USF Libraries
USF Digital Collections

Assessing level of service equally across modes


Material Information

Assessing level of service equally across modes
Physical Description:
54 leaves : ill. ; 28 cm.
Winters, Philip L
United States -- Dept. of Transportation
Florida -- Dept. of Transportation
University of South Florida -- Center for Urban Transportation Research
Center for Urban Transportation Research, University of South Florida
Available through the National Technical Information Service
Place of Publication:
Tampa, Fla
Springfield, Va
Publication Date:


Subjects / Keywords:
Urban transportation -- Florida   ( lcsh )
Local transit -- Florida   ( lcsh )
level of service   ( trt )
intermodal transportation   ( trt )
bibliography   ( marcgt )
technical report   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references (leaf 54).
Additional Physical Form:
Also available online.
Prepared for Fla. Dept. of Transportation under contract no.
Statement of Responsibility:
prepared by Center for Urban Transportation Research, University of South Florida ; Philip L. Winters ... et al..
General Note:
"December 2001."
General Note:
General Note:
White paper.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001927520
oclc - 48953525
usfldc doi - C01-00133
usfldc handle - c1.133
System ID:

This item is only available as the following downloads:

Full Text


Assessing Level of Service Equally Across Modes White Paper Prepared for Florida Department of Transportation Contract No. BC353 RPWO# l S Project Manager: Martin Guttenplan Prepared by Center for Urban Transportation Research College of Engineering University of South Florida Philip L. Winters Francis Cleland Edward Mierzejewski, Ph D., P.E. Lisa Tucker December 2001


TECHNlC\1. REPORTS'iANDARD 1'TJ'LE PAC 2. No, 3. No. 423-01 Tille efllttl Cocto 7. A.l.l!tor($) $ RtpQtt Nc, Philip L. Wint ers, Francis C lela nd, Edward Mierzejewski and Lisa Tucker 9. Pfonnln; 0rQ!IfiiZ&1cn Name andAcl::lrtM t 0. 'o\'o!t Unl! No. Cente r for Urban Transportation Research University of South Florida H Ccnnlcl 0t No. 4202 E. Fowler Ave. CUT100, Tampa, FL 33620-5375 BC353 RPW0#15 12. Agency tqmo an::l AdUcss 13. 1)?c ol R4f:01 an d PmcKI CO'\wl:d Florida Departme nt of Transportat ion 605 Suwanee Street Tallahassee, Florida 32399.0450 1 4 S9on&OIIn; A;encyCOOe 1 S.. Sl.lpp"emenl3ry Notes Supported by a gran t from the Florida Department of Transportation and the U.S Department of Transportation 18. Abmct The projec t 's focus was t o identify the feasibi l ity of and methods toward a L eve l of Service (LOS) system that can be assessed equally for the motor vehicle, bicycle. pedestrian, and trans it modes. Interviews conducted with key stakeho lders are used to Identify how LOS measures found they plan on continuing to use L OS meas ures to assess exis ting conditions identify roadways i n need of improve m ent and prioritize construc tio n projects. Some stakeho l ders want t o allow LOS trade -off s between modes and promote a balanced multi modal transportation system. The project advisory committee concluded that the c u rrent methods of de sc ribi ng levels of service are appropriate for thei r mode, understandable by b road audi ences, and profess i onally defensible. They also agreed thai a sys tem of measuring level of service equally across modes would be of significant value to policymakers, developers and the transportation industry. They applied a systemalic creat i ve thinking technique to identify and assess different concepts a n d approac h es that FOOT can use to develop a common LOS system. The comm ittee p r oposed a system that keeps the current LOS methods i n place but seeks to related each mode's LOS to the othe r modes' LOS by establis hing the re l ation to u ser needs as a common charac teri stic The advisory committee concluded that the identification of the common denominators across all the modes or users was necessary for a true me t hod of assess ing LOS equally across modes and permitt ing tr ade-offs across modes. The advisory committee strongl y recommended that FOOT seek to i ncorpo rate user perceptions to identify how t o respond to t he range of needs reflected by the var i ous LOS measures and dynamic needs of transportation u sers. In effect, the advisory committee recognized that t ransportation system users have a hierarchy of needs that are common t o all modes A subsequen t l iterature review demonstrated the importance of these psycholog ica l factors t o mode choice decisions and their omiss io n from mode c h o i ce models. The i mpli cation of the term "level of service" suggests t ha t we are concerne d wit h the appraisal of those existing and potentia l users of the variou s modes. It is clear that any method of comb ining l evel of service wi ll probably m is s the mar k if it fai ls t o take into account psycho logi ca l factors. An approach for a pilot project to i dentify, construct and appl y a Transporta t ion System User Hierarchy of Needs was proposed. 11. l<.e'/ Vltlrd' 18. Dkl:l'i:lvfO'I s-.ateMetv. Leve l of Service Th i s documen t is available t o the publ i c through the National Tec h nica l Information Service (NTIS), http://www.nt 19 (Of1hl:s.rei)C(I) 20 Secullty Clll$$if, (Of 11'11& Ptllt) 21. NO. Of Pill$$ u. Plloe Unclassified Unclassifie d Fonn DOT F 1700. 7 (3-69)


ACKNOWLEDGMENTS We woul d like to thank members of the advi sory committee for thei r active participation in this projec t : Dale Eacker, Flo rida Department of Community Affairs Gina Torres, Hillsborough County Metropolitan P lannin g Organization Mahdi Mansour, City of Tampa Taha Ataya, City ofTampa Martin Guttenplan, F l orida Departme n t of Transportatio n Doug McLeod, Florida Department of Transportation Bruce Landis, SCI Jonathan Paul, SCI DISCLAIMER The opinions, fmdings and conc lusions expressed in this publication are those of the authors who are responsible for the facts and accuracy of the data presented herein The conte n ts do not necessarily reflect the views or policies of the Florida Department of Transportat ion or the Federal Highway Administration. This report does not constitute a s tandar d s pec ification or regu lation. The report is prepared in cooperation with the State of Florida Departme n t of T ransporta t ion and the U S Department ofTransportation


Assessing Level of Service Equally Across Modes Introduction Transportation investments are influenced by level of service ratings of the current and expected system performance. According to the Highway Capacity Manual, Level of Service (LOS) is a .. quality measure describing operational conditions within a traffic stream, generally in terms of such service measures as speed and travel time, freedom to maneuver. traffic interruptions. and comfort and convenience ." LOS for a roadway facility or mode falls in to six letter grade levels with "A" desc ribing the highest quality and "F" describing the lowest. LOS for automobiles, transit, pedestrian and bicycle modes are based on a variety o f cri teria and, therefore, calculated on a d ifferent basis. Fo r automobiles LOS is measured using average stopped delay for intersections, average speed f or arterials and density for freeway segments. Automobile LOS "F" implies traffic is at a near standstill. For bicycles, LOS is a function of the typical roadway conditions, bicycle facilities, and safety perceived by users. U nlike automobiles where LOS "F" rep resents too many users, bicycle LOS "F" is just the opposite-only those w h o a bsol utely have to bike will do so, probably due to safety concerns or lac k of facilities all together. LOS for pedestrians is similar to that for bicyclists. Fo r fixed route transit, LOS measures access to transit routes based on po pulation within wal king distance to bus routes and service frequency. Unfortunately, the current classification schemes make the total transportation system performance and multimodal tradeoff decisions difficult to assess. For automobile travel, most users would consider LOS ''D" or "E" conditions as satisfactory. However, LOS 11D" or "E" for bicyclists is poor enough to deter all b ut skilled b i cyclists or those with no other mode choice from making the trip. Furthermore, the measures don't reflec t expectations (i. e., do trave lers interpret LOS "D" the same under all conditions?) or system reliability/volatility (i.e., how does LOS fluctuate over time? How sensitive is t he syste m to disruption?). Thus i nvestm ents are not optimized to provide the efficient and balanced transportation system that communities envision. These concerns were expressed in Florida House Bill 17 of t h e 2000 Legislative session, ref erred to as the Ur ban Infill and Redevelopment Act o r Growth Policy Act (GPA). This l egislation was I


amended to include the development of alternate LOS techniques to measure multi-modal p erformance. During the past several years, the Florida Department of Transportation (FDOT) has taken a leadersh ip role nationally in the development of analytical tools and processes to assist local governments and Metropolitan Planning Organizations (MPOs) in understanding facility, corridor and system levels of service for these alternative modes. More specifically, FOOT has funded the following activities: A national literature search of multimodallevel of service methodologies to determine o the best approach for Florida and o an areawide application. The application and validation of a methodology for assessing both bicycle and pedestrian quality of service and to measure the performance of roadway segments in all FOOT districts. The application and testing of the new Highway Capacity Manual (HCM) performance measures for transit in all FDOT districts The deve lopmen t of areawide multimodallevel of measures to be used for the evaluation of multimodal transportation districts. The develo pment of the Transit Level of Service software for detailed operational analysis of transit availability. Expert Review of FOOT's multimodallevel o f service measures by members ofTRB's Highway Capacity and Quality of Service Committee. Test applications for corridor multi-modal LOS techniques. The primary focus of this research is to assess the need for the deve l opment of a Level of Service (LOS) system that can be assessed equally for the motor vehicle bicycle, pedestrian, and transit modes. If a need for a common LOS system is i dentified, are there methodologies that FOOT can use to develop a common LOS system? In addition to the development of a co mmon LOS system, this re search is also examin ed how stakeholders currently use LOS measu r es to make decisions and how they would like to use them to make decision in the future. 2


Method The approach used to address these objectives were: (I) to gather information known about bow the automobile, b icyc l e, pedestrian, and transit modes are measured; (2) to assess how stakeholders use tbe measures and/or want to use the measures to make decisions; (3) to generate alternatives for applying or aligning current or proposed LOS measurement systems through a series of brainstorming/creative thinking sessions v.'ith an advisory committee of transportation e xperts; ( 4) to have tbe advisory committee identify the values, feasib i lity and benefi ts of each o f these alternatives; (S) have the ad visory commit tee identify the difficulties and cautions as sociated with each of the a l ternatives; and (6) recommend (an) approach(es) to FDOT for a pilo t study. Literature Review and Interviews The first two tasks were to gather informat i on about how the current systems are measured and assess how stakeholders use the measures to make decisions. Under the se tasks, the project team reviewed FOOT's Multimodal Level o f Service literature searches performed under FOOT's contracting with University of Florida (UF) and Reynolds Smith &Hill (RS&H); contacted agencies utilizing multimodal LOS analysis techniques; and assessed how stakeholders want to use the measures to make d ecisions in the future. Interviewees were queried about their use of multimodal LOS analysis techniques to assess the issues surrounding multi modal LOS tradeoffs in decision -making. The highlights from these ta sks follow. A sep a ra te report on t hese t asks was prepared The n eed for a LOS sca le th at ca n be assesse d equally across modes stems from the f act that the tenn LOS" has largely become associated v.'ith assessing tbe capacity of roadways for motor vehicle s. However, the resu lts of this report illustrat e t hat while capacity (based on dela y and speed) is the primary LOS evaluation measu re for motor vehicle s the frequency o f se rvice i s the primary evalua ti on measure for assessing transit, while safety /comfort are the primary evaluation measures for the bicycle and pedestrian modes. Thus, the deve lopment of a LOS score that can be applied equally across modes may not be practical without normalizing these scal es in some manner. Furthermore. interviewed stakeho1ders indicated that there is not a problem with using the LOS t ermino logy and sca l e to describe the quality of service for the transit, bike, and p edestr ian modes. 3


The other objective of this project is to assess how stakeholders currently use LOS measures to make decisions for each mode and how they want to utilize the measures to make decisions in the future. Stakeholders plan on continuing to use LOS measures to assess existing conditions, identify roadways in need of improvement, and prioritize construction projects. Some stakeholders also indicated that they would like to use the measures to allow LOS trade-offs between modes. In addition, stakeholders believe that multi-modal LOS measures should be used to promote a balanced multi-modal transportation system by modeling al l modes. The sentiment from selected members of the Transportation and Land Use Study Committee was that too much emphasis has been placed on evaluating motor vehicle mobility, whereas the transit, bike, and pedestrian modes have largely been ignored. The development of multi-modal LOS measures was recommended by the Transportation and Land Use Study Committee to focus on improving mobility through transit, bicycling, and walking, and to stop planning primarily for motor vehicle mobility. Generation of Alternatives In light of previous a ttempts to identify a means for assessing LOS equally across modes, the research team decided to extract new approaches by using a series of brainstorming/creative thinking sessions involving transportation experts and stakeholders. In consultation with FOOT, the researoh team developed an advisory committee consisting of Florida Department of Tr ansporta t ion, Florida Department of Community Affairs, Hillsborough MPO, and City of Tampa government representatives, Sprinkle Consulting, and CUTR. The research team facilitate d these sessions to identify the most widely accepted methodologies to pursue. The team held a series of six facili tated brainstorming/creative thinking sessions with the advisory committee. At the first series of meetings on February 1 2 -13, 2(){)1, the advisory committee was introduced to the creative thinking process based on the works of Dr. Edward De Bono, one of the world's leading authorities on conceptual thinking as the driver of organizational innovation, creativity, and problem solving. Dr. Edward de Bono divides thinking in to "vertical thinking" and "lateral thinking". The former is the more traditional methodusing the p rocesses of logic, removing 4


inhibitions and suppressing judgment. According to De Bono lateral thinking, however, is defmed as "a way of thinking that seeks a solution to an intractable problem through unorthodox methods or elements that would normally be ignored by log i cal thinking." The lateral thinking approach involves disrupting an apparent thinking sequence and arriving at the solutio n from another angle. In effect, it is desi gned to get people to think out of the box by starting out of the box There are d ifferences in approaches to the two types of thinking. The traditional form for brainstorming often results in "flopping about" with slight variations of known solutions. However, developing breakthrough ideas does not have to be reliant on luck or a shotgun effort in which traditional brainstorming sess ion participants often find themselves. The lateral thinking methods used in this project were designed to provide a deliberate, systematic process using proven "too ls that will result in innovative thinking One critical part of the lateral thinking process assumes that the application of creativity to assessing level of service equally across modes requires a systematic approach The advisory committee needed to focus on the same problem at the same time. Otherwise, the group would have been sidetracked with some debating the merits of a particu lar idea while o t her participants are noting its flaws or voids in understanding. Even if sidetracked, the group may have developed good ideas on all sorts of problemsexcept the one they bad been asked to think about. The first step was to i dentify the different focal points that this project could take both in general area terms and specific purpose or problem-focus The characteristic of the general area focus is that the area of thinking is defined but not the problem. This allowed the group to think about anything associated with the area General area foci can be broad or specific For example, "corridor analysis" might include ideas for defining a corridor, methods for conducting the analysis, corridor analysis focused on nonmotorized transportation alternatives, etc. A more specific genera l a re a focus m ight be "corridor analys i s approaches for b ic ycle facilities" This would narrow the creative thinking process. The purpose focus is the more commonly used focal type A purpose focus channels the creative thinking process to solving a particular problem or achieving a certain end. For example, the purpose may be to carry out some task (e.g., simplify the data collection requirements). 5


The advisory committee i dentified the following general area and purpose focal points on the task at hand assessing level of service equally across modes. General Area Focus 1 Assessing cvcryones mobi1ity 2 Low income/inner city mobility 3 Differences in measurement methods 4. Differences in the purpose of measuremen t s 5. Corrido r a nalysis 6. Corridor ana l ysis with and without bike facilities 7 Area wide analysis 8. Internal mobility 9. Concurrency application 10. User needs for LOS measures 11. Development Review/assessment mobility management appropria t eness 12 Equal assessment 1 3. Trade-offs 1 4. Crash data 15 Demand 16. App lication ofMultimo d a l levc l of service 1 '1. Cross impact analysis 18. Computational practicality 19 Need for common grading system 20. Target LOS 21. Area sensitive s t a ndards 22. Common LOS measures 23. Multi-dimensi onal measurement 24. Understanding of qua lity of service versus l evel of service 25. Future LOS 26. Cost benefit 27. Prioritizat ion 28. Transportation Land Use Study Committee 29. Facility selection 30 Scales ofLOS 6


Purpose Focu s I. Assess impact of multimodal facility. As tool for engineers pl anners 2. Accommodate all users in a corridor 3. Dev el op melhod to trad e-off between diff modes 4. Develop sing le method to measure same lhing, same way S De te nnining if LOS equally across modes is d e sir a ble or necessary 6. Evaluation of the effectiveness of Transportation Concurrency Exception Ar easlmu ltimodal districts 7 In trodu ce new lexicon to differentiate between d i fferent modes 8. Promote a b a lance d multimoda l system 9. Define what is meant b y a b a l anced multimodal system 10. Establish target lhresholds I I. Develop ins tructions to use 12. Develop me t hod for equitable fund. For melhod 13. Integrate dem and to TLOS, Bik e, Pedestrian LOS 1 4 Incorporate this project results i n state, local highway c a p acity manual (HCM), TCQSM IS. P rom ot e m u l timodall and use plann in g 16. Ad dre ss par king and vehic l e LOS 17. In corpora t e reliability i nto LOS 18. Reduce c rash rates From the generation o f the se lists, it becam e abundantly clear tha t there was a wide range o f perspectives that the group brought to the table ranging from implementation considerat i ons to use of the results The next ste p of the process was to vote to select t h e points to focus th e crea tiv e t h in king process from !he list of 48 responses that the group identified. Bach person selected the five bes t res po n ses to the question and r ank ordered them, with "S" representing the best response of the five se lect ed The vote s were tallied and are reported b elow in Ta ble I. The table a lso s hows how the r espo nd ents rated the respo nse. For example, s i x people selected the response "Determining if LOS equally across modes is desirable o r necessary" with four of the six people rating i t as their highest response. 7


Responses that did not rece ive many votes should not be interpreted as \Ulworthy. Some responses received few votes but several peopl e in the r oom gave them a high priority The responses were then ranke d based on a weight ed score Five points were awarded to each item rece i ving a highest priority ranking (#5); four points were awarded for the second highest priority item, and so forth Half of the 48 responses recei ved at least one vote from the committee members present. Table 1 Best D escriptio n of Focal Point-Voting Results #of Votes by Weighted Respome Response 5 4 3 2 1 Total Wtd votes Score Assessing everyone's mobility 0 0 0 0 0 0 0 Low income/inn e r city mobility 0 0 0 0 0 0 0 D i fference in measuremen t methods 0 0 1 1 0 2 5 Different purposes of the measurements 0 I 0 0 0 1 4 Corridor ana l ysis I 0 I 0 1 3 9 Corridor analysis with and without bike 0 0 0 0 0 0 0 facilities Area wide analysis 0 1 0 1 2 4 8 Internal mobility 0 0 0 0 0 0 0 Concurr ency app lica tion I 0 0 0 0 1 5 User needs for LOS measures 0 I 0 1 0 2 6 Development Review/ assessment 0 0 0 0 0 0 0 mob ility management appropriateness Equa l as se ssment 0 0 0 1 0 I 2 Trade-Qffs 0 0 0 0 1 I 1 Crash data 0 0 0 0 0 0 0 Demand 0 0 0 0 0 0 0 Application ofMultimoda l level of 0 0 0 0 1 0 I service Cross impact ana l ys i s 0 0 1 I 0 2 5 Computational practicality 0 0 0 0 0 0 0 Need for common grading system 0 0 I 0 0 I 3 Targe t LOS 0 0 0 0 0 0 0 Area sensitive standards 0 0 0 0 0 0 0 Common LOS measures 0 0 0 0 0 0 0 Mul tidimens i onal measuremen t 0 I 0 0 0 1 4 Understanding of quality of service 0 0 1 0 0 I 3 versus level of service 8


#of Vot es by W eighted Response Respo nse 5 4 3 2 1 Total Wtd votes Score Future LOS 0 0 0 0 0 0 0 Cost benefit 0 0 0 0 0 0 0 P rioritization 0 0 I 0 0 I 3 T r ansportation Land Use Study 0 0 I 0 0 I 3 Committee Fac i lity selection 0 0 0 0 0 0 0 Assess impact ofmu l timodal facil ity. 0 0 0 0 0 0 0 As too l for engineers, planners. Accommodate a ll users in a corridor 0 1 0 I 0 2 6 Develop method to trade-off between I 2 I 0 0 4 1 6 d i ffmodes Develop single method to measure same I I 0 I 0 3 II thing, same way Determining if LOS e qually across 4 I 0 0 1 6 25 modes is desirable or necessary Evaluation of the effective of 0 0 0 0 0 0 0 transportation concurTency exception areaslmultimodal districts Introduce new lexicon to differentia te 0 0 0 1 0 I 2 between different modes Promote multimodal balance 0 0 0 0 0 0 0 Defme ba lanced mu l timoda l system 0 0 0 0 0 0 0 Establish target thresholds 0 0 0 I 0 I 2 Develop i nstructions to use 0 0 0 0 0 0 0 Develop method for eq u itable fund. For 0 0 0 0 0 0 0 m ethod Integrate de man d to trans i t leve l of 0 0 0 0 2 2 2 service, Bike Pedestrian LOS Incorporate this project resu l ts in sta te, 0 0 0 I 0 I 2 local highway capacity manual, transit capacity quality serv ice manua l Pro m ote mu ltimoda l land use planning 0 0 0 0 0 0 0 A d d r ess parking and vehicle LOS 0 0 0 0 0 0 0 Incorporate re li ab i l i t y into LOS 0 0 0 0 0 0 0 Reduce cr a sh rates 0 0 0 0 0 0 0 Scales of LOS I 0 0 0 0 I 5 From this list, it b ecame readily a pparent that th e grou p agrees that assessing leve l of service equally acr oss modes i s a reasonable co urse of action. 9


CUTR developed a creative thinking meeting format to separate out thinking into six types (see Tabl e 2 below) Each type specifies a d i rection and focus for thinking. The advisory committee was told the sequence that the types would follow. The seq uence was designed to allow for f u ll discussion of the problem, generate alternatives and, ultimately, reach a conc lusion. By separating the thinking i nto six types, the Committee was forced to focus on one aspect of the problem at a time. Participan t s were directed when to switch thinking from o n e type to ano ther. Th i s sequentia l framewor k was used because it releases the group from inefficiency of the c i rcular d i scussions and argumentative natur e of traditional meetings Another benefit of this sequentia l approach is it allows the committee to b e able to explore alternatives in parallel. Table2 Six Types of Thinking and Their Purposes Six Types of Thinking Purpose of Thinking ll'let b od Managing the Thinking Proce s s Set s the agenda Sets the t i ming Defines the probl em/foca l po i nt Establishes the process for approac hing the problem Summarizes the decis i ons or conclus i ons Informa t ion Available and N eeded Gathers background data I d entifies areas where information is missing Al t ernatives & Cre ative Ideas Generate s a l ternative approaches D i scovers ways of overcom ing obstac les Lays out how else the concept might be carried out Benefits Pos itives and Value s Finds the benefits of the ideas Identifies the positives Identifies the value of the id eas Caution, Di f ficulties and Prob lems I aentit1es the fault s or reasons t hat we should not choose this approac h Intuition and Feelings Selects the a p proach/shows how we feel about the idea Signals intuition feelings and emotions The first step in the sequence was to i dentify what we know or take for granted a b out the problem o f assessing l eve l of service equally acr o s s modes. From t hat point, the group was ask ed to 10


describe the benefits of the idea of having such a method for measuring LOS equally across modes The third step was to inv i te the group to identify the faults or reasons for not doing it. The group was asked to generate possible solutions or alternatives using a c rea tive thinking technique (described below). The following li sts the response of the first three steps What Do We Knowtrake for Granted? The full group agreed to the following: That it's poss i ble to measure level of service equally across modes We c an determine a method where LOS C for one mode equals LOS C for another mode That there is a potential problem Modes incl ude auto bike, pedestrian, and bus (truck included in auto) We measure LOS separate l y/individually. LOS is differ en t across modes !!!!.Y! (e.g., Auto LOS F is not equal to Bike LOS F) LOS can be calcula te d for all modes People have access alternative transportation modes LOS shou ld be based on user p erception Government LOS should equal People LOS A single measure would improve transportation planning achieving 1 and 2 Six level scale is appropriate for all modes Decision makers \vill not discern difference s between LOS (modal) Bike C not equal Auto C Need more clarity Broad audiences can understand quality or LOS LOS measurements are professiona lly defensib le Bike, pedestrian, and transit should be routinely considered in planning design and operation of highways Assessing LOS is a worthwhi l e activity LOS does not equal quality of service Highway Capacity Manual is the definitive source on QOS/LOS concepts 11


Necessary to educate the transportation professional on QOS concepts Cost money to ftx. Will have to incorporate into QOS (ARTI'LAN) Current methods don't include Land use compatib i lity. Suburban vs. urban in non-auto QOSILOS Doesn' t consi der continuity of systems (e.g. missing sidewalk and/or bike facility segments Missing user response to mi:>tor vehicle LOS Methods to measure aren t static; need to be flexible like midblock crossing intersection FDOT has multiple related projects underway midblock, point, and corridor LOS mcastll'es Former Transportation Land Use Study Committee wanted a multi modal measure Growth management may c hange from concurrency to full cost accounting New intergovernmenta l coordination process will need to consider or use a method f or measuring LOS equally across modes What are the positives of having such a system? Gives impetus to multimodal planning Might give assessment across all modes to tie into impact fees Makes it easier to understand by public and elected officials, transportat i on professionals Simplifies the trade off process Provides cons i stency in measUl'ement across modes More environmentally sensitive Promotes livable communities Provides a better evaluation of concurrency (area wide) What are the negatives or drawbacks? Not assessing the same thing May need compl e t ely n e w methods to measure LOS/QOS May involve more subjectivity than 11exact" science Increases leve l of effort by peop l e w h o appl y it N ot everyone understands the current methods 12


Current research stands on own FDOT tried to match up Transit and Auto LOS without technical merit Auto LOS and transit lacks behavioral data Auto LOS is entrenched w i thin industry Wide funding disparity exists between modes Demand is more important than LOS What should we do? The advisory committee expressed concern in i tially whether assessing level of service equally across modes was desirable and/or feasible given the differen t needs revealed in the current methods. of measuring LOS. They agreed that the current levels of service are appropriate for their mode, understandable b y broad audience s, and profe ss ionally defensible. However, after they i dentified the advantages and drawb acks associated with a system of measuring level o f service equa lly across modes, the advisory comm ittee also agreed that this system would be of significant value to policymakers, developers and the transportation industry. ln pOl'licular, the participants noted that it would give more impetus to multimodal p lanning and simplify the process for making trade-offs among modes at the local level. Even with unanimous agreement to move forward with the development of such a method, the advisory committee recognize d several challenges inheren t in their dec ision. The group c i ted the lack of a common denominator among the current measures as one of the most significant obstacles. In part i cular, they noted that only the b icycle and pedestrian levels of service are based on system characteristics that have been correlated to user responses. At the same time, the group saw the resistance to any c hange in the other modes' LOS methods as another obstacle. Pointedly, the advisory committee concluded that a fundamental requirement for the development of such a method to assess LOS equa lly across modes would be to relate the LOS for each mode to user perceptions. The ad visory committee then began the process of identify i ng a lte rnati v e approa c h es to the prob lem. In effect, the committee began to develop a system that keeps the current LOS methods in place but seeks to rela te each mode's LOS to the other modes' LOS by establishing the relationship to user needs as a common characteristic. 13


What alternative approaches should we conside r? For this mode of thinking, the provocation operation (PO) or the stepping stone techn i que was the creative th inking approach select ed The purpose of the provocation statement is not to be "right" but to move the mind to new ideas (i.e., start thinking out of the box by sta rtin g out of the box). The PO technique helps individuals or groups generate new ideas by deli berately starting from a situation that forces them to consider "what happens next". Provocations need to be "uncomfortable". The technique relies on the brain's innate ability as a self -o rganizin g mechanism to make s ense of various inputs or a p articular situation Acting upon the normal thought processes and turning them inside out produces the PO. The process used to create these provocations is as tollows: one of the easiest ways to craft a provocation is to systematically take something that we take for granted and escape from it. This esc ape may be in the form of negating it, dropping it, or doing away with it Other approaches are to reverse the norma l direction of action or preface with the phrase "Wou ldn't it be nice if .. ". After the deve l opment of the PO s t atements and selection of the PO statements that make us the most "uncomfortable", the group then must move forward from the provocat i on to get new id eas The natural inclina tion is to use normal on the provocation. Given that the purpose of provocation is to deliberately think differently (i.e., out of the box), then jud gment would often result in the rejection of the statement. Even the suspension of j udgment a cornerstone of most brainstorming methods does not tell us what to do. "Suspending judgment" does not hel p us extract value from the provocation. We must actively move to new ideas There are four major forms of movement from response to idea. I. Top-of -Mind Some ideas just come from whatever pops into your mind based on the provocation. A phrase or image may stimu late an idea. 2. Extraction Pull out something such as a feature or princip le from the provoc a tion rather than only considering the "whole". 3. Focus on the DifferencesExamine what is different between the provocation and normal way of doing things. 14


4. Moment-to-Moment -Think what happens from moment to the next. In effect, this method is a movie in your mindyou can see the situation developing every step of the way. You can also rewind and pause at a particular poin t to examine closer what migh t be changed. The group developed the following Provocation Operation Stateme nts based on w h a t we stated that we knew o r took for granted (see above). Provocation Operation (PO) Statement PO System measures the transportation professional PO Perception of LOS is totally random PO Levels are unnecessary PO LOS has no relevance to dec i s i on making PO No one drives PO Don't need Highways PO Wouldn' t it be nice if transportation professionals started wit hout pre-.:onceived ideas PO LOS has great significance to the public PO N ot all modes need LOS PO Who are you kidding that it is professionally defensible PO Wouldn't it be nice if LOS defensi ble by children? PO HCM is useless docum ent PO Everyone has a copy of HCM Several of the provocation statements were used t o develop ideas on a "system" or approach to assessing LOS equally across modes. PO 1 System measures the transportation profe s sional Picture of designer by faci lity w/phone numb er to yield more accountability Failure of system equals a loss of credibility The current situation is a faceless community/ everywhere looks the same; this would put a "face" on it Objective measures to rate transporta t ion professiona l (to justify raises and budgets) Measures transportation professional so fmancially accountable Enhances professiona l credibility 1 5


Would actually be measuring a different time frame-it would be measuring the performance of transporta t i on professionals in the past Measure current projects to improve performance Public might vote on transportation professional Would create a system that is more fa ir, support for more money and reduce political pressure PO 2 No One Drives We get"Disney World" Focus moves to delay from capacity Differen t measures of effectiveness Improve lan d use patterns More use of horses Not everyone has to drive every day which could lead to "No drive Thursdays Improved local ization One less mode to deal with Would result in improved facilities to o t her modes as more money would be availa b l e for other modes Exp ectations are diffe r ent More centr a lized control of the transportation system; how people choose to ttavel No need for ttansportation management o rganization s (TMOs) and other groups Eliminates interrnodal carrie r s f dissimilarities Changes focusing to measuring peop l e rather than vehicles Would make aesthetics more important as capacity i s no longer an issue Would result in changes in urban design criteria Would allow for mor e accommodation of major users Change cou l d shift peak travel time of day Coul d shift emphasis from highway LOS t o address multimodal needs Measure LOS would focus on leading to solutions what would work Measure what we want to communicate to constituencies Changes to urban design criteria Not mea suring right thingnot mobility \ 6


Drivers want: Speed time Aesthetics Reliability comfort convenience There could be more movement from concurrency c.oncerns to urban design Focus would shift to internal trips I internal mobility Applicable to MMTF only? TCEA? DRI? Applicable of techniques to intersections 7 area wide Relate to targe ts PO 3 Levels are unnecessary Have continuous scale Don't measure at a ll Establish a polar scale with only good-bad score. Simplifies cross modes Easier and more accurate/ statistical reliability Gather user input Focus on specific area (geographic)(measure may not be necessary in some areas) Not necessary for auto mode, only calcu late other modes Straight calculationn o grouping Regulatory For well trained people such a measure is not needed Not needed for planning Use v ideos and simulation instead of Levels Not needed to determine solutions when we can see it Forces transportation professionals into t he field Communication with pu blic would be easier Better/worse 7 relative measure Videos maps Communicate using a "tim e" measure such as s p eed Move to using units (perception) such as minutes of delay per mile rather than LOS without units Crea t e DOW Jones Index Move to a more temporal/longitudinal measure to gauge progress over time 1 7


Everyone i s transportation expert Requires the use of a dynamic m eans of representing cond i t i ons Assessing the Alternatives Using the se quenti a l thinking proce-o;s, the n ex t step was to assess the alternatives a n d pr e pare pr op osed approach(es) for testing. At its seco n d series of meeting s on M arch 15-16, 2001, the adisory commi ttee focused on harvesting the ideas generated in the February sessions into specific p ro posals. Each of these alternativ es was then sub ject ed to identification of the positives theri the ne g a tives asso cia te d with each a l t ernative. The attributes for each of the following concepts se l ected by the adv iso ry commillee are described along with a listing of the advantages/strengths and d is advan t ages/problems with each i d ea. Dow Jones User Ratings Multimodal LOS Profile (a.k.a. Slide Rule) Weigh t ed Slide Rule Community Standards Based Method Modified Miami Metho d Sensory Method Icon Method 18


Dow Jo nes User Ratings The group c o llective l y desc ri bed the Dow Jones User Ratings option wi t h the followi ng a ttributes: Dow Jones User Ratings Measure user perceptions on a set of c haracteristics Use same characteristics for all modes Comfort Safety Speed or time Reliability Connectivity Convenien-ce Enjoyment/aesthetics Collect in siorilar manner Drive for Science Bike for S cience Ride transit for science Walk for science Weight by person trips (vo l ume) Figure 1Summary of Dow Jones Ra t ing method The group assessed the positives/benefits of the Dow Jones User Ratings (DJUR) method as: Makes all modes based on user perceptions by adding Drive for Science" and "Ride transit for science" efforts. It would use the same char a cteristics across all modes. Builds political consensus via a survey of people's assessment rather than transportation profession s assessment. Provides a co!lsistent measure to compare the system over time and across geography much like the Dow Jones stock index. Generates med i a interest. The media would love i t as a means to generate stories and lists such as the "10 Worst Intersections". Increases work f or con sultants. 19


The group a lso identified the following perceived drawbacks of such a system. Increases the costs for measuring LOS. Increases the data collection effort. Many factors are missing for some of the modes (e.g. Drive for Science) and, thus, the workload of local staff would have to increase. Relies on public opinions makes the DJURxm subjective (i.e., difficult to identify specific improvements). Masks differences and underlying interactions. Encouncers resistance to change, as the devel opment of a new way t o measure system performance woul d take a lot of effort to convince current transportation professionals and others who are invested in the status quo. The usefulness of some characteristics (e.g., travel time or delay for b i cycle users) is debatable. User opinions would be collected from people who are outside of that particular area and may not be familiar with prevailing conditions. 20


Multimodal LOS Profile (a k .a. Slide Rule) Under this method (Figure I), the actual levels of service of each mode are aligned (i.e., the LOS bars move). Multimodal Level of Service Profile f-Actual Level of S e rvice A B c D E F Auto T ra n sit Bicycle Pedestrian Location: Oshkosh Avenue@ 1st Street Figure 2-Mul ti modal Level of Service Profile The group assessed the positives/benefits of the MMLOS Slide Rule (Slide Rule) method as: The slide rule is more continuous method of showing bow cl ose a particular mode is to the previous or nex t LOS (i.e., we can see" whether the LOS is a high D vs. low D). Shows LOS clearly for each mode. Provides way to compare one mode vs. another 2 1


RecQgnizes value/prQgress of within range changes (i.e., moving from a high D to a low D). Provides way to compare one facility vs. another Now have a common scale-green, yellow, red. The slide rule provides an easy to understand method (Visual). Provides flexib i lity for interpreting the results. Provides more informatio n co n tent than j ust a number. Does not require additional data collection. The group also identified the following per ceived drawbacks of the slide rule system Implies tha t the range of experience is the same across modes due to the equa l length representation of the l etter grades ( i e., same level of effort required to move from one letter grade to another w ithin the mod es and across modes). Requires add i tional time and resources to portray LOS network-wide. Doesn't lend itsel f to trade-Qffs across modes Doesn t lend it to fund allocation at the area-wide and corridor leve l s Can't rank order & prioritize. Someone has to determine what is acceptable (Politicians?). Requires cross-country connections/cooperation. 22


Weighted Slide Rule This o ption was an extension of the Slide Rule concept discussed above. The major difference is the conversion of the above to a single quantified score that is weighted and that represents overall LOS. Weighting schemes could take into account one or more o f the following: Weight by # peop le Weight by C02/mile We i ght by cost Weight by CO Weight by energy Weight by geography The group identified the following advantages or benefits of the Weighted Slide Rule system: Converts several multi-modal LOS measures into a single score Makes it easy to prioritize when converted into a single score. Allows for the comparison of facilit i es Simplifies a complex process. Permits communities to emphasis one mode over another (or not) by the weight assignment process. Can be adjUSted to reflect change s in policies. Allows for applicability at different locations (i.e geographic based ) with different we i ghtings Provides a visual representation. Allows for single and system level cal c u lations. The group also identified problems or obstacles with the Weighted Slide Rule Masks modal characteristics Uses subjective classifications (e g., Green/Yellow/Red) Is not calibrated to user percept ions (good vs. bad). Demands a level of effort for creating all charts. 23


Community Standard Based Measurement Another permutation of the Slide Rule was discussed by the group (see Figure 3). The focus of the Comrmmity Standard Based Measurement aims at aligning the scales along the commu n ity standard axis and showing the gap between what actually exists and the community standard For example, in the exhibit below, only Pedestrian LOS A is acceptable to this particular community. However, Highway LOS of A through D would be acceptable. 11te scales are dis pro portionat ely sized to allow for the alignment with the various community standards. The gap could be measured as the area under the acceptable standard. Furthermore, the gap could be wei ghted by geographic locat i on What is acceptable could be determined by location or subarea. The group found the following benefits and advantages of the Community Standard a pproach to includ e: Permits the differences to be v iewed easily as all acceptable areas line up and the current conditions o f the area under question vary (zigzag) to v i ew differences. Makes i t easier to compare relative differences. Assigns priority by sizing by LOS modes (modes which more important/desired to be emphasized can be made by stretching). Leaves standards in p lace but shows actual performance. Provides a measure of quantifYing the gap between standards and performance. Shares many similar benefits to the Slide Rule. The group found the following problems and obstacles with the Community Standard approach to include: Could be misin te rpreted. Shows that changing from one LOS grade to another varies across mode (i.e., we don't move through scale at the same rate) . Makes i t more difficult to explain to citizens why the length of LOS is d i fferent (by re qu iring a more technical explanation) if you use A-F rather than "acceptable/to lerable/unacceptable" or "green/ye llow /red gradations on t he bars. 24


Community Standard Based Measurement PED BIKE TRAN AUTO Figure 3 Commnoity Standard-Based Measurement System 25 CJ Aoccptable LOS Conununil)' Standard Difference between Standard and Actual


Sensory Static Picture While the previous examples relied nearly exclusively on numbers or letter grades to communicate the concep t s oflevels of s ervice the group collectively described the Sensory LOS option with the following a ttributes: No Letters/Numbers Visual Use photos. of various conditions Use c<>lor screen to represent good/bad (hot/cold) for various characteristics of that mode Figure 5Summary of Senso r y LOS Concept The group i dentified the following advantages and benefits assoc i ated with the Sensory LOS option : Use o f visua l s makes it simpl e to understand and easy t o rela te. Does not require explanation. Demonstrates way LOS is be ing measured today (fo r motor vehic les}. EJtpl a in s the differences of peak vs. non-peak traffic on the same facility. Uses static images means it is cheaper and more transferable than mot ion Allows a community to capture shots o v er time at in tersections 'vi t h cameras Captures and communicates skill level using particular icons (Kids thru Adul ts). Could use descriptive 26


The group identified the following disadvantages and problems associated with the Sensory LOS option: Makes it easy to distort the truth. Can't use to forecast. Pictures not available for some conditions or modes. Does not easily measure time and Speed components of the experience. Uses same methods just different communication tool (does not provide anything new). Becomes more complicatedincreases resources neededmore paperwork. Diverts dis cussion to "my road doesn't look like that". Much of the focus of the group discussion was on the visual representation of the levels of service. However, the group also identified but did not assess the advantages and disadvantages of other "senses" such as sense of hearing (e.g., usc sounds and/or change the volume) and sense of touc h (e.g., use relief maps) to represen t differentlevels of service. The group also recognized the potential of using the other senses as another means of communicating LOS to the visually impai red. 27


Modified Miami Method The group also discussed the Modified Miami Method of m easuring l eve l of service th at focuses o n perso n carrying volumes and capacities (i.e., Person volume (d ivid ed by) Person capac i ty) The Mod ifie d M iami Method option was described as having the following attributes: Modified Person volume d i vided by person capac-ity A ssume c ar bas 1.6 per son ca pac i ty A dd b ike and wal k Develop stan dard c re dit s to add to p ers on capacity (6' bikeway vs wid e curb) Allow developers t o e arn int e re st in a bike/walk/trans i t bank (improvement becomes more va lua bl e over time) and sell c r e dit s to nearby d evelopers Limit to multi modal transpo rta t ion di s tric t s Geographic component-demand-determ ined Figure 6 Summary of Modified Miami Method The group identified the following advantages and benefits associa te d with the Modifie d Miami Method: Cred i ts given to d evel opers for capacity improvements even if capacity is n o t used. Simple to ca lc u late. Sounds good p olitic a ll y T ried elsewhere. Red uces burden on the gove rnmen t to inc re ase supply /p rovide infrastructure to meet peak demands base d on vehic les Results i n more multimodal facilities. Prom otes multimodal solutions Ti es t o geograph i c area Allow s for differen t modes to b e iden t ified in different areas. Provides an efficie n cy /uti lization measure 28


The group identified the following disadvantages and problems with the Modified Miami Method: Lacks bas i s on user perceptions. Fosters a "We don't care" image of pu b lic sector Requires bike and pedestrian capacity to be measured (when it may be di fficult or meaningless to mea s ure) Doesn't provide measures for indi v idual modes. Might result in the building of useless facilities (on purpose) to gain capacity w i thout solving the problem May prioritize wrong things May skirt concurrency 29


Icon Method The group also discussed the use of icons as a means of helping communicating LOS across modes. The leon Method option was described as having the following attributes: Grou of Icons Ferrari for fast moving highways Kid for safe walking or biking Kid with adult for supervised walking and biking Kid on trike, bike with training wheels up to racing bike Figure 7-Example of Icons to Represent Different LOS The group ident ifie d the following advantages and benefits associated with the Icon Method: Uses icons to relate to skill levels (safe for children, safe only for children accompanied by adult safe for adults only, etc.). Provide s opportunity to show non-real circumstances (3 buses to represent changes in service or frequency). Decreases data in tens i ty. Uses existing methods to quantify LOS. Shows all modes. 30


The group identi fied the following disadvantages an d prob l ems with the Ico n Method: L acks c.ontinuous symbols Loses gradations (no representation of different gradations of LOS C). Limits to displ ay capabilities (possibi li ty tha t icons w ou ld be too complex). 31


Results Advisory Committee Recommendations For the short term the advisory comm i ttee's p r eferred option is to pilot test the i nclusion of the basic slide ru l e as an ART-Plan enhancement with the provision of a method to allow communities to weight the various modes Furthermore FOOT shou ld provide gu i dance to local governments on setting targets to make sure not that the slide ru l e method for assessing equally across modes doesn>t make it an exe r c-ise of s imply translating the non .. auto modes int o an auto equiva lency LOS. The advisory committee recognizes that this system can perhaps adequately communicate the various modal LOS le vels on a single facility However, the committee also recognizes that i t still does not, by itse lf, allow for assessing LOS equally across modes. Th e "slid e rule" options will only truly be effective for that purpose when the characteristics are relate d to user perceptions To that end, the advisory committee concluded that the identification of the common denominators across all t he modes or users was necessary for a true method of assessing LOS equally across modes and perm itting tradc-offs across modes. The advisory committee strongl y recommends that FDOT seek to incorporate user perceptions to identify how to respond to the range of needs reflected by the variou s LOS measures and dynam ic needs of transportation users. The advisory committee would like FOOT to explore the ident i fication of the key factors or needs from a user perspective for each mode and the factors re la tive importance to each other The method to b e pilot tested should be designed to iden tify the linkages or common ground between modes. The advisory committee believes that these linkages imply that when there is a degradation of LOS for Mode X t hen changes in LOS for the other modes would be required to reach the d esired system performance How commun i ties will respond to the changing needs is the first step for allowing trade-offs (e.g., r esponding to an auto capacity problem with a safety improvement for bicyclists and pedestrians). Ultimate ly, the adv isory committee sees the development of a Dow Joneslike approach for assessing LOS across modes may be useful f or trade-offs and accessing LOS equally across 32


modes and the system It may y i e ld a mix ofperfonnanee indicators common to each mode similar to how Price E arni ngs ratios, 52week highs and lows, etc. are reported in the business section of the newspape r Furthermore, this Dow J o nes approach would foster track in g over time. The advisory commi ttee r ecognized that there w a s a need for a developing a method for multimodal weighting to reach policy targets. They further recommended that the me t hod sho uld be flexi bl e and be driven by policy to re flect different conditions For example some cons ideration should be given to different weighting or convers ion factors based on co rridor o r l ocation (i.e., the options and user trade-offs are quite differen t i n downtown streets than roadways in remote office parks or suburban subdiv isi ons) These target weightings should show people where the system stands t oday so they can communicate or determine wei gbt ings i n the future. The advisory committee recommended that the approaches be te st ed a t 3 different l oc a tions with lo cal involvement (i.e., hold mock test with local transportation staff). The advisory committee also wanted considera tion of using ic ons or other sensory representations when reporting LOS. Conclusions As previous ly d iscussed the advisory committee's various "slid e rules" depict different methods of representing LOS and assessing leve l of s e rvice acro s s modes. Fo r example, the Community Standard Based Measurement "slide rule" below groups the various levels of.service into three broad ranges of acceptabi li ty (Figure 3 is for illustration purposes only). The relative acceptability in terms of levels of service grades varies by mode. Under thi s example, only LOS A is accepta b le for pedestrians but LOS A through D would be a cceptable for au to users. 33


Community Standard Based Measurement PED BIKE TRAN AUTO Figure 3-Community Standard Based Measurement System EJ Acceptable LOS Community Standard Difference between Standard and Actual It was the consensus of the advisory committee tha t similar fact ors be used to measure LOS representat ions if leve l of serv i ce is to be assessed equally across modes (LOS C auto= LOS C bike, etc.) The factors used to calculate LOS for each mode currently a range of user experiences. Bicycle LOS is a safety and convenience function of the interac tion of the attributes of the adjacent roadway facility, i nclud ing pavement width, posted speed limit, etc that was sealed based on user reaction. On the other hand, auto LOS measure is based la rgely o n either vehi cle speed or delay, depend ing on whether LOS is being calcu l ated for a road segment or intersection. No single performance measure represen t s the predominant user-perceived measure of facility service of t h e other modes. The d ifferent mix offactors to calculate LOS for each mode i s a r eflection of the transportat i on profession's assessmen t of the most critical factors for that particular mode. As a credit to the 34


transportation engineering community, auto users rarely feel that their personal safety is threatened because of the design and maintenance of the facilities. If the levels of service were meant to be the same across modes then LOS F might represent the same as the potentia lly life threatening experie n ce b icyclist might encounter for a bicycle LOS F. For example, LOS F for an auto user might occur when the auto user is riding in a sub-compact on a busy high speed highway with semi-trailers on all sides and barreling along at 70 mph in order to be equivalent t o what a pedestrian might feel when attempting to cross a busy highway (ped estrian LOS F). As discussed in the intr oduction transportation investments are influenced by level of service ra t ings of the current and expected roadway facility performance are based on a variety of criteria and, in effect, weighted differently. LOS for automobiles transit, pedestrian and bicycle modes are based on a variety of criteria and, th erefore, calculated o n a different basis. For automobiles, LOS is measured using average stopped de l ay for intersections, average speed for arterials and density for freeway segments. Automobile LOS "F" implies traffic is at a near standstill. For bicyc le s, LOS is a function of the typical roadway co n d itions, bicycle facilities, and safety perceived by users. Unlike automobiles where LOS "F" represents too ma ny users, bicycle LOS "F" is just the oppositeonly those who absolutely have to bike will do so, probably due to safety concerns or lack of facil iti es all toge ther. LOS for pedestrians is similar t o that for bicyclists For fixed route transit, LOS measures access to transi t routes based on population w ithin walking distance to bus routes and service frequency. The advisory committee concluded the current LOS measurement methods make t he roadway facility performance and multimodal tradeoff decisions d iffic ult to assess based on the current methods used. In effec t the advisory committee recognized that transportation system users have a hierarchy of needs that are common to all modes. This hypothesis is similar to Abraham Maslow's Hierarchy ofNeedsone of the best-known theories explaining the actions of people. Dr. Abraham Maslow hypothesized that a hierar chy of needs motivates people. The hierarchy he desc ribed may be represented as follows in Figure 4: 35


Maslow's Hierarchy ofNeeds Need Netdfo Self-Actualization Need for sdfcstccm Sodal NeedsBdonglng Neod for Safety and Security Pbysital Survival Nttds Figure 4-Maslow's Hierarchy of Needs Motivation to Satisfy Need Challeoglng projects. Opportunities for innovation and cruthity. Important projHU. of strtngtfl-htCelllgtnct Status Pmtlgt. Atteptuce. Group membtrtbip. L<1ve aod Pby$ltal ufety. Eronomlt.s:urity. Frdom from tbr"ts. Comfort. Water. Food.Sirc:p. Heallb Exerche. Sex. Maslow's hierarchy states that once a person has met their most basic needs, they then begin to consi der higher needs. As Maslow said, "as one desire is satisfied, another pops up i n its place W e developed the following hierarchy of transportation system user needs drawing upon the discussions of the advisory committee and pattern ing it after Maslow's hierarchy The Transportation System User Hierarchy of Needs (Figure 5) is hypothesized as consisting of five levels of needs: safety and security (the most basic need ), time, social acceptance, cost, and comfort and convenience. The trav ele r's safety and security is considered to b e the most basic need. Timesavings, convenience etc. are nearl y meaningles s conside ra t ions if personal safety i s threatened. Furthermore, this need is thought to include the d egree of familiarity with the mode, route an d destinat ion The more unfamiliar they are with the route, for example, the greater the likelihood tha t they a re concerned with persona l security The next highest need relates to travel time, inc luding access time, waiting time, and i n vehicle time. Th e need to link trips or trip chain also may determine the level of serv ic e from the user's perspective. 36


The third level is categorized as social acceptance as reflected by personal and peer/society attitudes toward modes (for or agains t ). This level suggests that people may choose a mod e with a l ower measured level of service based on some personal belief (e.g., environmentally friend l y better for personal health). Of course the choices t hat provide a cost advan tage offer another need. These costs can include fixed expenses suc h as the cost of th e vehicle and variab le expenses such as gas, tnlnsit fare, tolls, and parking. Finally, as the traveler seeks to optimize the tnlvel ex perienc e then the needs of comfort and system reliability come to bear. This o r dering of the se five basic needs says that only once the most bas i c need is met will the other needs b e considered. Onc e the persona l safety needs of the transpo rt a tio n system user are met to his or her satisfactio n then the time need is addressed. When the time need is met then convenience need (e.g., how m u ch delay is encountered) pops up. Finally, we hypothesize that comfort and convenience is the low est ordered need In genera l, this hierarchy explains why only the highest l evels of services for bicyclists and p edestrians (e .g LOS A or B) could be comparable t o t he six LOS levels for autos. Only until conditions exist that app r oach the same "safety" threate ning conditions for autos that bicyclists or pedestrians face w ill the assessmen t of LOS equa ll y across modes be possi ble. Transportation System User Hierarchy of Needs Comfort & Con,'enimce Cost Time S a fety ud Stoirity Motivation to Satisfy Need Better tnvd cxperlc:n cc. Comfort. More: reliability. a3y t.C(.($$. Bert Valut. Fixtd insurance) and "ariablt (&as, rare. tollf, parkin&) Aepta.nee. Personal and pttr/sadety attitudes toward modes (for or against ) Trip Efficic:ncy.l n-vthidt tlmt Waiting time. TraoJ(c:f$. W lk time. Trip Pe:rsooal safety. Security for persooal property Famitl a rity with routt, and destlnatio n Figure 5 -Transportation System User Hierarchy of Needs 37


As the ad visory comm i ttee noted, the various methods of measuring level of service for auto, tra n sit, bike and walk modes do not consider the same factors (i.e., the transporta ti on system user's hierarchy of needs). In fact, as represented graphically in F i gure 6, the current LOS measurements by mode may not cover the full range of the hierarchy. Therefore, assessing LOS equally across modes will require fram i ng the LOS in the context of this hierarchy prio r to design the pilot testing for assessing LOS equally across modes. LOS Coverage of Hierarchy Auto LOS vs Bicycle LOS Auto LOS Bicycle LOS Figure 6LOS Coverage of Hierarchy A -F The advisory committee held several spirited debates about h ow the fundamental needs among a ll transporta t ion users are the same but that different factors come into play. In effect, the advisory committee was discussing t his concept of a transportation systems user hierarchy. According t o Maslow a satisfied need no l onger motivates. The advisory committee also recognized this basic truth; once a basic need was met then t h e user would seck to have the next level of need met. Once the safety and securi ty need is met for autos for example, then the ind ividual seeks t o address h i s need to address time constraints (e.g., How can I get there in the shortest time? Will I be able to link my trips? Etc.). 38


At the same time, the threat or failure of the transportation system to meet the most basic need (e.g personal safety) will make meeting the need of a d ifferent level moot For example, a tourist ex its his hotel room and decides to cross eight-lanes of Fowler Avenue to eat dinner at a local restaurant. He has two practical options: walk or drive. Even though it may be quicker to wal k across the busy highway, he assesses the trip as a real threat to his personal safety if he tries to cross. Therefore, the bas i c need of safety and security overrides othe r needs (e.g., quickest method from point A to po i nt B) and he chooses to drive. The same hierarchy holds as you move up the l evels. For example, a commuter may be circl ing downtown looking for an affordable parking space but once the price level need is met then the nex t need is convenience in terms of parking i n a nearby location. To be a hierarchy, the relationships must hold their positions relative to each other. The hierarchy does not exist by itself, but is affected by the situation and soci ety norms. Some environmentalists or heal th-conscious travelers for example, may be more willing to forgo the comfort that an auto or transit for a trip by bicycle because they perceive that opt i on to be more in line with their personal bel iefs or more acceptable to society. Level of service, therefore is relative. This hierarchy method of framing the issue helps shape the approach and focus the design of the pilot project to assessing LOS equally across modes. The expectation is the data collected for the pilot project will a llow for a quantitative-based "adjustment factor" so each LOS gradation means the same for all modes. Ideally the analysis will revea l a model that takes the form a step function, where there are clear demarcations when there are changes in LOS (e.g. when do an auto user, a transit rider, a bicyclist, and a ped e strian feel unsafe.) In effect, analysis would reflect the adv isory committee's thinking that normalizing the LOS gradations across modes via user opinions (albe i t via shifting of the sca les or app l ying some to be-determined discount factor) would create these bands of acceptable standards gradations or overlays (e g., green, yellow, and red) as shown on the slide rules Though Maslow's theory pervades the literature on employee motivation, a TRIS search did not reveal any transportation research related to a hierarchy of needs for transporta tion. However, time and cost based methods may overl ook psychological factors. A briefliterature search was conduct e d to establish the relative importance of these psychological factors or "motiva t ors". 39


Overall, the small body of literature ind i cates that there are most certainly cognitive processes at work when one is choosing a travel mode, and that these processes do not necessarily relate to an objective assessment of the characteristics of a particular mode or demographic factors. In fact, Kuppam, Pendyala, and Rahman declare attitudinal data so important in understand in g mode choices that "their omission from modechoice models may be more serious than the omission of demographic variables" (1999:75). According to Koppelman and Pas, attitudinal data related to beliefs or perceptions have for too long been the only focus of psychological factors in transportation research. Their examination o f the social psychology literature led them to state, "that other p sycho l ogical dimens i ons might be important determinants of travel behavior'' (1980:28-29). The psychological e l ements that contribute to mode cho ice are typically abstract concepts such as independence or freedom, values of time, and needs fulfillment. These elements are often individua ll y defined, and the sources reviewed here provide no ranking of their relative importance. However, those psychological aspects that are considered to contribute to travelers' mode choices do recur throughout the lite rature. Detecting such themes in published transportation research is in support of a Mas low-like Hierarchy of Needs. The following is a summary of several resources that discuss the psychological aspects related to mode cho ice. Ulberg ( 1 989) contends that values, beliefs, and psychological factors do play a role in mode choice decisions, but the relationships are not well understood. The author defines values as deep seated beliefs guiding human behavior. With regard t o mode choice bus riders are distingu i shed from non -bus riders on the basis of two valuesthey (bus riders) evaluate "equality" very high and "freedom" very low. "Fami ly security" was found to be a strongly held value among those who downsized cars and reduced their driv ing in response to the gasoline crisis of the 1970s. Those who did not participate in such behaviors placed a high value on "a comfortable life" and "an excit in g life." Values were also found to have a "significant effect" on the size and number of cars a family owned With respect to the hypothesized "societa l acceptance" level, Ulberg cites psychological research that provides evidence for two types of personaliti es>-int emals and externals. Internals generally feel a ble to control their fates, and externals feel relatively unable to do so. Internals would feel less anxiety about traveling v i a automobile than would ex ternal s, and internals would tend to 40


experience automobile travel as more interesting and involving. Internals are more likely than externals to participate in efforts for social change, they tend to have higher incomes, and they a re more likely to drive alone. '!bey also have a hig her n eed to be independent and in control, making t hem less willing to depend on others for their transportation Ulberg stresses the use of caution when promoting ridesharing and transit in appealing to envirorunental concerns, as those who may respond to environmental messages may also b e the lea st like ly to swi tch to modes requ i ring dependence upon others. Also, indivi dua l differences in the perception o f one's ab ility t o control outside forces have imp lications for how decisions are made. The author states that p l anners should not assume that everyone feels their individual decisions have impacts on transportation. The influence of stress on transportation decisions is also an important individual difference. The general assumption i s that driving alone in congest ion is the most stressful mode of travel, but some may fmd i t relax i ng. Some fmd comfort and relaxa tion in the contact with other people provided by ridesharing, but it could be extremely stressful for others. Perceptions tend to be individualistic. The small body of research that has been done on the fulfillment of psychological need s with regard to transportation indicates that this is an important component of transportation choices. One theme i s the need for independence. Ulberg cites research, conducted twenty years prior to hi s assessment, which revealed independence as one of the most important factors distinguishing bus riders from auto commuters. However, the author notes that given rising congestion issues in urban areas, this may no longer be as obvious. Another psychological need for which humans seek fulfillment according to Ulberg, is the need for community. In some cases auto mobiles can become t h e focus o f community, such as in carpooling. The desirability of carpooling has been found to correlate with the number of acquaintances one may have in the carpool. Self-expression is another psychological need and the type of automobile chosen and the various ways of personalizing or decorating it may provide some fulfillment for the need of self express ion. However, there is no se l f-expression permitted in the use of mass transit. Further, the automobi l e satisfies the need of control over one's envi ronment. The control over the 41


envirorunent the automobile offers through mobility is reinforced by the ability of people to control the machine itself. Ulberg states that because these needs will always exist for at least some part of the population, alternative modes to the automobile should consider these needs as well. In addition, because they are describe d as psychological factors that relate to travel mode c hoices, such variables should be included in a hierarchy of transportation system use r needs. Ulberg stresses the importance of individual differences in mode choice decisions. Psychological and cu ltur a l differences may be correlated with demographic variables but do not vary directly with them. Even when considering demographic and situational variables, people exhibit variation in decision-making. The author states that it may be "impractical" to measure such differences in planning-oriented research, but they should be considered in the interpretation of research results or promotion design. In addition to the perceptions, demographic differences, and individual variations that exist in mode choice, consideration must be given to the cognitive processes that people use. There are different ways of gathering, processing, and storing information, and these differences can have an effect on mode choice decis i ons. Ulberg recognizes four parts to the cognitive process: Combining mu l tiple several d i mensions and making one choice, such as evaluating the time, cost, and convenience of various modes, somehow weighing them with respect t o each other, and making a choice; How choice affects attitudes and perception-mode choice can affect attitudes and perceptions and v ice versa. People tend to evaluate their chosen mode most positively. Choice docs affect perception and people tend to place the grea test importance on the attributes that support their mode cho ice ; Using know ledge-despite individua l characteristics and behaviors or experiences, percepti ons are also determined by the actual attributes of a particu lar mode. The presentation of information does not guarantee that i t will be incorporated into the dec isi on-making process. If the ad d ed knowledge results in new behaviors, the new behaviors will not endure if t hey do not satisfy other transportation requirements, whatever those may be; and, 42


Influence of habit-once a choice is made, attitudes and perceptions are typically changed so belief s support th e choice. People tend to e mphasize the importance of attributes that support their current choices and seek no new information t h a t would tend to change those choices. Habi t should not be overemphasized however, and maj or lif e changes such as moving or having c hildr en, may re sult in a change in habits. Ulberg concludes that understanding how people process informa tio n is critica l in th e study of mode choice b eh aviors and in promoting alternative transportation choices. Models that do n o t cons id er these factors cannot interpret th e pr oce ss of mod e choice a ccurately or re a listicall y and promotions that do not consider these factors will not b e effective. Koppelman and Pas (1980) contend that traditionalttave l demand models p rovide o n ly limited understanding of the behavio ral processes underlying travel decision-making. These models tend t o focus upon objective measures of system performance and d emographic characteristics. Furthermore, because such models i gnore measures of traveler attitudes, including perception s of service attri bute s and p erso nal feelings toward d ifferent servi ces, they do not re flect the wid e range of strategies that can be d esigned t o influence consumer travel behavior. The authors' focus is the c onsumer p roces s i n mode cho ice fo r nonwork and nonschool trips to a suburban central bus iness ar ea (Evanston Illinois ) They analyze the relationships among perceptions, feelings preference, and choice. C hoices among travel alternatives ar e made based upon perce ptions of the alternatives rather t han o n objectively measured characteristics. The formation of perceptions i s influenced b y both measured (demographic) and unmeasured (experience, p s ychological make-up) i ndividual c haracteristics, i n addition to modal attributes. Perceptions of alterna t ive mode choices the refore differ among individuals. Koppelman and Pas claim that transportation researchers have typi cally f ocused on a single psychologic a l aspect in ex pl a i ning travel b e havior, t h at being perceptio n s abou t the attr i b utes of the object. They contend that research in social psychology indicates there may be other psychological d imens i ons at work in dete rmining travel behavior. In add i tion to individua l percepti ons abou t object characteristics, the lik ing a n d dislik ing of the object may affect the a tti tud es one has about that object Further, social norms, s u c h as what one b e lieves o thers expect,


and one's personal nonns contribute to ex plaining behavior. The authors assert these may be contributory factors in mode choice decisions. The authors' study reveals a generally very positive attitude toward car travel, a less positive affect with regard to walking, and a relat ively neutral stance toward bus trave l There was a "high degree of sensitivity" toward major increases in gasoline prices, but "little sensitivity" toward lower bus fares, despite positive reaction to improved bus service. There was indication of willingness to carpool for some trips, as well as to reconsider the car as the primary mode, given cost increases. Koppelman and Pas interpret these fmdings as a general lack of commitment to participants' existing mode choices. With regard to personal normative beliefs, the overall feeling was that walking should take precedence over car or bus travel. As for social norms, most respondents d id not feel thei. r peers would be "surprised" at their regularly traveling by car. Psychological measures of mode feelings, in addition to consumers' perceptions of various mode performance characte ristics contribute to preferences and t herefore to consumers' travel mode choice decision-making process. Koppelman and Pas con t end that this has great i mplications with regard to strategy development because it implies that preferences can be changed without necessarily a l tering service charac teristics However, the authors state tha t it could suggest service improvements resulting in mode use changes could be limited if deep feelings about modes are not changed in the process. Mitche l son and Gauthier ( 1980) state that understanding the decision-making process in travel is crucial in examining the effects of system changes on all urban travel groups. They contend that psychologica l and situational variables are great influences upon travel mode choice. In contrast to Koppelman and Pas (1980), who stated that attribute perceptions were a commonly stud ied variable among transportation researchers, Mitchelson and Gauthier claim that this is one set of psy chological variab les that has been neglected in research. The authors assert that perceptual variables are extreme l y important because they form the init ial aspect of mode choice which the authors define a s the transformation of an objective set of variables into subjective ones that provide th e basis for distinguishing between alternative trave l modes In addition, perceptual variables include the evaluat i on of those alternatives and the final mode c hoice. M i tchelson and Gauthier refer to this transformation as the psychophysic al function and contend that understanding this p roce ss as it relates to travel behaviors is very important 44


The authors present the results of a two-phase survey research study in Columbus, Ohio, designed to establish the broader dimensions of mode choice, as well as the relationships between mode characteristics and the broad dimensions of travel mode image. Statistical and interpretive conside rations allow the authors to establish a six -dimensional space of travel mode cognition encompassing travel burden, safety, convenience, privacy, flexibility, and comfort. Ba sed upon tradeoff data and pereeprual variations along group membership similarities, Mitchelson and Gauthier determine there to be three aggregate market segments: a relatively unbiased group, a bus-biased group, and an auto-biased group. Coinciding with the findings of the other literarure, these autho r s report tha t individual choi ces are reinf orced by perceptions. The bus-biased group generally gave the best bus ratings and the worst auto ratings. The auto-biased group, in general, provided the best automobile ratings and the worst ratings for the bus. The unbiased group provided mea n s cores that generally fell between the extremes of the auto and bus ratings. Mitchelson and Gauthier assert their findings are "suggestive a t best" because of small samp l e sizes and the restrictions of using the trade--off method of subjec tive evaluation. They do contend that their results provide added support to modeling urban travel demands through disaggreg a te behavioral approaches Further, the authors state that their approach is useful to the task of interpre ting mode choice decisions and their re l ationships to the physical characteristics of travel modes According to Kuppam, Pendyala, and Rahinan (1999), despite the recognition of the importance of traveler attirudes, perceptions, and values in mode choice behavior, there has been relative l y little use of attiru dinal and preference data in planning practice The authors attribute th is to two primary reasons: household travel data collection often ignores such variables; and attitudinal and preference data are typically not considered useful in travel -demand forecasting because they are not as easily quantifiable or predictable as demographic variables. Kuppam etal used the 1991 data set of the Pugct Sound Transportation Panel (PSTP) because it contains detailed information reg arding attirudes, demographics, and travel behavior. As determined in the other srudies reviewed here, the authors found that overall, people tended to favor their chosen travel mode. While different aspects of a p a rticular mode were valued by those 45


who use them, such as bus users being more sensitive to pollution and travel costs than were single occupant vehicle (SOV) users, relatively similar values were placed on time by all modal groups. With the exception of"nonmotorized transport users," each group ranked highly the ability to arrive to their destination on time. In general, SOY and car/vanpool users were found to feel that riding the bus did not "fit their lifestyle," but they did agree that the bus is an enjoyable mode of travel at significantly higher rates than did those who use the bus However, they were not likely to stop driving as a result of higher gasoline prices, and were not as willing as bus users to pay higher tal

Kuppam, P endyala a n d Ralunan suggest that these fmdings may in d i cate a group o f commut ers that f eels "captive" to the SOV mode, and that would potentially abandon solo driving in favor of transit use. As in the previous stud ies d iscussed here, these a u thor s found that, in general, attitu des and p re ferences vary across demographic variables and that they may be of significan t influence in mode choice decisions. Evidence of this influ e nce is based upo n the results of l.ikelihood ratio Chisquar ed statistical tests These tests show ed that bot h att itu d inal a n d demographic variables are valuable in int erpretin g mode c h o ice beha viors, b ut the statistical significance of attitudinal fact ors was near ly twice that of demographic factors. The authors therefo re suggest that the omis sion of s u ch factors in re search and plarming may be more serious than the omission of demographic factors. N eveu eta/. (1 979) use per ceptual mapping techniques to ex pl ore the influence o f such aspects as comfort, convenience, and relia bili t y with regard to work trips Whi le they state that previous re se arch has examined thes e factors individ u a lly, the a uthor s contend theirs is the first study to consider a ll three as they j o intly contribute to mode choice decisions. A survey was conducted among downtown Chicago commuters to collec t perceptual data for the study. Factor analysis, preference r egression, and first-preference logit model s revealed that commuters did not perc eive the three v ariables as ind e pen dent, and there was significant ove rla pping of the p u blic perception of such abstract concepts. Neveu, eta/. observed that none of the dimensions of each of these factors could be cons id ered strictly b elo nging to comfort, convenience, or reliability They found that elements of each appea red on more than one dimension, such as wait time or frequency, thereby contra sting the standard belief on the part of resear c hers th at these are co nsidered separate aspects of mode c hoice. Th is literature review demons trates the importance of these p s y cholog ical factors to mode choice decisions. Th e implication of the term "level of serv ice" suggests that we are concerned \vith the appraisa l of existing (and p otential) user s of th e various modes. It is clea r that any method of combining le v el of service and, frankly, measuring LOS for t h e ind i vidua l modes, will prob a bly miss the mark if i t fai l s to take into account these factors even though the various authors may use d ifferen t factors o r imply a different ranking than proposed i n the Trans p ortation Sy stem Us er Hierarchy of Needs. The follo,ving sectio n outlines an approac h for a pilot project to ide n tify, 47


construct and apply a TransJXlrlation System User Hierarchy of Needs to facilitate assessing level of service equa lly across modes. 48


Pilot Test Approach Phase I Transportation System User Hierarchy of Needs Task I Literature Review The purpose of this task would be to identify the levels and potential indica t ors for each of the levels Transportation User Hierarchy of Needs. For example, the following factors might constitute threats to personal safety for the auto user Absolute vehicle speed s Vehicle speed differential T im e of day (e.g., is there concern with impaired drivers in the late e vening/early morning hours? Sunlight) Aggressive driving behavior (e.g., ex tensive weaving) Truc k -auto mix Vehicle spacing to speed r atio (e.g., can a 2-car length gap be maintained?) Pedestrian crossing behavior (e.g., jaywalking) Cond i tion of pavement (e.g., potholes) Shoulders LightingNisibility Presence of Bicyclists Weather (e.g., rain slick roads) Perception of crime in area given segment Similar factors could be identified for the other modes and the remaining Traveler H i erarchy of Needs. Task 2 Qualitative Research to Identify Key Ind icators This task would usc qualitative research techniques to identify the re lative importance of each of the indicators for each of the modes and levels. 49


Focus groups are the mo st common me t hod for collecting qualita t ive data to he l p identify the factors or features that should be evaluated in the data collection task. The typical focus group consists of8 to 10 people that have some common characterist i c(s) relating to the subject (e.g., transit riders). A tra ined modera tor leads the group through a series of questions "focused" on specific aspects of the issues. The information generate d by a group will be more valuable b ec ause the intera ctions of the participants produce significantly more data than individual interviews. The participants may be required to take som e notes and/or fill out f orms, b ut audio and/ o r video recording capture most of the discussion data. After the sessions, the moderator transcribes conversation(s) .from the tapes into documents that become the documentation of the sessions Given the potentially wide range of indicators to be considered and discussed for each leve l FDOT should consider using a more dynamic method of collecting qualitative feedback t o guide the develo pment of the sampling plan (i.e., what needs to be measured), Under the recommend ed dynamic me thod, participants use hand-held keypad (similar to a television remote) to submit their responses to questions posed by the modera to r. These term inals transmit indivi dual responses back to a computer that tabulates the data and then disp l ay s a summary of the results for audience and/or moderator to review. Similar to the traditional focus group method, the moderator leads the group through the focus questions bu t may rely on the responses to determine where to probe dee per Th is "dynamic" focus group approach d iffers from the typical" focus group in three ways: fostering participat ion, direct in g the d iscussion and capnuing results. The dynamic version helps capture the re s ponses from all participants, not on l y the most vocal. In a tradi t ional focus group with 1 0 participants and a 30-minute segment for a specific ques t ion, each participant would be enti tled to an average of 3 minutes to d i scuss. In fact, most groups include people who are very vocal and others who prefe r to sit qu ietly. The "polling'' nature of the transponder permits participants to all "ta l k" a t the same time \vithout in te rrupting the flow, so they don't have to hold onto their opinions until it's their tum. Th e responses also a re anonymous as far as the other participants arc co ncerned. In effect, this feature provides a safe haven f o r all ideas and p r ovides a environmen t conduct i ve for the free flow of d iscussion, so there's less worry about feeling foolish or saying something wrong. 50


Though a good moderator will draw out the quiet participants or at least seek more balance among all the partic ipants the risk is the "response" captured may not accurately reflect how the group feels. Given the wide range of possible response s (see the list of potential auto personal safety needs/concerns above),llying to identifY the key issues will b e crit ical in deve loping a sampling plan. A dynamic focus group approach helps the moderate direct the discussion better. There are two distinct phases of the "discussion". In the first phase, participants use the responders to "quantify" the extent of their input or reactions to a quest ion The results can be d isplayed so t hat no one's input is overlooked or lost. The responses are anonymous as far as the other participants are concerned. In effect, this features pro vides political amnesty for aU ideas and provides a safe environment for the free flow of ideas, so there's less worry about fee lin g foolis h or saying something wrong. The second phase of the discussion is a verbal conversation, where the moderator uses the results to get participants to talk about the respon ses When participants see the responses that others have entered, this "triggers" additional discussion. A third major difference is capturing of the results. Part ici pants record their own responses; all "votes" are captured without filtering or interpretation The computer syste m records and saves all information entered by the participants. This will allow the analyst to later view or cross tabu late h ow various individuals r esponded to several questions. The audio and/or video recordings become supporting sources of information for the session rather than the primary sources. Respon ses can be readily transferred to analytical and evaluation tools that allow the moderator and analysts to understand, clarify, rate and prioritize the group's input. Task 3 Data Co llection The sampling p lan (include method, sample size, etc.) for each o f the modes (auto transit, bike and walk) will depend on the indicators identified in Task 2. While existing LOS can be one factor to consider for stratifying the sample, i t shouldn't be the only criterion. The data collection plan will focus on approaches that the advisory committee most often referred to as "Drive for Science", "Ride Transit for Science", etc. It should be pointed out that some o f 51


the need s (e.g., cost and comfort and convenience) are not represented in the Bi cyc l e a n d Pedestrian LOS so the data collection effort shoul d includ e those modes as well. There are at least two approaches that could be considered (though both approaches need not be mutually exclusive): I. Use a sin gle pool of participants to participate in all four For Science" data collection exercises (i.e Drive for S ci ence, B ike for Sc ience, Wal k for Science and Ride Transit for Science). The use of a common pool will allow for relative co mpar is ons across modes. 2. Use a si mulator to replica te conditions that may be difficult to find in the field (e g., when driving threa t ens p erson a l safety etc.) For exa mple if "percep t ion of crime in the vicinity of a g i ven segmenf is considered to be a r ela tivel y impo rtant indicator of"auto safety need" based on the qualitative i mpact of Task 2, t hen the sample of segments to be represen te d under such sc e nario should take that i nto consideration. Task 4 Data Analysis The expec t at ion is the data coll ecte d under Task 3 will allow f or a quantitative-based "adjustment factor" so each LOS gradation means the same for all mode s Ideally, the analysis will reveal a model that takes the fonn a step function, where the re a re clear demarcations when there are changes in LOS. In effect, ana lysi s would refl ect the advisory committ ee's thinking that normalizing t h e LOS gradations across modes via user opinions (albei t via shifting of the sca les or applying s om e to be-de termined discount f a ctor) would create these ba nds of acce ptable standards gradations or overl ays (green, yell ow, red) as shown o n the slide rules. Task 5 Final Re11ort for Phase I This task would generate the models and adjustment factors. Phase II w ill apply the model in the field


Phase II Pilot Testing The second phase of the project would p ilot test the model application in up to 3 areas in Florida i n consultation with lo cal transportation staff and FOOT D i strict offices Final Thoughts As stated earlier, upon comp l etio n of the pilot testing, we hope to va l idate the quantitative-based "adjustment factor" so each LOS grada tion means the same for all modes. In tum t his approach to assess i ng l eve l of service equally across modes will reflect the adv i sory comm i ttee's thinking that normalizing the LOS gradations across modes via user ''hierarchy o f needs" factors (albe i t via shifting of the scales or applying some tobe determined discoun t factor) would create the bands of acceptable standards gradations or overlays (green, yellow, red) as shown on the sl i de rules If the pilot tests p r ove successful, such a sy stem would h elp Florida communities improve the applicat i on of LOS measures to assessing existing conditions, identifying roadways in need of improvement, and prioritizing construction projects It should also facilita t e the ability of dec i s i on makers to allow LOS trade-offs between modes and promote a balanced mul ti -modal transportation sy s tem. 53


References Cited Highway Capacity Manual 2000. Transportation Research Board. Mitchelson, Ronald L. an d Howard L. Gauthier. Examination of the Psychophy sica l Function in T ravel Mod e-C ho ice Behavior. In Transportation Research Record 750, 1980, pp. 27-33 Kuppam, Arun R., Ram M Pendyala, and She la Rahman. Analysis of th e Role of Trave ler A ttitu des and Perception s in Explaining Mode-Choice Behavior In Transportati o n Research Record 1 676, 1999, pp 68-76. Koppe lman, FrankS and Eric I. Pas. Travel-Choice B e havior: Models of Perceptions Feel i ngs, Preference, and Choice In Transportation Research Record 765, 1980, pp 26-33. Neveu, Alfred J., FrankS. Koppelman and Peter R Stopher Perceptions of Comf ort, Convenience, and Re lia bility for the Work Trip. In Transportation Research Record 723, 1979, pp. 59-63. Ulbcrg, Cy. Psychological Aspects of Mode Choice. Wash i ngton State Transporta tion Center, Seattle, in cooperation with the United Stat es Department of Transportat ion, Federal Highway Administration, 1989. 54

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader cam 2200349Ia 4500
controlfield tag 001 001927520
005 20080327154857.0
008 020207s2001 flu bt 000 0 eng d
datafield ind1 7 ind2 024
subfield code a C01-00133
2 local
b A87 2001
0 245
Assessing level of service equally across modes /
prepared by Center for Urban Transportation Research, University of South Florida ; Philip L. Winters ...[ et al.].
Tampa, Fla. :
Center for Urban Transportation Research, University of South Florida ;
[Springfield, Va. :
Available through the National Technical Information Service,
54 leaves :
ill. ;
28 cm.
White paper.
Prepared for Fla. Dept. of Transportation under contract no.
BC353 RPWO#15.
"December 2001."
Includes bibliographical references (leaf 54).
Also available online.
Urban transportation
z Florida.
Local transit
level of service.
intermodal transportation.
1 700
Winters, Philip L.
United States.
Dept. of Transportation.
Dept. of Transportation.
University of South Florida.
Center for Urban Transportation Research.
8 773
t Center for Urban Transportation Research Publications [USF].
4 856