 19 25 GIS GPS used to collect data NO 15 9 6 YES 14 19 21 GIS used to analyze report performance data NO 22 13 7 YES 7 15 19 AVL possessed NO 21 17 6 YES 8 11 20 AVL has improved on time performance NO 3 0 6 YES 5 10 13 AVL can be used to measure transit performance NO 0 2 0 YES 28 25 26 AVL is currently used to measure transit performance NO 23 20 12 YES 4 5 14 APC possessed NO 25 17 7 YES 4 11 19 APC has produced useful information NO 0 0 2 YES 4 10 16 APC can be used to measure transit performance NO 4 0 0 YES 22 29 27 APC is c
10. for our first example Then for the Normalization dropdown choose Acre or an area field for your polygons By doing this we have created a measure of 2000 Population per Acre Figure 24 shows the symbology created by this procedure 52 FIGURE 24 SYMBOLOGY Layer Properties General Source Selection Display Symbology Fields Definition Query Labels Joins amp Relates Show FAIRE Draw quantities using color to show values Import Categories Fields Classification Natural Breaks Jenks Classes 5 Quantities Value mn POP2000 Graduated colors Graduated symbols Proportional symbols Dot density Color Ramp aa Charts Multiple Attributes Normalization lace o l _ 0 178372595 3 29621249 0 178372595 3 29621249 I 3 29621250 6 90936397 3 29621250 6 90936387 D 6 20936388 12 2156643 6 90936388 12 2156643 WE 22156644 22 8036809 12 2156644 22 8036809 D 22 036810 53 2342131 22 8036810 53 2342131 F Show class ranges using feature values Advanced 7 OK Cancel Apply There are a number of different ways to format categories Color Ramp Symbol Range and Label can be changed in the Symbology tab You can also change the type of classification and the number of classes that your variable is split up into by clicking the Classify button in the upper right of the Symbology tab Specific formatting can be done by
11. of a transit agency The guide uses demographic data from the United States Census Bureau and geographic data from Waukesha Metro s transit system to give visual representations and provide examples of performance measures which can be analyzed with GIS 14 SUBJECT TERMS 15 NUMBER OF PAGES Geographic information systems GIS Automatic Vehicle Location Systems 67 Performance evaluation public transportation mass transit 16 PRICE CODE 17 SECURITY CLASSIFICATION 18 SECURITY CLASSIFICATION 19 SECURITY CLASSIFICATION 20 LIMITATION OF ABSTRACT OF REPORT OF THIS PAGE OF ABSTRACT Unclassified Unclassified Unclassified Use of Geographic Information Systems for Transit Performance Measurement October 2006 Prepared by Zhong Ren Peng Eric Lynde and Edward A Beimborn With contributions from Jack Reilly and Robert Schmitt Center for Advanced Spatial Information Research School of Architecture and Urban Planning and Center for Urban Transportation Studies University of Wisconsin Milwaukee Milwaukee WI 53211 414 229 5887 http www casir uwm edu http Awww uwm edu Dept cuts Prepared for U S Department of Transportation Federal Transit Administration Through the Great Cities University Transportation Consortium ABSTRACT This study addresses the use of Geographic Information Systems GIS for transit performance measurement Transit agencies that are members of the American Public Transpor
12. of the Route_1_Stops_ Buffer layer 58 i Layers Route_1_stops 50O Stops ka E M Route_t O Routes E Route_1_stops_Buffer C Stops_Buffer_Intersect E Block_Groups o H Data Management Tools E Geocoding Tool E Geostatistical Analyst Tools om Spatial Statistics Tools Step 11 Overlaying Route 1 Bus Stops Buffer with Census Block Group Data Step 11 is a similar procedure to step 6 You now want to overlay the Route_1_Stops_Buffer layer with the Block_Groups layer In ArcToolbox select Analysis Tools select Overlay then Intersect For Input Features select both the Route_1_Stops_Buffer layer and the Block_Groups layer Change the name of the Output Feature Class to Route_1_Stops_Buffer_Intersect Select the defaults for Join Attributes and Cluster Tolerance For Output Type make sure that Input is selected Select OK Once the process concludes you will be able to see the results of the Intersect Figure 31 shows these results 59 FIGURE 31 INTERSECT RESULTS ZOOMED IN gt Transit txomplemmnd ArMop Arcino OOO ziei File Edit View Insert Selection Tools Window Help Oa S BX 9 amp 117358 W 4 HBONM QQuUUNSeDSR OMS x xl amp Layers ig ArcToolbox SS xe B M Route_1_stops H 30 Analyst Tools poy e EH Analysis Tools oe O Stops H Ext
13. select all the stops that are within 150 feet of the features in Route_1 You should visually inspect the selected features to make sure that they are indeed part of the Route 1 s stops To do this click the Zoom In button on the Tools toolbar and click and drag a box around the area you want to inspect If you want to add an additional stop or unselect a stop go to the Selection menu then select Set Selectable Layers and uncheck the box next to each layer except Stops To select an additional stop go to the Selection menu then select Interactive Selection Method and Add to Current Selection To unselect a stop go to the Selection menu then select Interactive Selection Method and Remove from Current Selection To add or unselect the stop click the Select Features button then click on the stop Figure 29 shows the stops selected for Route 1 57 FIGURE 29 ROUTE 1 BUS STOPS SELECTED File Edit View Insert Selection Tools Window Help Dee S BBX a a g 13868 MIRAE AOE A E ag aM Stops m EN S E M Route_1 amp O Routes E CO Stops_Buffer_Intersect E Block_Groups Oo rawn k O Bl Oy Aw fojaia wl pz ulAr r gr te 2471760 40 364755 69 Feet Once the stops have been selected you must export them as a new shapefile Right click on the Stops layer
14. using with the different options until you find something that best suits your agency Once you have found the best format click OK The symbology selected will be displayed fitting each polygon s attributes into one of the categories that were created Figure 25 shows a zoomed in view of the 2000 Population per Acre display 53 FIGURE 25 ZOOMED IN VIEW OF SYMBOLOGY RESULTS Untitled ArcMap ArcInfo DER Bile Edit Yiew Insert Selection Tools Window Help Dengi aex noy hs Viv Peo r7 QQrurT esd Gh OMS S SS e foo ge i x a ag ArcToolbox B Stops E 30 Analyst Tools 8 amp Analysis Tools Routes Cartography Tools H Conversion Tools Stops_Buffer_Intersect Coverage Tools POP2000 Acre i Data Interoperability Tools 70 178372595 3 296 Data Tools E 3 29621250 6 9093 E 6 90936388 12 215 WE 12 2156644 22 803 WE 22 8036810 53 234 B Block_Groups oO 1 Spatial Analyst Tools Spatial Statistics Tools lt i gt f Displey Source Selecion _Favortes Index Search fonen proving k O EO Ay amp fiojaia a pz ul Ar amp fy sy Draw a rectangle 2457232 77 370630 98 Feet This step can be repeated for all the system wide performance measures that you wish to map out would recommend saving a new mxd file for each variable Step 8 Selecting A
15. 1 3 15 14 Third Street 28 16 1 1 74 18 Delaware Avenue 19 10 4 2 24 Source Capital District Transit Authority Albany N Y Social Measures While most of the work in system wide performance measures is focused on business measures such as revenue to cost ratios there is also a need for transit systems to use social indicators Examples of these would include what proportion of households without autos has transit service of a specified quality One particularly helpful social measure is an estimate of the number of households with particular social characteristics in a transit district within a quarter mile of a bus stop this is a five minute walk at three miles per hour For example one can determine the proportion of households without autos served by daytime service the proportion of households with direct access by transit to grocery stores and the proportion of jobs in the transit service territory It should be noted that different measures of transit accessibility could generate drastically different results A quarter mile of air distance from a bus stop without considering the actual street network would over estimate the accessibility of residents who may have to walk much longer than a quarter mile on the street network to access the bus stop To address this issue the Transit Level of Service software TLOS developed by the Florida Department of Transportation Figures 1 and 2 provides a technique for measuring transit acce
16. 1 CAR 100 57 55 41 53 35 WORKERS 100 44 42 29 40 24 AREA 100 5 4 2 3 2 Source Capital District Transit Authority Albany N Y FIGURE 3 PROPORTION OF POPULATION IN SERVICE AREA BY SERVICE PERIOD 100 90 80 TOTAL TOTAL ELDERLY EMPLOYMENT HOUSEHOLDS HOUSEHOLDS WORKERS AREA POPULATION HOUSEHOLDS POPULATION 0 CARS 0 1 CAR PEAK MIDDAY BEVENING BSATURDAY amp SUNDAY Social measures can also be made at the route level By using geographic information systems one can compute the number of households within 0 25 miles of the route the 16 distribution of auto ownership within the route s market area etc Further the characteristics of the route service can be compared with the characteristics of the system service area The tables below illustrate this In practice three kinds of measures are commonly used TCRP 88 The first are revenue and cost measures Examples include gross profit margin net income percent of revenue from fare box and cost per rider mile trip The second category is system and change monitoring It can include secondary data such as inventory on hand number of complaints accidents per mile and number of vehicle washings Thirdly there is customer satisfaction and loyalty This must answer questions like does the service meet customers expectations Also will customers recommend services or continue to repurchas
17. 24 Steps in Assembly of National Transit Database Data Go to FTA website http www fta dot gov Click Transit Data and Info which takes you to http www fta dot gov 14512 ENG HTML htm Click on Transit Agency Financial amp Operating Data Tables which takes you to http www fta dot gov transit data_info national transit _db 1082 611 ENG HT ML htm Click on Transit Agency Financial amp Operating Data Tables again which takes you to http Awww ntdprogram com NTD NTDData nsf DataTableInformation OpenForm amp 2001 On the left side of the page Right Click O1tabxls exe and Click Save Target As Choose where to save the file and click Save Once the download completes find where you saved the file and double click it Click Run Choose the folder to Unzip to and click Unzip After successful unzipping find you agency s row in each Microsoft Excel spreadsheet to locate your data For an explanation of the tables follow the first four steps then Right Click Data Tables for Individual Transit Agency Statistics and Click Save Target As Choose where to save the file and click Save Once the download completes find where you saved the file and double click it to open it Steps in Assembly of TIGER Data Topologically Integrated Geographic Encoding and Referencing TIGER TIGER files are from the U S Census Bureau Defines location relat
18. 52 Step 8 Selecting A Ro te cicisissisieriardessansndieradessauataniavsvansasocataeresasuaxinrcousebnencenitonanes 54 Step 9 Creating A Route 1 Bus Stops Layer ccccccceeeeecet nee eetecnneeeeeeeeneneees 56 Step 10 Buffering Route 1 Bus Stops 0 cccccceesssseeceeeeeeeseennnneeeeeseeesssesnntssaaees 58 Step 11 Overlaying Route 1 Bus Stops Buffer with Census Block Group Data 59 Step 12 Route Level Transit Service Analysis ccccccccceseeeeteeeeeeteeneeteees 60 Step 13 Making a Map Document with the Layout VieW cccccccssseees 61 Step 14 Exporting Map as A PDF File ccccccccccesssseceseeesessensnseseseeeeseesensnaees 61 APPENDIX 1 SURVEY FORM USED IN PROJECT 0 0ccccccccccceeseeeeeeeeenees 63 INTRODUCTION It is important for public transit agencies to measure transit performance easily and accurately This fact is emphasized by Transit Cooperative Research Program Report TCRP 88 which will be used for much of the support of this paper This report outlines three reasons for measuring performance First it is required for transit agencies as a condition for federal funding Secondly it is useful for a transit agency to assess its performance in order to maintain and improve their service And thirdly accurate information is needed for decision making bodies to oversee transit service The report sums up the use of performance measures by stating that they c
19. 7 N 523 634 Source TLOS Tallahasse F L 13 Ausman Jon TLOS Demonstration Urban Transportation Planning Short Course Tampa Florida March 2002 14 FIGURE 1 AIR DISTANCE BUFFERS FROM TRANSIT STOPS 5 MINUTE DISTANCES TALLAHASSEE TRANSIT SYSTEM ie TLOS MAP File Data ILOS Mobility Tools Window SAWN i n ener b a N ij AY i AN b p J n ERNA z 3h Hap Options Start TLOS Values TCOSM Values Time El 5 j0 ama Monday AirBuffers e Service End Frequency A Time o fs fem Sonday Walk Buffers p Senice Taz Cover f Ai C We age Mins Total 1 898 901 360 889 373 520 ir Buers akaa lhe Group 3 c His OF Mins Served 349 731 333 231 814 558 oa egy to xSevet aa zat Calculate TLOS Help Shade Hep C jag Astan HVTAPIaniation Wak Bulle STLOS MAP Jase WS WS naM FIGURE 2 ACTUAL WALK BUFFERS FOLLOWING LOCAL STREET PATTERNS TALLAHASSEE TRANSIT SYSTEM i TLOS MAP BEE File Data ILOS Mobility Tools Window alB j Ba fa 2 E m Ot Full Pan Identiy _TLOS Maps Reports Legends Hilte Clear Gro Path 1 I moea l M Local Roads M TAZ Borders Class 1 M Bus Routes Class 2 I Bus Stops A Chass 3 Oo F Buff
20. ATA DIALOG BOX Export Data Export All features v Use the same coordinate system as this layer s source data the data frame the feature dataset you export the data into only applies if you export to a feature dataset in a geodatabase Output shapefile or feature class Z New_Block_Groups shp 5 Cancel A popup message will ask if you want to add the new exported data to the map as a layer Click Yes Once the layer is added right click on the old block group polygon layer and choose Remove Step 5 Buffering Bus Stops The next step will be to locate the transit service catchment area by buffering all locations within a quarter mile of a bus stop Buffering creates a polygon of an area in a new layer by drawing a constant distance around a feature a point a line or a polygon To buffer open ArcToolbox Click on Analysis Tools then Proximity and choose Buffer This will launch the Buffer dialog box shown below in Figure 21 When this box comes up you will first need to choose your bus stops layer as the Input Features layer This is the layer that will be buffered Then choose the location for saving the buffer layer to by clicking the Browse buttonl and navigating to the folder where the files and layers are saved Name the layer Stops_ Buffer Under Distance type in 25 for Linear unit and choose Miles from the drop down menu This will make
21. Another set of questions related to the collection of transit performance data This includes sources of the data how the data has been collected in the past problems with collecting the data and sources for analysis Of the 107 responses 98 91 6 say they use transit performance measures in their service and operation planning and evaluation while 9 of the respondents did not Follow up questions were then asked for those that do use transit performance measures When asked why do you use transit performance measures the most popular response was that the measures are useful to maintain and improve service as shown in Table 6 and Figure 4 The other two popular responses are that they are needed to report to decision making bodies and to measure whether agency goals are being met TABLE 6 WHY PERFORMANCE MEASURES ARE USED REASONS FOR TRANSIT PERFORMANCE MEASURES COUNT PERCENT RANK hey are required for the National Transit Database The total percentage does not add up to 100 because the question asked the respondents to check all that apply 19 FIGURE 4 WHY PERFORMANCE MEASURES ARE USED CZ 90 80 70 60 50 40 30 20 They are useful to They are needed to They are needed to They are required They are needed to They are needed to Others maintain and report to decision measure whether for the Nat
22. C possessed 19 97 0 000 0 722 0 523 0 000 APC has produced useful information 1 66 0 436 1 000 0 109 0 134 APC can be used to measure transit performance 9 06 0 011 1 000 0 133 0 033 APC is currently used to measure transit performance 13 42 0 001 0 656 0 454 0 000 We also wanted to find out if agencies of different size use different performance measures Tables 21 and 22 along with Figure 17 display the results of association tests comparing agency size with different performance measures The six performance measures chosen were the six most popular measures from the survey The results show that except for on time performance there is not much difference in performance measures used at different sized agencies On time performance however was found to be associated with the size of agency Large agencies seem to be more concerned about using on time performance as a measure while smaller agencies are less concerned with this measure This may be because on time performance in small agencies is not as big a problem as in large agencies TABLE 21 AGENCY SIZE AND PERFORMANCE MEASURES Agency Size Performance Measure Response Small Medium Large Ridership NO 7 6 3 YES 15 22 24 Cost NO 9 9 14 YES 13 19 13 Farebox Recovery Rate NO 11 14 15 YES 11 14 12 On time Performance NO 15 16 9 YES 7 12 18 Customer Satisfaction NO 11 17 19 YES 11 11 8 Accidents NO 9 20 13 YES 13 8 14 37 FIGURE 17 AGENCY SIZE AND PERFORMANCE MEASURES 25 4
23. D WITH REVENUE DATA 1 2 94 3 DID NOT VERIFY COMPLAINTS 1 2 94 3 34 100 00 Question If so what method s did you use List all methods FIGURE 7 PAST COLLECTION METHODS ESTIMATED WITH DID NOT VERIFY REVENUE DATA _ COMPLAINTS 3 r 3 LESS ADVANCED TECHNOLOGIES 26 Question six asked about problems that have been encountered in collecting performance data This question aimed to find the problems associated with gathering data for transit performance assessment Just over 3 4 of agencies 65 out of 86 reported that they had experienced data collection problems A follow up question asked what the problems were that had occurred These results are shown in Table 10 and Figure 8 Data inconsistency time requirements and lack of proper equipment were cited by a majority of respondents as problems TABLE 10 SPECIFIC DATA COLLECTION PROBLEMS PROBLEM COUNT PERCENT RANK Data inconsistencies or measuring error 47 72 31 1 Time consuming 43 66 15 2 Inadequate equipment 38 58 46 3 Limited data measures available 31 47 69 4 High costs 24 36 92 5 Unwillingness of participants to cooperate 15 23 08 6 Others 5 7 69 7 Question If so what problems have occurred Check all that apply 24 FIGURE 8 SPECIFIC DATA COLLECTION PROBLEMS 45 40 35 30 25 20 15 10 5 E e m E Ey m Data Time consuming Inadequate Limited High costs Unwillingness of
24. ENCIES In order to understand the state of the practice in the use of transit performance measurement and the use of new technology in measuring transit performance a survey was conducted for transit agencies in the U S The purpose of the survey was to determine the best practice of transit agencies across the country It is split up into three parts use of transit performance measures and measurement techniques the use of GIS to measure transit performance and the use of other technologies such as AVL and APC to measure transit performance The survey was sent to all of the transit agencies on the American Public Transportation Association membership list totaling nearly 400 agencies The agencies were given the option of returning the paper survey or filling out the same survey made available on the internet Of the agencies surveyed 107 of them responded and the data collected from their surveys was compiled and coded for analysis A copy of the survey used is given in the appendix of this report This section will present the findings of the survey Transit Performance Measures and Measurement Techniques The questions in Part of the survey were intended to determine the use of transit performance measurement by U S transit agencies We wanted to know if and why transit performance measures are used We also wanted to find out what the agencies feel are the most important transit performance measures that they use or would like to use
25. ETM S a ee _FTA Co oe Han U S Department Federal Transit of Transportation Administration Geographic Information Systems for Transit Performance Measurement A report of the Center for Advanced Spatial Information Research and Center for Urban Transportation Studies University of Wisconsin Milwaukee October 2006 Form Approved REPORT DOCUMENTATION PAGE OMB No 0704 0188 Public reporting burden for this collection of information is estimated to average 1 hour per response including the time for reviewing instructions searching existing data sources gathering and maintaining the data needed and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of this Collection of information including suggestions for reducing this burden to Washington Headquarters Services Directorate for Information Operations and Reports 1215 J efferson Davis Highway Suite 1204 Arlington VA 22202 4302 and to the Office of Management and Budget Paperwork Reduction Project 0704 0188 Washington DC 20503 1 AGENCY USE ONLY Leave blank 2 REPORT DATE 3 REPORT TYPE AND DATES COVERED October 2006 4 TITLE AND SUBTITLE 5 FUNDING NUMBERS Use of Geographic Information Systems for Transit Performance Measurement 6 AUTHOR S Zhong Ren Peng Eric Lynde and Edward Beimborn 7 PERFORMING ORGANIZATION NAME S AND ADDRESS ES ieee Center for Advanced Spatial Information
26. IS and its Applications in Transportation an article written by Peng and Beimborn of the University of Wisconsin Milwaukee GIS is being used to integrate analyze and display spatial data This use can be applied to transit using the proper data to show the effectiveness of a transit system This report will focus on the use of Geographic Information Systems GIS for measuring the performance of transit systems The report will begin by defining transit performance and illustrating how it is measured and address issues with transit performance and demonstrate the positive aspects of GIS use in transit performance assessment Next the data necessary for proper implementation and analysis will be identified and discussed Then state of the practice within transit agencies will be examined through the use of surveys Transit Cooperative Research Program Report 88 A Guidebook for Developing a Transit Performance Measurement System FTA Transportation Research Board 2003 Ibid p 5 3 Bertini Robert L and El Geneidy Anmed Using Archived Data to Generate Transit Performance Measures Transportation Research Board November 2002 pp 1 24 Beimborn Edward A and Peng Zhong Ren Internet GIS and Its Applications in Transportation TR News March April 1998 pp 1 10 both direct and online of transit agencies across the nation The final topic will present a case study of the use of GIS as a means of transit performance ass
27. Milwaukee 1996 98 This material was written by Jack Reilly of the Capital District Transportation Authority Albany N Y Edward Beimborn of the University of Wisconsin Milwaukee and Robert Schmitt of RTR Associates in Pittsburg This report was developed as part of work being conducted by the Great Cities University consortium under the lead of the University of Alabama at Birmingham using funds provided by the Federal Transit Administration of the U S Department of Transportation The opinions expressed are the product of independent university work and not necessarily those of the sponsoring agencies or of the agencies supplying data for the project TABLE OF CONTENTS PSN De acta saa ecg sna a aban eeccuas ens A 4 AGKNOWEEDGMENT 83 ccccsuasivsiaieinteiiederinnatoentoncomeieenaieeise Sass 5 INTRODUCTION nissiininineninnrnannn e e e AR a a A E f TRANSIT PERFORMANCE MEASURES ccccccccccsecssecseesscesecesecssecseeeseeeeeeseseseesseens 9 Financial Indicators 1 ccceeeeeeeenneeeneceneccaeccaeceaeeesaeesaecsaeceaeedaeceseeeeeeeseeeseeesteeeteesseeeees 10 Non Financial Indicators 1 ceseeeeeeeneeeneeceneceeceeseaeeceaeeaaeesseedsesseeeeeeeeseesseeeeeeseteesseesees 11 Bu sin ss MCASUIES eerie EEE inne REER 12 OCIA Measure S oorner aA E R 13 SURVEY OF TRANSIT PERFORMANCE MEASUREMENT IN TRANSIT AGENCIES orean e EEA AE AA E EN A E ERA e 19 Transit Performance Measures and Measurement Techniques 000 19 Use
28. NCE INDICATORS FROM APC COUNT PERCENT RANK Passengers per stop 21 65 63 1 Load Factor 16 50 00 2 Passengers per vehicle hour vehicle mile 15 46 88 3 Others On time performance 3 9 38 4 Fare box recovery rate 2 6 25 5 Others Point boardings 1 3 13 6 Schedule related measures 1 3 13 6 Question If yes what performance indicators does APC produce Check all that apply FIGURE 14 APC PRODUCED PERFORMANCE INDICATORS 25 20 Passengers per Load Factor Passengers per On time Fare box recovery Point boardings Schedule related stop vehicle performance rate measures hour vehicle mile The final question of the survey asked the agencies if they used other technologies and if so what those technologies were Less than half of the agencies use other technologies 44 Out of those who do use other technologies the top five technologies as shown in Table 18 and Figure 15 are farebox technology scheduling software video cameras stop problems have been encountered in collecting performance data announcements and hand held data collectors 32 TABLE 18 PERFORMANCE INDICATORS FROM OTHER TECHNOLOGIES TECHNOLOGY TOTAL RANK Fare box technology 15 1 Scheduling software 11 2 Video cameras 8 3 Announcement information display technology 5 4 Wireless handheld devices for data collection 5 4 Question If so what are these other technologies List all others FIGURE 15 PERFORMANCE INDICATORS FROM
29. Others inconsistencies or equipment data measures participants to measuring error available cooperate In addition we asked what additional existing data are used for analyzing transit performance The results are shown in Table 6 and Figure 9 It shows that the National Transit Database NTD local planning agencies and the U S Census were the most commonly used data sources TABLE 11 DATA SOURCES USED FOR ANALYSIS SOURCE COUNT PERCENT RANK National Transit Database NTD 75 76 53 1 Local planning agencies 62 63 27 2 U S Census Bureau 56 57 14 3 State Department of Transportation DOT 34 34 69 4 Local public works department 13 13 27 5 Others Other outside groups sources 11 11 22 6 In house data 9 9 18 7 The survey asked to check all that apply Therefore the total does not add to 100 25 FIGURE 9 DATA SOURCES USED FOR ANALYSIS 70 60 50 40 30 20 ia i National Transit Local planning U S Census State Department tee public works Other outside In house data Database NTD agencies Bureau of Transportation department groups sources DOT Summary of part 1 Overall the results of the first part of the survey show that most transit agencies use transit performance measures to evaluate the performance of their transit systems The major purpose of using transit performance measures is to improve transit services while a secondary goal is satisfyin
30. P Service Evaluation and Monitoring Report prepared for the Pennsylvania Department of Transportation TCRP 88 op cit 8 Transit Cooperative Research Program Report 100 Transit Capacity and Quality of Service Manual Second Edition FTA Transportation Research Board 2003 622 pages ibid spatial availability where service is and whether customer can get to it temporal availability when service is available information availability customer is informed on how to use service and capacity availability space is available for trip With comfort and convenience some factors that a transit agency has some control over include e service delivery frequency of service e travel time transit trip time versus other modes e safety and security risks of being injured or being a victim to crime and e maintenance reliability of system s vehicles The second point of view is that of the community Here the concern is with areas in which transit service benefits the community as a whole as well as the costs and negative aspects of transit service From the standpoint of community support a transit agency does well to provide evidence of its performance and the way in which it deals with negative concerns The third viewpoint is that of the agency itself which has a distinctly different view from the customers and community The agency must constantly ensure that its operations run smoothly and is concern
31. PC C YES _ NO If no please go to 15 gt If yes do you feel that this technology has produced useful information C YES CINO 15 Do you think that APC can be used to measure transit performance C YES CINO If no please go to 16 gt If yes do you currently use APC for the purpose of transit performance measurement C YES CINO If no please go to 16 gt If yes what performance indicators does APC produce Please check all that apply _ A Fare box recovery rate LC B Load Factor _ C Passengers per stop L D Passengers per vehicle hour vehicle mile L E Others please specify 16 Do you use other technologies C YES CINO gt Ifso what are they Please list all others 66
32. Route For the routes layer there should be a RouteNumber field if you do not have this field you will need to obtain data matching each link of the Routes layer to a route number This field has the route number associated with each link of the Routes layer Links for a specific segment are selected by going to the Selection menu and choosing Select By Attributes For the Layer dropdown choose Routes layer The Method should be Create a new selection Where it says SELECT FROM Routes WHERE type in the following RouteNumber 1 be sure that to use double quotations around the route number field that you put one space before and after the equals sign and that you put the route number of the route you wish to analyze after the equals sign When everything looks like Figure 26 below click Apply then OK 54 FIGURE 26 SELECT BY ATTRIBUTES DIALOG BOX Select By Attributes 2x Layer Routes X 7 Only show selectable layers in this list Method Create anew selection FID FNODE_ TNODE_ LENGTH LINKID SRC ee EN ES Ba E Ta jet Unique Values GoTo SELECT FROM Routes WHERE Lele JINAN FEE RouteNumber 1 Clear Verify Help Load Save Apply Close After hitting Apply ArcMap will select each link of Route 1 or the route chosen and the route will be highlighted in li
33. Routes layer You should see your route in the display you may want to change the width and color of the line by double clicking the line symbol under Route_1 Step 9 Creating A Route 1 Bus Stops Layer To select all the bus stops that are part of Route 1 go to the Selection menu then choose Select By Location In this box choose select features from check the box next to Stops choose are within a distance of choose Route_1 leave the Apply a buffer to the features in Route_1 box checked choose Feet and type in 150 in the box next to Feet Click Apply Figure 28 shows the Select By Locations dialog box 56 FIGURE 28 SELECT BY LOCATION DIALOG BOX Select By Location 21 xi Lets you select features from one or more layers based on where they are located in relation to the features in another layer want to select features from the following layer s I Only show selectable layers in this list that are within a distance of the features in this layer Route_1 Use selected features 0 features selected IV Apply a buffer to the features in Route_1 of fi 50 000000 Feet z Preview The red features represent the features in Route_1 The highlighted cyan features are selected because they are within a distance of the red features Paints Lines Polygons ome This will
34. a describing workers in each tract They also separated trip types into three categories The first being work trips of residents in the tract Second there are work trips of workers in the tract Third there are non work trips in the tract GIS Analysis for Performance Measures A Case Study of the Waukesha Metro Transit System Details of the process and data sources are given below and should be consulted if the process is to be repeated in another location Step 1 Gather the Necessary Data The first data that are needed are vector data for the area the agency serves Vector data describe individual geographic features which can be in the forms of points lines and polygon objects which are defined by mathematical formulas The most important points in a transit GIS are bus stops The lines are the existing bus routes The polygons will be the representations of different areas such as census tracts and block groups The next data that will be incorporated into the GIS are demographic data for residents which are collected by the U S Census Bureau Details of data needs and steps to produce results are given in the following tables 43 Table 23 Data Needs for a Transit GIS General data Point file that represents locations of landmarks Point file that represents locations of bus stops Line file that represents the network of bus routes Line file that represents the network of streets Polygon file that represents the
35. an provide perspective understanding and context to what has gone on and what is going on within and organization The easier it is to extract these measures the less time and money it will take to produce them and also for others to reproduce them for their purposes So obviously if this analysis can be done easily and accurately it is to the benefit of everyone involved There are a number of critical problems with the current methods of measuring transit performance First the ways of collecting data on performance have been an issue Bertini and El Geneidy in Using Archived Data to Generate Performance Measures state that it has been difficult and costly to collect comprehensive performance data They go on to say that agencies have had to use limited general aggregate measures for reporting performance to external funding and regulatory agencies Second each agency must create its own way of measuring and implementing a system to assess the performance of their agency Transit agencies would benefit by uniform ways to measure and assess performance that will simplify the data input and easily produce comprehensive reports on their performance The purpose of this report is to examine how transit performance measurement could be enhanced through the use of geographic information systems GIS The importance of GIS use is that it allows for easy accurate measurement of transit performance According to Internet G
36. ary in a small transit agency D Software is not user friendly Too costly LI L1 L1 L1 _ E L F Others please specify 10 Do you use GIS to analyze or report performance data YES CINO gt If yes please go to question 11 gt If no why don t you use GIS Please check all that apply and skip to 12 Lack of personnel to implement maintain software program LI A L B oa is not user friendly L C Others please specify 11 How important is GIS in measuring performance at each of the following levels Not Important Very Important A System Level eneee 1 2 3 4 5 B Route Route Segment Level 08 1 2 3 4 5 PART Ill USE OF OTHER TECHNOLOGIES 12 Do you have an Automatic Vehicle Location AVL system C YES C NO gt If yes do you feel that this technology has improved on time performance C YES CINO 13 Do you think that AVL can be used to measure transit performance C YES CINO If no please go to 14 gt If yes do you currently use AVL for the purpose of transit performance measurement 65 C YES CINO If no please go to 14 gt If yes what performance indicators does AVL produce Please check all that apply C A Average dwell time at stops L B On time performance L C Percent of route deviation L D Response time to accidents or breakdowns L E Others please specify 14 Do you have Automatic Passenger Counters A
37. ated a number of reports that deal with transit performance issues Of these two of them have a significant relevance to our research Both TCRP Report 88 and TCRP Report 100 have provided very useful information on transit performance TCRP 88 A Guidebook for Developing a Transit Performance Measurement System was published in 2003 TCRP 100 is the Transit Capacity and Quality of Service Manual and was published in its second edition in 2003 The TCRP 88 method of defining and measuring transit performance requires looking at the performance from different points of view The viewpoints are those from the customer the community the agency and the driver vehicle The first viewpoint is that of the customer The customer is very important as it is the customer who uses transit and for whom it was created for According to the report the customer has two main concerns service availability and the comfort and convenience of service when it is available There are four factors that affect service availability This material is adapted from a NTI course Improving Transit System Performance Using Information Based Strategies developed at the University of Wisconsin Milwaukee 1996 98 This material was written by Jack Reilly of the Capital District Transportation Authority Albany N Y Edward Beimborn of the University of Wisconsin Milwaukee and Robert Schmitt of RTR Associates in Pittsburg Schmitt Robert
38. ay be useful in designing stops and assigning equipment e Passenger Trips Population of Service Area An indicator of the level of transit use in an area Service Quality e Number of Complaints Number of Drivers A rough measure of consumer dissatisfaction e Stops On Time Total Stops A measure of on time performance e Vehicle Miles Road Calls A measure of miles between road calls a surrogate for fleet age and maintenance effectiveness Level of Service e Revenue Miles Revenue Hours A measure of concentration of services e Vehicle Miles Year A useful measure for comparing level of service over time e Vehicle Hours Year A useful measure for comparing level of service over time and as an element in calculating additional indicators Safety e Vehicle Miles Vehicle Accidents The number of vehicle miles between accidents an important safety indicator 11 e Avoidable Accidents per Year A useful safety indicator and one often used for setting safety standards The indicators may be expressed in many different ways depending on what the manager wants to measure Total vehicle miles and revenue vehicle miles are different indices they measure slightly different things farebox revenue and total revenue also differ significantly in some systems Most indicators are appropriate at the system and route level while others are useful primarily at the route level Performance indicators have two major uses First
39. being met D They are needed to measure whether community goals are being met E They are needed to report to decision making bodies F They are needed to report to the public G Others please specify 2 How important are the following transit performance indicators Please circle the number on a 5 point scale from 1 Not Important to 5 Very Important Not Important Very Important A Avoidable accidents per vehicle hour vehicle mile 12 3 4 5 B Complaints per driver trip cceeeeeeeeeeeeeeeeees 12 3 4 5 C Expense per vehicle hour vehicle mile 1 2 3 4 5 D Fare box recovery rate 1 2 3 4 5 E Load Factores ee EG 1 2 3 4 5 F Passengers per vehicle hour vehicle mile 1 2 3 4 5 G Revenue per vehicle hour vehicle mile 1 2 3 4 5 H Others please specify 1 2 3 4 5 3 What do you feel are the 5 most important performance measures that your system uses Please list 5 measures they do not need to be in any particular order 4 What are your sources of data Please check all that apply L A Accident Reports L B Automatic Passenger Counter APC Systems L C Automatic Vehicle Location AVL Systems 63 D Complaints E Data collected in house or from other field surveys F Fare box data or other accounting reports G Others please specify 5 Did you collect it differently in the past C YES CINO
40. ct 1 Ridership e Received 107 responses 24 21 e Examples of measures included Average annual ridership Boardings per hour Passengers per mile or per hour or per trip e Received 70 responses 15 84 e Examples of measures included Cost per mile or per hour or per passenger Expense per mile hour passenger Operating cost per hour or per trip 3 Farebox Recovery Rate e Received 44 responses 9 95 e Examples of measures included Cost recovery Farebox recovery rate ratio standard 4 On Time Performance e Received 43 responses 9 73 e Examples of measures included On time performance rate percentage 5 Customer Satisfaction e Received 41 responses 9 28 e Examples of measures included Complaints per boarding or per driver or per mile or per passenger or per route or per trip Compliments comments Customer satisfaction 6 Accidents e Received 39 responses 8 82 e Examples of measures included Accidents per mile or per 100 000 miles Avoidable preventable accidents per mile or per 100 000 miles Table 8 and Figure 6 show the various sources of data and how popular they were among the respondents The most common data sources of data are farebox data or accounting reports like payroll data followed by passenger complaints data collected from the field 22 and accident reports About a third of the respondents indicated a use of AVL and APC data or data generated from some other technologies
41. e hour vehicle mile 0 0 7 Expense per vehicle hour vehicle mile 1 2 13 Avoidable accidents per vehicle hour vehicle mile 5 5 13 Fare box recovery rate 4 6 18 Complaints per driver trip 5 8 23 Revenue per vehicle hour vehicle mile 6 8 22 Load Factor 3 7 28 Others Maintenance measure On time performance measure Ridership measure Cost measure 1 Road calls measure 1 Connectivity Customer satisfaction measure Debt service coverage ratio Wheelchair passenger trips 4 45 30 31 29 31 38 44 i oot 5 46 52 44 39 31 24 15 myo BP AVERAGE 4 40 4 33 4 06 3 97 3 77 3 67 3 63 5 00 4 57 4 50 4 40 4 00 4 00 4 00 4 00 4 00 Q uestion How important are the following transit performance indicators Circle the number on a 5 point scale from 1 Not Important to 5 Very Important FIGURE 5 IMPORTANCE OF SPECIFIC MEASURES 5 00 4 50 Passengers per Expense per Avoidable Fare box recovery Complaints per vehicle vehicle accidents per rate driver trip hour vehicle mile hour vehicle mile vehicle hour vehicle mile Revenue per vehicle hour vehicle mile Load Factor The next question was aimed at determining the five most commonly used or most important performance indicators without pre specified indicators The question was What do you feel are the 5 most important performance measures that your sy
42. e or use the service TABLE 4 PERFORMANCE MEASURES OF ROUTE 50 Business Measures Passengers per hour 8 5 Cost per passenger 6 55 Public support cost annual 105 633 Social Measures Households in service area 3 191 Households without autos in service area 270 Public support cost per household 33 Public support cost per household w o autos 391 Source Capital District Transit Authority Albany N Y 14 Reilly Jack Improving Transit System Performance Using Information Based Strategies developed at the University of Wisconsin Milwaukee 1996 98 17 TABLE 5 RESIDENTIAL ROUTE ANALYSIS ROUTE 50 Overall Route 50 Route Service Area Key Indicators Service Percent Area Area of total Population density pop sq mi 4 105 2 116 Percent of households without autos 13 2 Percent of households with 0 1 auto 43 12 Percent of population over age 65 13 4 Households without autos per sq mi 415 26 Percentage of region s workers living in route service area using transit 6 0 2 Supplemental Indicators Total population 468 719 8 377 2 Total area sq mi 264 7 3 Total population over age 65 68 667 1 333 2 Total households without autos 30 310 270 0 9 Total workers 187 283 3 190 2 Population over age 65 per sq mi 575 345 Workers per sq mi 1 675 236 Source Capital District Transit Authority Albany N Y Problems with the performance measurement Assessing transit performance measurement
43. e timely even real time transit performance measures But these hurdles have to be overcome Procedures that are less technically complex should be provided to transit 39 agencies in order to make full use of modern technologies Technical training and funding are critically needed to support transit agencies particularly small transit agencies to produce better transit performance measure to improve transit services 40 PART II CASE STUDY 41 CASE STUDY USE GIS TO ANALYZE TRANSIT PERFORMANCE This section will illustrate the step by step process used to create a Geographic Information System GIS to evaluate the performance of a transit agency Once completed the GIS will provide the ability to measure a transit system s performance at the system level as well as at the route level To create a GIS an off the shelf GIS program is needed to perform the necessary functions and display your results This case study will use ArcGIS 9 0 to demonstrate the creation of a specialized GIS The guide can be applied to other GIS programs but there may be slight differences in the steps GIS Data for Performance Evaluation Specific data are necessary for transit performance measurement These data can come from a number of sources First there are data collected by an agency in house These data as the TCRP 88 report states can be calculated from information an agency would normally have on hand p 130 Examples of this type of da
44. ed with its organizational performance It can use performance measures to guide its actions and should also be aware of customer and community concerns The final viewpoint comes from the driver or vehicle Within it comes the need to gauge the interaction between transit and automobiles because the way they affect each other can determine the quality of transit service The report suggests that vehicle oriented measures taking into account the likely increased person occupancy on a bus should be addressed when studying the impacts of transit and autos on each other In a system as complex as transit hundreds of measures to assess performance could be developed However experience has shown a relatively small number of measures usually no more than 20 can be used effectively More than 20 can make it difficult for the end user to make an accurate judgment Considering the vast number of possible indicators it will be helpful to classify performance measures into a smaller set of categories In a study published in 1982 a classification of performance indicators was devised An abbreviated version of this classification is given below Financial Indicators Expense e Total Operating Expense Cost Total Passenger Trips This is a measure of how well the system is serving riders with available resources e Total Operating Cost Vehicle Miles or Vehicle Hours Measures of productivity of 10 TCRP 88 p7 1 Pennsylvania D
45. epartment of Transportation Rural Public Transportation Performance Evaluation Guide Bureau of Public Transit Harrisburg PA November 1982 pp 4 10 10 transit service provision which is useful in setting benchmarks or comparing services including the services of peer systems e Administrative Expenses Total Expenses A measure of the appropriate balance between these two expenses As a rule of thumb administrative costs should not exceed 15 20 percent Revenue e Total Revenue Total Passenger Trips A measure of the average revenue for a passenger trip e Total Fare Revenue Total Revenue An indicator of the percentage of revenue accounted for by fares e Revenue Expense Cost Also called operating ratio or cost recovery A measure of the degree to which operating expenses are covered by revenues Subsidy e Total Subsidy Total Vehicle Hours The average subsidy per vehicle hour of service e Total Subsidy Total Passenger Trips A measure of the average subsidy for every passenger trip Non Financial Indicators Ridership e Total Passenger Trips Total Vehicle Hours The average number of trips served per vehicle hour A measure of transit productivity e Total Passenger Trips Total vehicle Miles A productivity measure useful for comparing services especially in rural areas or on longer suburban routes e Elderly Passengers Total Passengers An indicator of the use of transit by elderly passengers M
46. er Air Distance F Buffer Walking Distance dO Se eI MMA eA OPop lob i a Start TLOS Values TCOSM Values Mins in last 7 080 Last Query Help Time a 5 foo a al Monday Pass Pop J En DADOTATOTOO CO a i Service c rs a afio 59 PM SfMond Al a p pet Total p uzsa 623434 Time 2910 159 LIP SyMondsy DE r Service Ka Ci G Coverage Meta an Air Buttes Wak Buttes A Group i Route s bs Service 7s ate Calculate TLOS Help Ga Los Astr A Si DY DW Microso Word Bto7 c04 07 gt CD Piave S TLOS MAP E Calendar PTS43IA C ESI 201 PM 15 In other cases the accessibility measure can identify the concentration of the captive riders For example Table 3 along with Figure 3 shows the system wide performance measures by service period The table and figure show that although the transit service territory includes only 4 of the transit district fully 73 of the households without autos are within 0 25 miles of a transit route TABLE 3 PROPORTION OF POPULATION IN SERVICE AREA BY SERVICE PERIOD CHARACTERISTIC TOTAL PEAK MIDDAY EVENING SATURDAY SUNDAY TOTAL POPULATION 100 42 39 27 37 23 TOTAL HOUSEHOLDS 100 44 42 29 39 24 ELDERLY POPULATION 100 47 45 30 42 24 EMPLOYMENT 100 62 62 50 60 46 HOUSEHOLDS 0 CARS 100 73 71 58 70 52 HOUSEHOLDS 0
47. er s attribute table so you can for example symbolize the layer s features using this data What do you want to join to this layer Join attributes from a table v 1 Choose the field in this layer that the join will be based on x 2 Choose the table to join to this layer or load the table from disk lus Census Block Group 5 V Show the attribute tables of layers in this list 3 Choose the field in the table to base the join on 7 Advanced About Joining Data Cancel Open the new joined attribute table which is under the block group polygon layer It will contain all of the original attributes from the original attribute table plus it now has all of the socio demographic information from the Census block group data Step 4 Exporting Layer as New Feature Class Next this layer with all of its changes will be exported as a new feature class This will make the join permanent Until this is done it is possible to remove the join which will return the block group polygon layer s table to its original state Right click on your block group polygon layer select Data and then select Export Data Click the Browse button 2I see Figure 20 to select where the feature class should be saved Change the Save as Type to Personal Geodatabase feature classes Change the name of the layer I will use New_Block_Groups Click Save then click OK 48 FIGURE 20 EXPORT D
48. essment A conclusion of the report will summarize our findings and provide a guide to further studies would be helpful to advance the understanding of this topic TRANSIT PERFORMANCE MEASURES Transit performance is very important to an agency in order to have a system that is as efficient and attractive as possible Performance can be measured by numerous indicators A performance indicator is a measure usually quantitative which reveals information about certain characteristics of a service Sometimes the measure is a ratio of two other measures For example miles per hour is a measure of the average vehicle bus miles traveled per hour It is composed of total vehicle miles traveled divided by the total hours of operation in travel With careful precise measurement and analysis these performance indicators can reveal the level of effectiveness and efficiency of the transit system The purposes of measuring transit performance are abundant Robert P Schmitt in his Service Evaluation and Monitoring s gives the goals of transit performance measurement There are six of them according to Schmitt To control costs and ensure the integrity of the system To justify changes in service levels To maintain or improve the quality of service To monitor subcontractors To guide marketing efforts and To report the status of transit service performance to policy boards at od nah he The Transit Cooperative Research Program has cre
49. feature where there is overlap and retain all areas from the two layers that do not overlap In an Intersect operation only the areas that overlap will be preserved For this procedure Intersect will be used In Arc Toolbox select Analysis Tools select Overlay then Intersect For Input Features select both the Stops_Buffer layer and the block group polygon layer Change the name of the Output Feature Class to Stops_Buffer_Intersec Select the defaults for Join Attributes and Cluster Tolerance 50 For Output Type make sure that Input is selected Figure 22 shows what the dialog box should look like Select OK FIGURE 22 INTERSECT DIALOG BOX Intersect DEK Input Features e Features amp Stops_Buffer 2 Block_Groups le l gt LX Le Output Feature Class AGIS Instructions Layer Examples Stops_Buffer_Intersect shp us Joinattributes optional ALL v Cluster Tolerance optional Feet Gad Output Type optional INPUT OK Cancel Environments Show Help gt gt Once the process concludes you will be able to see the results of the Intersect if you zoom in The gray areas below in Figure 23_show the areas of overlap or intersection 51 FIGURE 23 INTERSECT RESULTS ZOOMED IN Untitled ArcMap Arcinfo File Edit View Insert Selection Tools Window Help Deas x 2 ra AL OBO RAKTAI Prk oes pozz aa x x ag EB ArcT
50. g reporting requirements The most commonly used transit performance indicators are ridership and cost related measures particularly passengers per vehicle hour vehicle mile and expense per vehicle hour vehicle mile Other important measures include farebox recovery on time performance customer satisfaction and accidents Furthermore the survey indicates that most current methods of data collection and analysis have problems About one third of transit agencies have started to change the way they collect performance data moved away from manual methods to more automated technology based methods and started to use higher level technologies such AVL and APC for data collection Use of GIS to Measure Transit Performance The questions in Part II of the survey were aimed at determining the use of Geographic Information Systems GIS for transit performance assessment and to find out how transit agencies perceive the benefits of GIS The reason that we want to focus on the use of GIS is because GIS has a high potential for use to integrate analyze and report transit performance efficiently and effectively We wanted to know if GIS is readily accessible to transit agencies and if accessible is it put to use Also how do transit agencies perceive it as a transit performance assessment tool This relates to performance data collection as well as to the use of GIS to analyze or report that data The difference in the use of GIS at the system and ro
51. ght blue to show that it is selected You can right click on the Routes layer and choose Selection and Zoom To Selected Features to show the route zoomed in Route 1 for Waukesha County is displayed in Figure 27 55 FIGURE 27 ROUTE 1 SELECTED AND ZOOMED IN a amp Transit Example mxd ArcMap ArcInfo Ele Edt view Insert Selection Tools Window Help Dmg B3 X a amp amp 1398566 MIRAE fal RQA PEVE KERERE E r Tal Pt ag a Stops Routes CO Stops_Buffer_Intersect 8 Block_Groups o Display Selection Drawing RO BOF A v re o Aria Amas z ulA amp by ee 2478474 78 364858 98 Feet The selected route should be exported as its own layer To do this right click on the Routes layer and choose Data then Export Data You want to Export Selected Features and Use the same coordinate system as this layer s data source Click the Browse button to choose the folder to save in choose Save as type Shapefile and rename it Route_1 or your route number Click Save and OK Click Yes when asked if you want to add the new exported data to the map as a layer When the Route_1 layer has been added right click on the Routes layer and choose Selection then Clear Selected Features Also uncheck the box next to the
52. he respondents as a barrier 28 TABLE 14 WHY GIS IS NOT USED TO ANALYZE OR REPORT DATA REASON FOR NOT USING GIS TO ANALYZE REPORT DATA COUNT PERCENT RANK Lack of personnel to implement maintain software program 27 45 76 1 Others Technology software not available implemented yet 7 11 86 2 Software is not user friendly 6 10 17 3 Others No funds to support 6 10 17 3 Will in near future 5 8 47 5 Lack of data performance measures 3 5 08 6 Small agency 3 5 08 6 Question If no why do you not use GIS to analyze or report performance data Check all that apply FIGURE 12 WHY GIS IS NOT USED TO ANALYZE OR REPORT DATA 30 25 20 Lack of personnel to Software is not user No funds to support Will in near future Lack of Small agency implement maintain Technology software friendly data performance software program not measures available implemented yet The survey also asked the importance of GIS in evaluating transit performance at the system level and the route level The respondents who use GIS indicated that the technology is more important at the route or route segment analysis level than at the system level Table 15 shows these results 29 TABLE 15 IMPORTANCE OF GIS AT DIFFERENT LEVELS LEVEL 1 2 3 4 5 AVERAGE System 3 2 12 16 11 3 68 Route Route Segment 1 1 7 13 21 4 21 Question How important is GIS in measuring performance at the following levels Note There was 1 no score for Route
53. hem around and resize them to your liking Step 14 Exporting Map as A PDF File The final step to creating a map document is to export itas a PDF A PDF does not retain the functionality of an ArcMap mx file To export as a PDF go to File then Export Map When the dialog box pops up locate the folder you want to save in and change the Save as type dropdown to PDF pdf You can name the file the same name as your mxd file for that measure but make sure you do not change the pdf at the end of the name Click Save Do this for each map You can then view them by opening them in Acrobat Reader 61 REFERENCES Ausman Jon TLOS Demonstration Urban Transportation Planning Short Course Tampa Florida March 2002 Beimborn Edward A and Peng Zhong Ren Internet GIS and Its Applications in Transportation TR News March April 1998 pp 1 10 Beimborn Edward A Greenwald Michael J Jin Xia Transit Accessibility and Connectivity Impacts on Transit Choice and Captivity Transportation Research Board Record March 2005 pp 1 18 Bertini Robert L and El Geneidy Anmed Using Archived Data to Generate Transit Performance Measures Transportation Research Board November 2002 pp 1 24 Florida Department of Transportation Transit Level of Service Software Users Guide version 3 0 2001 http www 1 myflorida com transit Pages transitlevelofservicesoftware htm Accessed July 22 2002 Kurt Car
54. igure 13 30 TABLE 16 AVL PRODUCED PERFORMANCE INDICATORS PERFORMANCE INDICATORS FROM AVL COUNT PERCENT RANK On time performance 26 96 30 1 Percent of route deviation 11 40 74 2 Average dwell time at stops 8 29 63 3 Response time to accidents or breakdowns 6 22 22 4 Others Average Speed 1 3 70 5 Question If yes what performance indicators does AVL produce Check all that apply FIGURE 13 AVL PRODUCED PERFORMANCE INDICATORS 30 4 25 4 20 4 On time performance Percent of route deviation Average dwell time at stops Response time to accidents Average Speed or breakdowns The survey also asked if Automatic Passenger Counters APCs can be used to measure transit performance A vast majority of agencies 93 believe it can be Two follow up questions were then asked The first was of those who said yes and asked if they currently use APC for the purpose of transit performance measurement It was found that almost two thirds currently do not use APC for assessing their performance The 32 agencies which do currently use APC for transit performance assessment were asked what performance indicators APC has produced The most commonly used performance indicators derived from APC include passengers per stop 66 load factors 50 and passengers per vehicle mile or hour 47 These indicators are found in Table 17 as well as in Figure 14 31 TABLE 17 APC PRODUCED PERFORMANCE INDICATORS PERFORMA
55. ional report to the public measure whether improve service making bodies agency goals are Transit Database community goals being met are being met The results from Table 6 show that transit performance measures are mainly used to improve services and assess agency goals The need to report to the national transit database and to the public is also important but to a lesser degree Over four fifths of the agencies use transit performance measurement for business purposes while social measures are used by about half of the respondents To obtain the most commonly used transit performance measures we asked two questions The first question asks the respondents to rank the 7 given performance indicators including an open ended other factor The respondents were asked to assess the transit performance indicators on a scale from 1 Not Important to 5 Very Important The scores obtained from each agency responding to the question were totaled and divided by the total number of responses to get the average score for each measure Looking at Table 7 and Figure 5 the average scores indicate that the most important measures are passengers per vehicle hour vehicle mile 4 40 and Expense per vehicle hour vehicle mile 4 33 The lowest two scores were Revenue per vehicle hour vehicle mile 3 67 and Load Factor 3 63 All measures scored relatively high 20 TABLE 7 IMPORTANCE OF SPECIFIC MEASURES MEASURE 1 2 3 Passengers per vehicl
56. ionship of streets rivers railroads etc to each other and to numerous geographic entities which the Census Bureau uses Primary features are street segments with address ranges along them Found at www census gov geo www tiger To make use of TIGER Line data creation of maps etc one must have mapping or GIS software that can import the data TIGER Line does not include demographic statistics 45 Step 2 Display Data Using GIS Programs In this case study ArcMap was used The specific process is as follows To begin open ArcMap When the opening ArcMap dialog box pops up make sure that A New Empty Map is chosen Then check the box at the bottom that says Immediately Add Data Click OK This will bring up the Add Data Dialog Box you may also click the Add Data button to bring up this dialog box From here navigate to where your vector data the points lines and polygons are being stored You may need to click the Connect to Folders icon if you cannot navigate to your data from the original Add Data Dialog Box Highlight all the necessary files using the Control or Shift keys and click Add The files should show up under Layers to the left side of your map display and they should be visually present on the map display itself Next repeat the above process to add your U S Census Block Group data This data can be seen by clicking the Source tab on the lower
57. l E and Li Qiang GIS Based Itinerary Planning System for Multimodal and Fixed Route Transit Network Mid Continent Transportation Symposium 2000 Proceedings pp 47 50 Schmitt Robert P Service Evaluation and Monitoring Strathman James G Dueker Kenneth J Kimpel Thomas Service Reliability Impacts of Computer Aided Dispatching and Automatic Vehicle Location Technology A Tri Met Case Study Center for Urban Studies Tri Met December 1999 pp 1 27 Sutton John C Transit Cooperative Research Program Synthesis 55 Geographic Information Systems Applications in Transit FTA Transportation Research Board 2004 60 pages Transit Cooperative Research Program Report 88 A Guidebook for Developing a Transit Performance Measurement System FTA Transportation Research Board 2003 368 pages Transit Cooperative Research Program Report 100 Transit Capacity and Quality of Service Manual Second Edition FTA Transportation Research Board 2003 622 pages 62 APPENDIX 1 SURVEY FORM USED IN PROJECT PART I MEASURING TRANSIT PERFORMANCE 1 Do you use transit performance measures in your service and operation planning and evaluations C YES C NO If no please skip to question 8 gt If yes why do you use transit performance measures Please check all that apply A They are required for the National Transit Database B They are useful to maintain and improve service C They are needed to measure whether agency goals are
58. ld you want the variable normalized by we will use Total Households By doing this we have created a measure of Households with No Vehicles per Total Households As previously discussed in step 7 there are a number of different ways to format your categories Once you settle on a format your Households with No Vehicles per Total Households will display with the symbology you have chosen This step can be repeated for all the route level performance measures that you wish to map out It is recommended that you save a new mxd file for each variable 60 Step 13 Making a Map Document with the Layout View The next step is to make a map document of each system wide and route level performance measure you have used Go to the View menu and select Layout View instead of Data View You can now add the proper items to your map document Title Legend North Arrow Scale Bar etc To do this first and select Layout View From here there are an unlimited number of ways to arrange the items of your map document You can resize your map image by clicking one of the blue squares along the outside of the image and you can move the image by clicking and dragging it Make sure you keep what you want displayed in the map document inside the outlined box To add items to your map document use the Insert menu To add a title go to the Insert menu and select Title Type in what you want the title to be and press
59. left portion of your screen At this point your screen should look similar to Figure 18 FIGURE 18 LAYERS ADDED TO ARCMAP 6 Untied Arcan are FRIES Ele Edt view Insert Selection Tools Window Help DSS t Baxloals fPiss AL MQOolwr QQuxuTOes gt PROMS aeoe e e a Ble E se ag Layers f S SB Z users eric data minal eRe 7 5 Wauk_Stops LS k K Ee H a Wauk_Routes S ie E block_groups_data ina Ea y 4 E Z lusers eric data Wauk r t 8 Wauk_Block_Group T So 5 met ees mr i al oa N Beat i Lae Sa es me Se j j j 2 it ey lee a ee lL JE rami aS z Ji sr a pF Ss Ai E l u D x JEA LA aaah a 2 T E i iS Ry e a ee a ae be Ls ae y h j q i il m o aN Z ere a ae va ame a oes BE wn pe ae eae dea AAE l Ce E E J E p l S l a A D a Re We Ls 2 L ue AE A Ta ka E yA x AC A A ey ri E l f aa i AA A aa ay L Wy B l S fi j ae g K p e Display Souce Selecion fg nj ow gt f prewing gt e O E Oly Ay E e ava wife pz u aol 2369641 17 304664 42 Feet The map should be saved to the same location as the data that was added This map file will have an mxd extension 46 For each layer the
60. location of geographic areas defining census block groups U S Census Bureau Block Group Data that includes demographic data for residents within each block group Orthophotos which are aerial photos that have been geometrically corrected for relief displacement of the topography They have all the image qualities of a photo but have all the geometric qualities of a map FTA Bus GIS Database A GIS line database of the streets of the county or counties where the service is located from Bureau of the Census 1992 Enhanced TIGER files A condensed file of the TIGER based street network that takes up less data storage and provides quicker access and display A GIS network built from the county street database A GIS route system of the bus routes built from selected TIGER street segments as indicated on the transit agencies route maps and schedules All of the above GIS data products are condensed into a self executable zip file for archiving and data transfer FTA Level of Service LOS Information Days of the week of service Start and end time of weekday service Frequency of service for Morning peak Mid day peak Afternoon peak Evening Information on Saturday and Sunday service if available Information on accessibility of the route e g dedicated lift equipped buses for all or part of the day Description of fares if available 44 Table
61. of GIS to Measure Transit Performance ccccccccessssecceeeseeeensnsesseeeeeeees 26 Use of Other Technologies for Transit Performance Measurement 30 SUMMARY AND CONCLUSIONS is csscssszniscuanconndnasesenisaneiwas sanvateasbeadslastinadenncoeiediiauutlennns 39 PART IIZCASE STUDY ernennen e aa a a a aie 41 USE GIS TO ANALYZE TRANSIT PERFORMANCE cccccccsccssccssessceseeeseceseeseens 42 GIS Data for Performance Evaluation cccccccccccceseeeesssnseeetesssssessssnnseesesesees 42 GIS Analysis for Performance Measures A Case Study of the Waukesha Metro Transit System ss socsnccsansavatsanasehpeessundteaydnesndansosncerenes ndenenaase pvsteanadedvndesucsmsrsensnaccavedaopes 43 Step 1 Gather the Necessary Data ccccccsccecsesseeeeeeenneeeeeeeaeeeeseeeeeeenensenenees 43 Step 2 Display Data Using GIS Programs cccseseseeeceeeeeesensnneeeseceeeeesentnaees 46 Step 3 Join Tables oc ssicscaisesrisesosaseecnnziiavascnovaciseracinnsadaue sade tenhacsassd dabndeasveiadunseaveicaave 47 Step 4 Exporting Layer as New Feature ClaSS cccccccccsssssscceceeeeesnsntsssaeees 48 Step 5 Buffering BUS Stops 0 cccccccccecssessseeeeceeesenseaneeseeseesssecesssasesseseseeeseniaaes 49 Step 6 Overlaying Bus Stops Buffer with Census Block Group Daia 50 Step 7 System wide Transit Service ANA lYSiS ccccccccccsssssecceecseesensnnsseseees
62. oolbox Stops 30 Analyst Tools Analysis Tools A Routes amp Extract amp Overlay Stops_Buffer_Intersect A Erase o identity Stops_Buffer A Intersect o Symmetrical Differens 5 Block_Groups A Union oO AP Update a amp Proximity amp Statistics Cartography Tools Conversion Tools w Coverage Tools Data Interoperability Tools Data Management Tools Geocoding Tools H Geostatistical Analyst Tools Linear Referencing Tools Network Analyst Tools Samples Spatial Analyst Tools E Spatial Statistics Tools lt 5 Display Source Selection Favorites Index Search en r E E Drawing gt A y AY I0 Arial Tit se Bz ulAy Oy wy 2 2470839 63 366537 45 Feet Step 7 System wide Transit Service Analysis Once step 6 is complete we can analyze the transit service catchment area based on the socio demographic data from the U S Census Bureau To do this begin by right clicking on the Stops _Buffer_Intersect layer and choosing Properties Then select the Symbology tab and choose Quantities and Graduated Colors Under Fields in the Symbology tab click the dropdown arrow for Value and scroll down until you find the first variable you wish to use we will use 2000 Population
63. r Systems AVL and Automatic Passenger Counters APC These technologies relate to the questions from Part that dealt with the sources of performance data and how it is collected Responses in Part indicated that past collection methods were done manually and or with less advanced technologies In Part Ill we focused more on the use of technologies to improve and measure performance Close to half of the survey respondents 43 3 have an AVL system available Among those who have AVL technology over three fourths reported that AVL has improved on time performance Among all survey respondents regardless of whether they have AVL or not almost all of them 98 responded that AVL can be used to measure transit performance Those who felt it could were then asked if they currently use AVL for that purpose Surprisingly almost three fourths of respondents said they did not currently use AVL to measure transit performance Even for those 45 agencies that have AVL only 27 of them currently use AVL data to measure transit performance The 27 agencies that have AVL were then asked what performance indicators were measured using AVL data Nearly all 96 of the respondents replied that AVL was used to produce on time performance About 40 said they used AVL to determine route deviation Other indicators that could be derived from AVL include average dwell time at stops and response time to accidents or breakdowns as demonstrated in Table 16 and F
64. r use GPS technology is the main reason for not using GIS or GPS to collect data The next two reasons are both 27 monetary The agency either does not have a budget for software programs or the technologies are too costly for the agency It should be noted that 8 of these agencies indicate that they will be using them soon Table 13 and Figure 11 show these results TABLE 13 WHY GIS OR GPS ARE NOT USED TO COLLECT DATA REASON FOR NO GIS GPS TO COLLECT DATA COUNT PERCENT RANK Lack of personnel to implement maintain software program 21 45 65 1 No budget for software programs 16 34 78 2 Too costly 14 30 43 3 Others Will have in near future 8 17 39 4 Not necessary in a small transit agency 7 15 22 5 Software is not user friendly 3 6 52 6 Question If no why do you not use GIS and or GPS to collect data Check all that apply FIGURE 11 WHY GIS OR GPS ARE NOT USED TO COLLECT DATA Lack of personnel to No budget for software Too costly Will have in near Not necessary in a Software is not user implement maintain programs future small transit agency friendly software program 25 20 Among all respondents more than half 57 reported that they did not use GIS to analyze or report transit performance data The 59 who did not use GIS for analysis or reporting were then asked why not These results are shown in Table 14 and Figure 12 Lack of personnel was again cited by about half of t
65. ract gt amp Overlay eS B A Route_1 P Erase a a 7 l Identity Se lees O Routes A Intersect PSS aa s A Symmetrical Differen M Route_1_Stops_Buffer_ A Union Update O Route_1_stops_Buffer Proximity a H Statistics S CO Stops_Buffer_Intersect E Cartography Tools m E9 Conversion Tools 5 Block_Groups E Coverage Tools o E Data Interoperability Tools E Data Management Tools E Geocoding Tools E Geostatistical Analyst Tools H Linear Referencing Tools E Network Analyst Tools H Samples 41 4 Spatial Analyst Tools E Spatial Statistics Tools a Display Selection Drawing e O Bl Oy A E o Aria 2471626 65 360905 15 Feet Step 12 Route Level Transit Service Analysis Once step 11 is complete we can analyze the transit service catchment area based on the socio demographic data from the U S Census Bureau To do this begin by right clicking on the Route_1_Stops_Buffer_Intersect layer and choosing Properties Then select the Symbology tab and choose Quantities and Graduated Colors Under Fields in the Symbology tab click the dropdown arrow for Value and scroll down until you find the first variable you wish to use we will use Households with No Vehicles for our first example Then for the Normalization dropdown choose the fie
66. rce utilization measures such as customer boardings per revenue mile or revenue hour Other measures of some interest in this area include revenue to cost ratio or cost per customer transported In fact several transit systems have some type of service standards against which actual performance levels are measured Providing cost information below the route level requires the use of some type of cost allocation model This can either be a short run avoidable cost model to determine the short run impact of changes in the level of service or fully allocated costs in which all system costs including short run avoidable costs and fixed costs are pro rated among routes Albert Gan Ike Ubaka et all FTIS Florida Transit Information System Version 2004 Developed for the Public Transit Office Florida Department off Transportation by the Lehman Center for Transportation Research Florida International University http Ictr eng fiu edu ftis FTGIS htm accessed 8 16 06 12 By using farebox data for example one can develop business measures for routes in the system The table below shows a set of measures for Saturday routes in a transit system TABLE 1 EXAMPLE OF ROUTE COMPARISON CAPITAL DISTRICT TRANSPORTATION AUTHORITY KEY INDICATOR SUMMARY Saturday Total Period May Route Revenue Passengers Margin Cost Hour Passenger 2 West Albany 26 14 2 1 60 4 Pine Hills 20 12 8 1 98 3 Quail Street 37 21 4 1 23 8 Arbor Hill 14 10
67. re is an attribute table that has data associated with the layer Right click on a layer and choose Open Attribute Table You can scroll up down and left right to see each Field column and Record row Do this for each of the layers to become familiar with the data represented within each layer Step 3 Join Tables Data from the U S Census block group data table should be joined to the block groups polygon layer so that the data can be shown spatially The data table and the polygon layer will need to have matching unique identifier fields in each of their respective attribute tables for this to work This field will be the basis for joining the table to the layer To join the tables right click on the polygon layer select Joins and Relates and select Join This will load the Join Data dialog box as shown in Figure 19 Under the first drop down menu you want Join attributes from a table Next choose the unique identifier field in the block group polygon layer that the join will be based on Then select the Census block group data as the table to join Finally select the unique identifier field in the Census block group table as the field to join Click OK Your tables will now be joined If you receive a warning concerning indexing click ok and continue don t worry if you don t get that warning 47 FIGURE 19 JOIN DATA DIALOG BOX Join Data Join lets you append additional data to this lay
68. rtant reasons are the lack of standard and easy to use procedures the lack of user friendly software and the lack of technical personnel Does Agency Size Make Any Difference One would suspect that responses to the survey would vary by size of the transit agencies Does the agency s size affect what technologies are used by the agency Does the agency s size affect what performance measures are used by the agency These are two important questions when discussing transit performance assessment and they can be addressed using statistical methods To find the answers to these questions we used the agency information at the National Transit Database NTD to determine the agency size The measure of agency size used was vehicles operated in maximum service VOMS This measure according to the NTD definition is the revenue vehicle count taken during a transit agency s maximum season of the year on the day of the week that this maximum occurs It is a useful measure of agency size because it reflects the level of operation of an agency Using VOMS we split the responding agencies into three groups small medium and large Tables 19 and 20 and Figures 16a c were then created by running cross tabulations between each technology question and the agency size classifications Table 19 contains the count data which show how many respondents in each size category responded in each way to each question while Figures 16a c represent these data
69. s using these indicators easily and accurately is important But measuring these indicators is not an easy job particularly the social indicators Traditionally general fare box and payroll data are used to measure business performance indicators at the system level There are a number of problems with the current methods of measuring transit performance As Bertini and El Geneidy 2002 observed that it has been difficult and costly to collect comprehensive performance data Therefore transit agencies have had to use limited general aggregate measures to assess transit performance Furthermore each agency usually has to create its own way of measuring and implementing a system to assess the performance of their agency Transit agencies may benefit by use of advanced technologies to simplify data collection and produce comprehensive reports on their performance Technologies like GIS automatic vehicle location systems AVL and automatic passenger counters APC appear to have promise as ways to improve the measurement of transit performance A key question for this report is are transit agencies taking advantages of these technologies This report explores the use of these technologies and how it relates to the size of the system and other factors The next section will report a survey about the actual usage of transit performance in transit agencies in the United States 18 SURVEY OF TRANSIT PERFORMANCE MEASUREMENT IN TRANSIT AG
70. ssibility and availability of service based on actual street networks walk buffers and transit schedules rather than an air distance buffer The software uses GIS to determine if transit service is in fact available to users at different times of the day and week at any location within the system s service area Work with the TLOS system has indicated that far fewer individuals have adequate transit service to meet their travel needs than had been 13 previously thought Figures 1 and 2 compare the use of air distance buffers to the use of walk buffers for the Tallahassee Transit System TLOS uses two distances from residence and from job site to determine the accessibility of a person to a transit stop According to the data used in the TLOS software the difference between the use of air distance buffers and the use of walk buffers was quite significant For residences the population served for the air distance buffer was about 19 greater than that served based on actual walk distance For accessibility to job sites that difference was over 75 Table 2 shows the difference between the two methods for all routes and times available in the TLOS software for the Tallahassee Transit System TABLE 2 MEAN POPULATION AND JOBS SERVED BY AIR AND WALK BUFFERS TALLAHASSEE TRANSIT SYSTEM AIR DISTANCE BUFFER VS WALK BUFFER All routes and all times Buffer Type Population Mean Job Mean Air Distance 1 227 2 409 Walking Distance 699 2 02
71. stem uses In response to this we received 37 different measures with 442 total responses Respondents generally 21 provided specific measures which we grouped according to broader categories Each of the six types of measures listed is perceived to be important to transit performance evaluation Ridership measures were deemed most important as indicated by the number of responses and the percentage of total responses received These measures relate to the number of passengers that choose to use transit over other forms of transportation The cost measure second on the list is also an important indicator of performance especially when normalized by service mileages hours trips or passengers These measures indicate the expense required to provide transit services to transit users Altogether the results of this question indicate the measures that are viewed as most important to the nation s transit agencies These top two measures as well as farebox recovery rate are highly consistent with the previous question The importance of other measures is not consistent with the previous question For example the accidents indicator is ranked as the least important and commonly used indicator in this question but ranks third in the last question This may indicate that some transit agencies consider avoidable accident rates important but do not actually measure them One possible explanation is that the data required are difficult to colle
72. ta are schedule data maintenance records financial data fleet data and so on Another source is the National Transit Database NTD to which many agencies are required to submit information This database includes service area characteristics labor hours cost data fleet information amount of service provided amount of service consumed and many others TCRP 88 warns that although it is a good source to compare measures across agencies the NTD may have inconsistencies from the reporting of different agencies Other agencies may have demographic data traffic data and GIS data that they can provide Demographic data include population and household data usually from the U S Census Bureau or local planning authorities and employment data also from planning organizations or state employment departments Traffic data including daily traffic volumes traffic speeds sidewalk inventories traffic signal timing information and the number of lanes provided on streets 132 can be found at local public works departments or state departments of transportation GIS data are the demographic or traffic data with spatial location information that can be analyzed using GIS software The difficulty of collecting data in the past has been that it has required a great deal of labor which leads to a small number of trips being sampled because of the high costs The alternative to this is automated or semi automated data collection such as the
73. tation Association APTA were contacted via survey to determine the use of performance measures GIS and other technologies such as Automatic Vehicle Locators AVL and Automatic Passenger Counters APC The results of the survey revealed a distinct gap between agencies believing the technologies are useful and agencies actually implementing them The National Transit Database NTD provided agency size and revenue data which matched with the surveyed agencies demonstrated a sharp contrast in regards to an agency s technology use and its size or revenue generated One reason that may discourage agencies from implementing GIS and other technologies is a lack of procedures and methods for implementing GIS for performance evaluation This study also includes a step by step guide to using GIS for performance assessment at the system and route level of a transit agency The guide uses demographic data from the United States Census Bureau and geographic data from Waukesha Metro s transit system to give visual representations and provide examples of performance measures which can be analyzed with GIS ACKNOWLEDGMENTS We would like to thank the people at transit agencies around the country that helped us with the survey portion of this project The descriptive material about performance indicators was adapted from a NTI course Improving Transit System Performance Using Information Based Strategies developed at the University of Wisconsin
74. the buffer a quarter mile around each stop Finally for Dissolve Type choose ALL to dissolve all of the overlapping boundaries Click OK to begin the buffering process NOTE The time period required for the buffering process to complete depends on many factors such as computer speed and number of stops so it may take a few minutes 49 FIGURE 21 BUFFER DIALOG BOX Buffer Input Features Output Feature Class Distance value or field Linear unit Unknown C Field Side Type optional FULL End Type optional ROUND Dissolve Type optional NONE Dissolve_Field s optional aired Cancel Environments Show Help gt gt Once the process has concluded the transit service catchment area represented by the polygons of the stops buffer layer can be seen Step 6 Overlaying Bus Stops Buffer with Census Block Group Data In order to be able to relate socio demographic information to the transit service catchment area you must overlay the stops buffer layer with the new block group polygon layer Overlays identify the overlaps between features in two or more layers and create a new dataset based on where the overlap occurs There are two basic types of overlay Union and Intersect A Union will create a new layer with all features from the two layers whether or not they are overlapped For example if portions of two polygons overlap it will create a new
75. they may be employed to assess how well the system is doing with respect to the standards established by management Second they may be used to identify areas within the system that need attention or remedial action Both of these approaches come together in the transit system evaluation process Business vs Social Measures Both business and social measures of transit service can be performed at the system route or route segment and even stop level Most of the transit performance work to date using on board data has focused on business measures of service cost per customer transported etc The recent advent of geographic information systems GIS has enabled transit analysts to make a number of social measurements of transit service For example coupled with a description of the service stop lists frequencies etc an analyst can determine the proportion of households without autos served by daytime service the proportion of households with direct access by transit to shopping and the proportion of jobs in the transit service territory The social mobility measures when coupled with more traditional business measures of transit service revenue to cost ratio for example provide a more complete picture of transit performance in metropolitan areas Through the use of these technologies one can estimate the cost to achieve certain levels of mobility Business Measures The most important business measures to transit operators are resou
76. ts of our survey reinforce these reasons but in a different order Of the agencies responding to the survey nearly all of them utilize performance measures to assess their systems and of those almost all of them find the measures are important primarily because they help to maintain and improve the quality of the services that they provide Reporting to funding agencies and decision making bodies was also important but to a lesser degree in spite of the fact that they are required Of a lesser importance was the use of indicators for social purposes to report to the public and to relate to community goals This reflects the current status of performance measurement that is it is more important to measure transit performance for business purposes rather than for public service or social purposes TCRP 88 also includes an extensive list of performance measures that can be used The report acknowledges that different agencies may have different needs for their performance measures In our survey ridership and cost measures were found to be the two most popular measures but there was a wide range of specific measures that agencies found important and used This was the case for all transit agencies regardless of size Most agencies have had difficulty with collecting the data for the various performance measures This is consistent with Bertini and El Geneidy s 2002 finding that collecting comprehensive performance data has been difficult and expensi
77. urrently used to measure transit performance NO 19 17 8 YES 3 11 16 FIGURE 16a AGENCY SIZE AND TECHNOLOGY GIS 30 GIS accessible GIS GPS used to collect data GIS used to analyze report performance data B Small BMedium W Large 35 30 FIGURE 16b AGENCY SIZE AND TECHNOLOGY AVL AVL possessed AVL has improved on time AVL can be used to measure AVL is currently used to measure performance transit performance transit performance E Small BMedium W Large FIGURE 16c AGENCY SIZE AND TECHNOLOGY APC APC possessed APC has produced useful APC can be used to measure APC is currently used to measure information transit performance transit performance E Small Medium W Large 36 TABLE 20 AGENCY SIZE AND TECHNOLOGY ASSOCIATION Measure of Association Relating to Agency Size Question topic Chi square Significance Gamma Kendall s tau c Significance GIS accessible 7 51 0 023 0 501 0 268 0 004 GIS GPS used to collect data 5 53 0 063 0 418 0 264 0 016 GIS used to analyze report performance data 13 43 0 001 0 605 0 435 0 000 AVL possessed 14 40 0 001 0 597 0 430 0 000 AVL has improved on time performance 4 51 0 105 0 132 0 061 0 714 AVL can be used to measure transit performance 4 10 0 129 0 037 0 002 0 789 AVL is currently used to measure transit performance 11 30 0 004 0 585 0 348 0 001 AP
78. use of Automatic Vehicle Location AVL technology instead of manual collection AVL equipment according to TCRP 88 has two main functions They are to track the real time location of AVL equipped buses and to collect and store data about bus arrival and departure times 136 Automatic Passenger Counters APC are another form of technology to aid in data collection This can save on labor costs as well as increase the availability of ridership data at the system route and stop level Data collected with this method according to TCRP 88 include stop route and system level ridership maximum passenger loads and their locations passenger miles how long standing loads occur during a trip and how often loads exceed a pre determined level 138 In research by Beimborn Greenwald and Xia at UW Milwaukee a number of data sets related to transit performance are examined Their work Transit Accessibility and Connectivity Impacts on Transit Choice and Captivity describes the implementation of a Geographic Model for Bus Service Planning and Marketing Four data sets are required to integrate into the model The first set is geographic data which consists of a digital map of 42 census tracts by latitude and longitude The second set is transit service data which is a network of bus routes and frequencies Third is residential data consisting of details on households and residents The fourth and final set is worker dat
79. ute levels was also included in this part of the survey 26 The survey indicated that GIS technology is available at about two thirds of the agencies The major reasons for not having GIS are high cost lack of budget for software purchase and lack of technical personnel to operate GIS In addition some small transit agencies felt that GIS was not needed due to the size of their size Table 12 and Figure 10 show these results TABLE 12 REASONS WHY GIS IS NOT ACCESSIBLE REASON FOR NO GIS COUNT PERCENT RANK Too costly 14 40 00 1 No budget for software programs 13 37 14 2 Lack of personnel to implement maintain software program 12 34 29 3 Not necessary in a small transit agency 6 17 14 4 Others Will have in near future 5 14 29 5 Software not user friendly 2 5 71 6 Software not available 1 2 86 7 Question If no why do you not have a GIS software program that you can use Check all that apply FIGURE 10 REASONS WHY GIS IS NOT ACCESSIBLE 12 9 6 l J E Too costly No budget for eee of oe Not necessary ina Will have in near ee not user Software not software programs small transit agency future friendly available mieuna software program Over half 56 of transit agencies reported to use GIS or Global Position System GPS to collect transit performance data Of those who responded did not use the systems a majority responded that the lack of personnel to run a GIS program o
80. ve Agencies have had to rely on measures that are incomplete or aggregated which limits their usefulness It is interesting to see however although most transit agencies still rely on traditional methods to collect performance data there is some progress in shifting to more sophisticated technology oriented methods The advancement and wide implementation of technologies for transit services provides advanced tools to not only evaluate traditional transit performance at the system level but also create new transit performance e g social measures at the route route segment and even stop level However the results of the survey revealed a distinct gap between agencies believing the technologies are useful and agencies actually implementing them For example GIS is available to about two thirds but less than half use it to integrate analyze or report performance data Similarly although AVL can provide rich data for performance measurement at the fine grain level and transit systems consider AVL a great source of data for performance evaluation only 60 of those who have AVL actually use it The major barriers seem to be the lack of well defined procedures and methods of measuring performance the lack of well trained technical personnel and the lack of budget The expansion of advanced technologies like GIS AVL and APC provide great potential to improve the efficiency of evaluating transit performance and to create more accurate and mor

Download Pdf Manuals

Related Search

Related Contents

2015総合カタログ - ヒビノデータコム株式会社  PPS payment update provides a slight increase, but few  User Guide - Riverstone Schofields Memorial Club  EMCS Web service, user's guide  AX4E-Centos5.4-Dahdi-User Manual-V1.0-EN  Entretien sur la démocratie - Académie des sciences morales et  取扱説明書 24mm ハンマドリル  Unison Echo Relay Panel v1.0.0 User Manual  

Copyright © All rights reserved. Failed to retrieve file