Embed Size (px)
Citation preview
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
4
The author is Director ofEnterprise Risk andSafety Management,Washington StateDepartment ofTransportation, Olympia,and serves as Chair ofthe TRB Highway SafetyPerformance Committeeand of the NationalCooperative HighwayResearch ProgramProject Panel onDevelopment of aComprehensive Approachfor Serious Traffic CrashInjury Measurement andReporting Systems and ofthe Project Panel onInterchange SafetyAnalysis ToolEnhancement.
In December 1999 a group of highway safety aca-demics, practitioners, and agency representativesconvened in Irvine, California, to discuss the cre-
ation of a comprehensive and authoritative manualthat would serve the field of highway safety in thesame way that the Highway Capacity Manual hadserved the field of operations. This group became thecore of a larger team that has focused substantialamounts of professional and volunteer time on thisvision.
The vision became reality 11 years later, with thepublication of the Highway Safety Manual (HSM).The HSM represents a dedicated effort to create anddrive the science of safety into practical use and tointegrate safety as a key consideration in any high-way project, program, or activity. The developmentof the HSM focused on the quantification of safetyperformance, beyond the traditional safety paradigm.
To succeed, the HSM had to become a tool forprofessionals in planning, design, operations,
maintenance, and system management. The plannerand designer would be able to consider reconstruc-tion and upgrades to facilities and to compare theanticipated safety performance with that of the cur-rent facility. Summarizing crash history and relyingon nominal assessments of what would improvesafety performance no longer sufficed. The HSM hadto improve the reliability of information, enablingsubstantive assessments that would reduce fatal andserious injuries. In addition, the HSM had to offerapproaches to assess a facility’s safety performance.
The HSM represents the cooperative efforts of theAmerican Association of State Highway and Trans-portation Officials (AASHTO), the Federal HighwayAdministration (FHWA), and the TransportationResearch Board (TRB). Technical content was devel-oped through a major research effort funded andmanaged through the National Cooperative High-way Research Program (NCHRP). The TRB High-way Safety Performance Committee, the AASHTOHSM Task Force, and the FHWA Office of Safetycontinue the collaboration as new partners becomeinvolved in implementation and in the developmentof the second edition of the HSM.
Scientific BasisBefore the HSM, safety program and project deci-sion making relied on a variety of sources and meth-ods to assess steps and actions to reduce the potentialfor crashes. The assessments varied in effectivenessand did not maximize reductions in total crashes.Finding science-based, state-of-the-practice meth-ods, approaches, and knowledge was difficult.
When safety decisions are not quantified, trade-off decisions among traffic operations, environmen-tal stewardship, or cost are difficult and often havebeen based on perception. A scientific basis wasneeded for decision making. Highway and roadwaysafety is a priority for most state DOTs, but budgetsare shrinking for all programs. Decision makingmust be reliable and optimal.
The Highway SafetyManualImproving Methods and ResultsJ O H N C . M I L T O N
Blueprints to Improve
HIGHWAYSAFETY
The Highway SafetyManual, published in2010, was developed as acomprehensive tool forthe integration of safetyinto highway planning,design, operations,maintenance, and systemmanagement.
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
5
The HSM was designed for use throughout a proj-ect’s development. The intended users are profes-sionals in planning, design, construction, operations,and maintenance of the roadway and highway sys-tems at the state, county, and local levels.
Absolute safety does not exist; the primary objec-tive of highway safety is to lower the risk for crashfrequency and crash severity. The safety professionalrelies on three primary components to reduce risk:the human, the vehicle, and the roadway. The HSMfocuses on the roadway but recognizes the impor-tance of the vehicle, of human factors, and of theinterventions of education, enforcement, and emer-gency medical services.
Core PrinciplesThroughout the development of the HSM, particularattention was given to the reliability of the analysisas key to reducing fatalities. As a result, the HSM isa fundamentally unique document, presenting theevolution of approaches, methods, and knowledge inhighway safety. At its core are two principles:
u Allow the user to base decisions on actual antic-ipated changes in crash frequency and severity; and
u Address the statistical issues that have a directimpact on reliability.
Nominal and Substantive SafetyMost highway programs have addressed safetythrough the application of standards. Standards oftengenerically refer to design policies, guidelines, andcritieria that are part of an agency’s design and traf-fic operations manuals; to the AASHTO Policy onGeometric Design of Highways and Streets; to theAASHTO Roadside Design Guide; and to the Manualon Uniform Traffic Control Devices. Each of these doc-uments, however, differs in purpose, use, and inter-pretation.
The assumption was that the application of adesign standard provided safety. Ezra Hauer aptlytermed this “nominal safety” in the 1990s. Many ofthese standards were not developed to address crashfrequency and severity but operational, environ-mental, economical, and other considerations.
Hauer advocated that “substantive safety” wasmore appropriate for addressing crashes. Substan-tive safety provides a statistically reliable assessmentof safety performance to inform an agency’s plan-ning, project development, and traffic operations.
In applying design standards, the designerassumes that the resulting design will be safe—thatmeeting the design manual requirements equalssafety. Within this context, however, the designedfacility is only nominally safe. For example, 12-ftlanes are the design standard for a particular facilitytype; a design for 11-ft lane widths, which wouldrequire a design exception, represents an absolutechange in safety—the nominal crash risk would besignificantly higher.
Substantive safety, in contrast, relies on scientificfindings for determining the anticipated safety per-formance, measured in crash frequency and severity.With the substantive safety approach, the designer
Efforts to reduce crashrisk focus on the human,the vehicle, and theroadway.
PHOTO
: IS
TOCK
PHOTO
: N
ATIONALH
IGHW
AYTR
AFF
ICSA
FETY
ADM
INISTR
ATION
A first responder records information at the scene ofa crash. From emergency response services toroadway design, highway safety decision makingmust be reliable.
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
6
uses the HSM to determine the effect of a change inlane width on crash frequency and severity; for agiven facility type and volume, the difference in theanticipated crash outcomes may be neglible. In thesubstantive safety approach to design and systemmanagement, crash frequency and severity are criti-cal in decision making.
Figure 1 (above) illustrates the difference betweennominal and substantive safety with the 12-ft versus11-ft lane width example. The line labeled as nomi-nal safety indicates that the crash risk associated withthe 11-ft lane is substantially higher than that for the12-ft lane. Under the nominal safety approach, the11-ft lane would be deemed unsafe, and the 12-ftlane safe. The curve labeled substantive safety showsthe actual anticipated crash performance as a con-tinuum—as lane width increases, the crash riskdecreases.
Regression to the MeanHistorically, potential projects underwent a rankingof crash frequency or rate, followed by a “black spot”analysis that used short periods of observed crashhistory—typically two to five years—to identify loca-tions that might benefit from changes to the road orroadside environment that would reduce crash fre-quency or severity. This method has been moder-ately successful, but using limited data can lead tolarge fluctuations in crash rates, as shown in Figure2 (below, left).
Determining locations for safety intervention fromsmall sets of data can lead to unreliable assumptionsabout safety performance if the analysis does notaccount for regression to the mean. An agency cannotknow with certainty that the site will respond, or thatany changes are directly caused by the treatment.
Crashes are random events; therefore crash fre-quency at a given site will vary from year to year; aproject selected on the high side of this variance willshow an average number of crashes that exceedsexpectation. As shown in Figure 3 (below, right),accounting for regression to the mean produces a dif-ferent expected average crash frequency, but one thatis more indicative of the true safety performance at thesite. Therefore the project is more likely to show acrash reduction with the appropriate countermeasure.
Accounting for regression to the mean and incor-porating the substantive safety philosophy improvesdecision making. The result is fewer crashes on thehighway system and a reduction in fatal and seriousinjuries. Improving an agency’s ability to incorporatesafety into decision making translates into increasedeffectiveness in expenditures, increased awarenessof trade-offs, and the ability to inform the public—including stakeholders and elected officials—aboutthe benefits and impacts of programs and projects.
Nominal safety isan ABSOLUTE
Substantivesafety is a
CONTINUUM 12-ft lane width
11-ft lane width
Design Dimension(Lane Width, Radius of Curve, Stopping Sight Distance, etc.)
CR
ASH
RIS
KG
reat
er
Greater
FIGURE 1 Nominal versussubstantive safety.(Source: CH2M Hill,2011.)
Ob
serv
ed C
rash
Fre
qu
ency
Years
Ob
serv
ed C
rash
Fre
qu
ency
Years
FIGURE 2 Average short-term crash frequency (redhorizontal rules).
FIGURE 3 Expected average crash frequency (bluehorizontal rule).
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
7
New Thought ProcessThe HSM presents a new thought process for manyorganizations that will require some changes to pol-icy, procedures, and methods. The document’s struc-ture follows the project development process formost transportation organizations. The three-vol-ume HSM comprises 17 chapters sorted into fourparts (see box, above).
Basic ConceptsPart A introduces the basic concepts of highway safe-ty, with chapters on human factors and safety fun-damentals. Improving safety performance requires anawareness throughout the project developmentprocess of user capabilities, limitations, and interac-tions with the roadway.
Chapter 1—Introduction and Overview definesthe purpose and intended audience of the HSM. TheHSM is a technical document that requires a level ofunderstanding for its use. The chapter highlightstypes of applications, the scope of the document,and its relationship to project development.
Chapter 2—Human Factors introduces the basicprinciples, indicating the effect of human factors andof the ways that designs and operations can accountand compensate for limitations. The goal is to reducethe potential for errors—and consequently reducecrashes—related to human factors. The NCHRPReport 600 series, Human Factors Guidelines for RoadSystems, supplements this chapter.1
Chapter 3 introduces the concepts that supportthe application of the methods, approaches, andknowledge presented in the HSM.
Road Safety ManagementChapters 4 through 9 address road safety manage-ment, following the steps commonly taken in systemand project planning, program management, pre-liminary engineering, and operations.
The first step generally involves screening for theprojects most likely to respond to design modifica-tions. Chapter 4—Network Screening outlines theprocess, which includes establishing the reason,determining the reference population, selecting per-formance measures, and selecting the screeningmethods. The chapter summarizes the state of thepractice and shows how to incorporate safety into theranking process, to focus resources where the returnis highest.
After the projects are ranked, the next step is tounderstand the factors contributing to crashes acrossthe location, corridor, or system. Chapter 5—Diag-nosis describes how to analyze historic safety per-formance to gain insight into the circumstancescontributing to observed crashes. Properly identify-ing the patterns, target crash types, and the con-tributing circumstances increases the probability that
Highway Safety ManualContentsVolume 1Part A. Introduction, Human Factors, andFundamentals
Chapter 1. Introduction and OverviewChapter 2. Human FactorsChapter 3. Fundamentals
Part B. Roadway Safety ManagementProcess
Chapter 4. Network ScreeningChapter 5. DiagnosisChapter 6. Countermeasure SelectionChapter 7. Economic AppraisalChapter 8. Prioritize ProjectsChapter 9. Safety Evaluation
Volume 2Part C. Predictive Methods
Chapter 10. Predictive Method for RuralTwo-Lane, Two-Way Roads
Chapter 11. Predictive Method for RuralMultilane Highway
Chapter 12. Predictive Method for Urbanand Suburban Arterials
Volume 3Part D. Crash Modification Factors
Chapter 13. Roadway SegmentsChapter 14. IntersectionsChapter 15. InterchangesChapter 16. Special Facilities and Geomet-
ric SituationsChapter 17. Road Networks
1http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_600A.pdf.
A multilane, dividedhighway in a rural area.Understanding thespecific factors thatcontribute to crashes in ahighway system is anessential step in theproject planning process.
PHOTO
: AAA F
OUNDATION
FORTR
AFF
ICSA
FETY
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
8
the factors contributing to crashes can be addressedin the best way possible.
Selection of a potential countermeasure is thethird step, discussed in Chapter 6. Countermeasuresaim to reduce the frequency and severity of crashesfor a specific site, or they may represent a trade-offbetween crash severity and frequency. Selecting theappropriate countermeasures maximizes the poten-tial for reducing crashes and their severity.
The economic analysis methods in the HSM allowcomparisons of a countermeasure’s anticipated bene-fits and costs. Chapter 7—Economic Appraisal pre-sents cost-effectiveness and benefit–cost analysismethods. Although costs are typically valued in mon-etary terms, agencies often incorporate nonmonetaryconsiderations. Project comparisons can use commonperformance metrics and analysis techniques.
Chapter 8 outlines project prioritization, neces-sary when several candidate sites have been identi-fied on the network, the economic evaluations havebeen performed, and the economically feasible can-didate sites have been selected and are consistentwith organizational policies. The process allows anagency to identify the projects with the greatestpotential benefit across the system.
Evaluating the change to the system in terms ofcrash or injury reduction helps an organization tounderstand safety performance and to account forsafety in decision making. This approach is morelikely to succeed in improving system performance.Chapter 9 presents different evaluation methods andapproaches, reviews the strengths and limitations ofeach, and supports the development of crash modi-fication factors (CMFs). Evaluating safety effective-ness provides agencies with feedback for policies anddecision making.
Assessing Safety PerformancePart C—Predictive Methods applies to rural two-lane, two-way roads (Chapter 10), rural multilanehighways (Chapter 11), and urban and suburban
arterials (Chapter 12). The predictive methodassesses anticipated safety performance—that is, thepredicted average crash frequency—and theexpected average crash frequency, where applicable.
The estimates can be used to assess a network,corridor, or site, for existing and future conditions,with or without proposed countermeasures, as wellas the effectiveness of potential countermeasuresunder current and future conditions. Part C allowsthe assessment of the safety performance of newfacilities under various traffic volumes.
Safety TreatmentsPart D—Crash Modification Factors presents thepotential effects of safety treatments and of opera-tional and other site-specific changes on crashes forroadway segments (Chapter 13), intersections(Chapter 14), interchanges (Chapter 15), specialfacilities and geometric situations (Chapter 16), androad networks (Chapter 17). Each chapter exploresthe potential safety impacts in terms of the qualityand the statistical reliability of the research.
Part D presents a subset of the CMFs available inpractice. All information in the HSM underwentthorough screening and expert review to assure reli-ability of the information and guidance. Part D issupplemented with information from the FHWACrash Modification Clearinghouse2 and the NCHRPReport 600 series.
The HSM is scalable; some organizations or
Developed for trainingpurposes as part of aNational CooperativeHighway ResearchProgram (NCHRP) project,spreadsheetsincorporating locallyderived values forOregon can be used as acompanion tool to PartC—Predictive Methods ofthe HSM. Part C assistspractitioners in assessingthe safety performanceof rural, two-lane roads,as well as rural highwaysand urban and suburbanarterials.
Raised medians were installed along New HampshireAvenue in Washington, D.C., in 2009; safetytreatments are examined in the HSM Part D—CrashModification Factors.
PHOTO
: O
REG
OND
EPARTM
ENT
OFTR
ANSP
ORTA
TION
PHOTO
: WAYANV
OTA
2www.cmfclearinghouse.org.
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
9
agencies may choose to implement only a portion ofthe tools. For instance, an organization may decideto focus on the planning and scoping of projects cov-ered in Part B and on the reliable approaches incor-porated in Part C; others may focus on design andapply Part C and Chapter 2—Human Factors.AASHTO developed Figure 4 (right) to show poten-tial links to the project development process.
Software ToolsSeveral software tools were developed to ease appli-cation of the HSM and are an integral part of thepackage. Following are some of the noncommercialsoftware applications.
Safety Analyst is a highway safety managementsoftware tool that performs functions outlined inPart B, Chapters 4 through 9, of the HSM. SafetyAnalyst was developed through a pooled-fund effortby states and is offered as part of AASHTOWare,AASHTO’s suite of software for designing and man-aging transportation infrastructure projects.3
The Interactive Highway Safety Design Model(IHSDM), a software tool developed and made avail-able at no cost by FHWA, is used to evaluate thesafety and operational effects of design decisions andincludes modules on crash prediction, design con-sistency, intersection review, policy review, trafficanalysis, and driver–vehicle issues. The IHSDM–HSM Predictive Method 2011 (Version 7.0.0) is avail-able online.4 The crash prediction module incorpo-rates Part C, Chapters 10 through 12, of the HSM.
Under NCHRP Project 17-38, Highway SafetyManual Implementation and Training Materials,Karen Dixon of Texas A&M Transportation Institutedeveloped spreadsheets to assist in performing thePart C predictive methods. Alabama, Virginia, Illi-nois, and Washington State DOTs have applied andexpanded the spreadsheets.
The FHWA Crash Modification Factors Clear-inghouse is a web-based database that allowssearches for CMFs and assists in selecting the mostappropriate measure.2 The site employs a five-starrating system and includes supporting research.
Supporting DocumentsTo assist in the implementation of the HSM, docu-ments have been created for a variety of levels withina state DOT, focusing on the HSM as part of the proj-ect development process. The documents are avail-able on FHWA’s HSM website5 or AASHTO’sHighway Safety Manual website.6
The HSM outreach materials provide a quickpoint of reference and summary information to easeadoption of the HSM, including the HSM OverviewBrochure, HSM Overview Fact Sheet, and An Intro-duction to the Highway Safety Manual.
FHWA has developed three HSM guides. TheHSM Managers Guide (2011) provides information tostate DOT managers charged with incorporating theHSM into the project development process. Integrat-ing the HSM into the Highway Project DevelopmentProcess (2012) helps transportation professionalsapply the HSM in planning and scoping, environ-mental studies, predesign, final design, and day-to-day maintenance and operations activities. The HSMTraining Guide (2011) was developed for state andlocal agencies considering implementation of theHSM; it documents training needs, sequences oftraining, delivery methods, and courses offered byFHWA’s National Highway Institute and the Insti-tute of Transportation Engineers.
Case studies complement the guides, demon-strating applications of the HSM, including the road-way safety management process, predictive methods,and the development of safety performance func-tions and implementation plans. FHWA also spon-sors a user discussion forum on the AASHTO HSMwebsite.7
HSM Application: Part B• Identify sites most likely to benefitfrom safety improvement
• Identify targeted crash patterns forthe network
• Prioritize expenditures for efficiency
System PlanningIdentify needs and program projects
HSM Application: Parts B and C• Identify crash patterns at existinglocations
• Evaluate safety effectiveness ofpotential countermeasures
• Modify policies and design criteria forfuture planning and design
Operations and MaintenanceModify existing conditions to
maintain and improve safe and efficient operation
HSM Application: Part B• Identify targeted crash patterns forthe project
• Evaluate countermeasures’ costs andeffectiveness
• Compare change in crash frequency topredict safety effect of alternatives
Project Planning and Preliminary Engineering
Identify alternatives and choose thepreferred solution
HSM Application: Part C• Evaluate how performance measuresare impacted by design changes andconstruction
• Assess potential change in crashfrequency during design exceptionevaluation
Design and ConstructionDevelop design plans and build projects
FIGURE 4 HSM and theproject developmentprocess. (Source:AASHTO.)
3www.aashtoware.org/Pages/default.aspx.4www.ihsdm.org.5http://safety.fhwa.dot.gov/hsm.6www.highwaysafetymanual.org/Pages/default.aspx. 7www.hsmforum.org.
Research Supporting the HSMTRB’s Transportation Research Circular E-C142,Methodology for the Development and Inclusion ofCrash Modification Factors in the First Edition of theHighway Safety Manual (2010), provides backgroundon the development of Part D of the HSM anddescribes the CMF literature review and inclusionprocess.8 This document offers a framework for thereview of future safety publications, to determinestatistical reliability, the characteristics of reliableCMF results, higher-quality methods to advance thescience of safety, and improvements for later editionsof the HSM.
During the development of the first edition, sev-eral professionals expressed concern about employ-ing the monetary costs of crashes to justify programsand initiatives. In moving from a qualitative to aquantitative approach, however, economic appraisalis a key in decision making. Societal cost figures areoften used to show the benefits of crash reductionand to prioritize projects. In 2011, NCHRP Project20-24(068) developed a simple calculator to assiststate DOTs in estimating the crash costs by state,using site and safety attributes such as posted speedlimit, geometry, traffic control, signalization, crashtype, and maximum severity of injury.9
CMFs are critical in quantifying the impacts onsafety and directly affect the reliability of analyses.FHWA published A Guide to Developing QualityCrash Modification Factors (2010) to explain theprocess and the issues to consider in selecting eval-uation methods; the document covers the back-ground, definitions, purpose, use, and concerns indeveloping CMFs.10
In addition, FHWA sponsored research in support
Safety AnalystSoftware Tools for Safety Analysis of Specific Highway SitesD O U G L A S W . H A R W O O D A N D D A R R E N J . T O R B I C
Safety Analyst consists of a set of software tools available from the Amer-ican Association of State Highway and Transportation Officials (AASHTO)
to help highway agencies analyze the safety effects of infrastructure improve-ments at specific sites on a highway and street network.a The tools were devel-oped through a Federal Highway Administration (FHWA) pooled-fund studythat ran from 2001 to 2009 with the participation of 27 state highway agen-cies and other stakeholders.
The software includes four modules that together address the six steps ofthe safety management process described in Part B of the Highway SafetyManual (HSM):
u Module 1: Network Screening reviews the entire highway network andidentifies sites with the greatest potential for safety improvement.
u Module 2: Diagnosis and Countermeasure Selection diagnoses thesafety concerns at particular sites and assists in selecting effective counter-measures to reduce the frequency and the severity of crashes.
u Module 3: Economic Appraisal and Priority Rankingperforms cost-effec-tiveness and benefit–cost analyses for a specific countermeasure or for sev-eral alternative countermeasures at specific sites and ranks countermeasuresand sites to assist highway agencies in setting investment priorities.
u Module 4: Countermeasure Evaluation performs before-and-after eval-uations to document the effectiveness of implemented safety improvements.
The modular package allows for flexibility—users can apply Safety Ana-lyst at any stage of the safety management process.
The Safety Analyst database consists of the highway agency’s records ofroadway segment, intersection, and ramp characteristics; traffic volumes; andcrashes for its entire system. The software includes tools to import and man-age data from the highway agency’s databases. When Safety Analyst identi-fies a need, the project can be designed with supporting analyses from FHWA’sInteractive Highway Safety Design Model (IHSDM; see article, page 11).
The data requirements for Safety Analyst are less extensive than for thepredictive models in Part C of the HSM and in the IHSDM; project designrequires more extensive data. Agencies that lack the detailed crash data tosupport Safety Analyst can apply the usRAP Tools software, available from theU.S. Road Assessment Program (see article, page 15), to perform the steps ofnetwork screening, countermeasure selection, and economic appraisal.
The Safety Analyst software can be licensed for use by highway agenciesthrough the AASHTOWare program.b In addition to permitting use of thesoftware, the license provides access to technical support, including engi-neering support from MRIGlobal and computer software and data manage-ment support from ITT Excelis.
The authors are with MRIGlobal. Harwood is Transportation ResearchCenter Director, Kansas City, Missouri; and Torbic is Principal TrafficEngineer, State College, Pennsylvania.
awww.safetyanalyst.org.bFor licensing information, contact Vicki Schofield at AASHTOWare, [email protected].
8http://onlinepubs.trb.org/onlinepubs/circulars/ec142.pdf.9http://usroadsafety.com/crash-cost-calculator.10www.cmfclearinghouse.org/collateral/CMF_Guide.pdf.
Brochures, fact sheets, and otheroutreach materials help statesand localities assist in adoptingthe HSM.
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
10
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
11
T he Interactive Highway Safety Design Model (IHSDM) is asuite of software tools that support project-level decisions
about the geometric design of roadways. The model providesquantitative information on the expected safety and opera-tional performance of a design. Produced by the Federal High-way Administration’s Safety Research and DevelopmentProgram, the IHSDM is a resource for the predictive methoddescribed in Part C of the American Association of State High-way and Transportation Officials’ Highway Safety Manual (HSM).
IHSDM assists projectdevelopers in making de-cisions that improve thesafety performance of de-signs. The software toolsalso help project planners,designers, and reviewersjustify and defend deci-sions about geometric design.
The 2011 IHSDM com-prises six evaluation mod-ules: Crash Prediction, Pol-icy Review, Design Consis-tency, Intersection Review,Traffic Analysis, and Driver–Vehicle. The Crash Predic-tion Module (CPM) is a software implementation of HSM PartC, which includes crash prediction methodologies for rural two-lane and multilane highways and for urban and suburban ar-terials.
The IHSDM includes a recently developed calibration util-ity to assist agencies in implementing the proceduresdescribed in the Part C Appendix. Agencies can enter theirown safety performance functions and modify the defaultcrash distributions. Efforts are under way to extend the CPMto include freeway, ramp, and interchange crash predictioncapabilities from new HSM chapters developed under NationalCooperative Highway Research Program Project 17-45,Enhanced Safety Prediction Methodology and Analysis Toolfor Freeways and Interchanges.
Two recent case studies highlight the use of IHSDM to imple-ment HSM Part C methods:
u The Louisiana Department of Transportation and Devel-opment (DOTD) conducted a safety analysis to quantify thebenefits of constructing an alternative route from Interstate 12to Bush. The CPM produced estimates of the crashes expectedin 2035 for the study area network, for the no-build condition,and for four alternative scenarios. The total network crash per-
formance was evaluated and compared for each scenario. Anestimated cost per crash by severity was applied to the pre-dicted number of crashes, and the alternatives were ranked bycost savings. Louisiana DOTD was able to identify the safestand most cost-effective solution.
u The Idaho Department of Transportation (DOT) conducteda comprehensive review of conditions on the State Highway 8corridor to identify and prioritize operational improvementsover a 10-year period. The IHSDM crash predictions were based
on the existing traffic, roadway geometry, and recent crash his-tory. The output indicated that for more than half of the 11-milecorridor, the calculated crash rate was higher than the statewideaverage. Potential locations were identified for improvement.“The advantage of employing IHSDM was the opportunity toperform a detailed and simultaneous review within the corridoron a variety of critical elements—for example, traffic operations,geometry, and safety—to isolate potential problem areas andallow the development of strategic mitigation strategies,” theproject manager noted.
IHSDM training consists of a two-day course onsite or a web-based version led by experienced instructors. Participants learnabout key IHSDM capabilities and limitations, evaluate high-ways using IHSDM, and recognize when and how the modulecan be used during the project development process.
Resourcesu For more information about the Idaho DOT case study, visit http://
safety.fhwa.dot.gov/hsm/casestudies/id_cstd.cfm.u For more information about training courses, see FHWA-NHI-380071
and -380100 in the National Highway Institute catalog, http://nhi.fhwa.dot.gov.
u Additional resources are available at http://www.fhwa.dot.gov/research/tfhrc/projects/safety/comprehensive/ihsdm/index.cfm andhttp://www.ihsdm.org.
u Contact: Clayton Chen, 202-493-3054; [email protected].
The Interactive Highway Safety Design ModelC L A Y T O N C H E N
The Interactive HighwaySafety Design Model’s CrashPrediction Module presentsmethodologies for highwaysand arterials; efforts areunder way to incorporatecrash prediction capabilitiesfor ramps, freeways, andinterchanges as well.
Blueprints to Improve
HIGHWAYSAFETY
PHOTO
: W
ASH
INGTO
NST
ATE
DOT
The author is Highway Research Engineer, Federal HighwayAdministration, McLean, Virginia.
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
12
of the Highway Safety Improvement Program. Inves-tigation of Existing and Alternative Methods for Com-bining Multiple CMFs (2010) presents issues in theapplication of multiple CMFs and offers guidance forestimating the effects of combined treatments.11
NCHRP Project 20-7(314), Recommended Pro-tocol for Developing Crash Modification Factors, hasoutlined a process for developing CMFs. The find-ings describe the necessary documentation and howto address potential biases.
The NCHRP Report 600 series focuses on roaduser capabilities and on limitations in road designand operations.12 The guidelines are useful for diag-nosing the contributing factors in collisions and forthe selection of countermeasures.
Training and ImplementationTraining is a critical component of the HSM.
NCHRP Project 20-7(290), Highway Safety Train-ing Synthesis, surveyed the safety training avail-able and developed a roadmap and central databaseof the information. NCHRP Project 17-38, Devel-opment of Overview Training for the HSM, assistedFHWA and others in preparing to implement theHSM.
The first of many research publications sup-porting HSM implementation was NCHRPResearch Results Digest 329, Highway Safety Man-ual Data Needs Guide.13 The guide assists users inunderstanding the data needed for the Part Cmethodologies and explains the difference betweenavailable data and the data needed in the future.
NCHRP Project 17-50, Lead States Initiative forImplementing the HSM, started after the successfulSafety Performance Function Summit hosted by Illi-nois DOT through the efforts of State Safety EngineerPriscilla Tobias; the summit focused on the use ofquantitative tools in highway safety. The project hasestablished a dialogue among states that have expe-rience in quantifying safety, to inform other states,
11www.cmfclearinghouse.org/collateral/Combining_Multiple_CMFs_Final.pdf.12http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_600A.pdf; http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_600B.pdf; and http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_600C.pdf. 13http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rrd_329.pdf.
WIK
IMED
IACOM
MONS
A pedestrian refugeisland in a roadway.Safety for nonmotoristsalso is part of the NCHRPresearch portfolio.
T he Roadside Safety Analysis Program(RSAP) was developed to update and
enhance the cost-effectiveness analysisalgorithms and procedures of the ROAD-SIDE software package included in theRoadside Design Guide, published by theAmerican Association of State Highway andTransportation Officials (AASHTO). The pro-gram and its interface recently have under-gone an extensive upgrade, incorporatingnew research results and enhancing thecomputing capabilities and ease of use.
Version 3 of RSAP (RSAPv3) soon will be available electroni-cally at no charge from AASHTO to purchasers of the RoadsideDesign Guide. The National Cooperative Highway Research Pro-gram is preparing to publish a final report that contains a user’smanual with detailed, how-to explanations and an engineer’smanual tracing the technical background behind the code, aswell as the methodologies for the analyses. The software, man-uals, and example problems are available on the website of theAASHTO Technical Committee for Roadside Safety.a
As the benefit–cost tool for the 2011 edition of the RoadsideDesign Guide, RSAPv3 assists in performing roadside safety eco-
nomic analyses and serves as an alterna-tive to the warranting approach alsoincluded in the guide. RSAPv3 incorpo-rates substantive advances in roadsidesafety in the decade since the previousversion; this required rewriting the codeand developing new methods and tech-niques.
An encroachment-based approach,RSAPv3 divides collisions into three independent events:
1. The encroachment, when the vehicle first leaves the road;2. The traversal of the roadside, where hazards may be
located; and3. The severity of the crash when a vehicle intersects a road-
side hazard.
RSAP performs this series of calculations many times, simu-lating tens of thousands of encroachments for a typical roadwaysegment and estimating the crash costs of each possibleencroachment. After generating all the encroachments and theestimated crash costs, the program produces an estimate of thetotal crash cost for the segment.
RSAPv3 can evaluate up to five design alternatives on up to
Roadside Safety Analysis Program, Version 3Upgrading a Tool for Roadside Safety Design
M A L C O L M H . R A Y , C H R I S T I N E E . C A R R I G A N , A N D C H U C K P L A X I C O
ahttp://design.transportation.org/Pages/RoadsideSafety.aspx.
Version 3 of theRoadside SafetyAnalysis Program(RSAPv3) wasdeveloped as a toolfor the 2011Roadside Design
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
13
and is developing a user guide for the HSM. TheFHWA Pooled Fund Study for HSM Implementationis under way to advance the lead states initiative andto expand implementation to all states.14
Toward Future EditionsNCHRP Project 20-7(279), Work Plan for the 2ndEdition of the HSM, is a comprehensive effort toassess future research needs for the manual. Surveysof safety professionals have helped the AASHTOTask Force on the HSM and the TRB Highway SafetyPerformance Committee prioritize research efforts.
The first edition of the HSM included only a fewroadside countermeasures. NCHRP Project 17-54,Consideration of Roadside Features in the HighwaySafety Manual, recognizes the role of roadside featuresin crashes and is exploring the differences between theHSM and the NCHRP-developed Roadside SafetyAnalysis Program (RSAP), including the comparativestrengths and weaknesses of each and ways to addressthe differences. The project also is identifying ongoingroadside research, CMFs research, and needed CMFs
and aims to develop CMFs.15
NCHRP Project 17-56, Development of CrashReduction Factors for Uncontrolled Pedestrian Cross-ing Treatments, looks to quantify the relationshipbetween pedestrian crashes and crossing treatments.The research will evaluate various crossing treatmentsand will develop CMFs for crash type and severity.16
The ability to assess crash injuries accurately iscritical in the selection of appropriate countermea-sures; however, police reports of crash injuries oftenlack accuracy. NCHRP Project 17-57, Development ofa Comprehensive Approach for Serious Traffic CrashInjury Measurement and Reporting Systems, is devel-oping a framework for moving to International Sta-tistical Classification of Diseases and Related HealthProblems codes and to provide a process for linkingcrash data with hospital discharge data. The findingsmay provide a new injury scale for the HSM.17
14Transportation Pooled Fund 255; www.pooledfund.org/Details/Study/484.
20 highway segments and can compare the benefit–cost ratioof each alternative. The program identifies the alternative thatmakes best use of the funds.
The encroachment module in RSAPv3 takes into account thefrequency of roadside encroachments by highway type andtraffic volume. Adjustment factors are included for the effectsof horizontal curvature, grade, number of lanes, lane width,access density, and posted speed limit.
RSAPv3 incorporates data on vehicle trajectories during anencroachment. The trajectories are superimposed on the user-entered data for roadside terrain, and the program assesses allpossible interactions of trajectories with user-entered hazards.
RSAPv3 replaces the subjective severity index with an objec-tive fatal crash cost ratio based on observed, police-reportedcrash data. RSAPv3 also accounts for unreported crashes—whichcan be considered roadside safety successes. The programincludes a preloaded selection of crash severity models for manycommon roadside hazards, such as trees, utility poles, guardrails,and bridge piers.
The final report and its appendices document the methodsand procedures in developing RSAPv3 and include a selectionof case studies to illustrate use of the program. RSAPv3 providesroadside designers with an effective tool for making decisionsabout roadside safety designs.
ResourcesMak, K. K., D. L. Sicking, and B. A. Coon. NCHRP Report 665: Identifica-
tion of Vehicular Impact Conditions Associated with Serious Ran-off-Road Crashes. Transportation Research Board of the National
Academies, Washington, D.C., 2010. http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_665.pdf.
NCHRP Project 17-43: Long-Term Roadside Crash Data Collection Program. http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=1637.
Roadside Design Guide, 3rd Edition. American Association of StateHighway and Transportation Officials, Washington, D.C., 2006.
The authors are with RoadSafe LLC, Canton, Maine.
A median cable barrier on I-470 in Ohio prevented this tractor-trailerfrom reaching the opposite lanes of the highway in a crash. Roadsidetraversals are among the collision events analyzed by RSAPv3.
PHOTO
: MIK
ELEE
15http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=2979.16http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3178.17http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3179.
TR N
EWS
282
SEPT
EMBE
R–O
CTO
BER
2012
14
NCHRP Project 17-59, Safety Impacts of Intersec-tion Sight Distance, is reviewing the relationshipbetween safety and available intersection sight dis-tance. Products include guidelines for various inter-sections and other factors and conditions, plus CMFsor other appropriate functions.18
NCHRP Project 17-62, Improved Prediction Mod-els for Crash Types and Crash Severities, is develop-ing models to supplement or to replace those listed inthe HSM. The research will produce new SPFs or dis-tributions for predicting crash severity on a facility.19
NCHRP Project 17-63, Guidance for the Develop-ment and Application of Crash Modification Factors,is exploring procedures for formulating, calibrating,and using new CMFs for multiple-treatment applica-tions.20
Potential New ChaptersNCHRP Project 17-45, Enhanced Safety PredictionMethodology and Analysis Tool for Freeways andInterchanges, has produced two new chapters for theTRB Safety Performance Committee and AASHTO toconsider for the HSM. The chapters will likely becomepart of the IHSDM predictive methods module. Inaddition, the research has enhanced the InterchangeSafety Analysis Tool.21
NCHRP Project 17-58, Safety Prediction Modelsfor Six-Lane and One-Way Urban and Suburban Arte-rials, is pursuing the continued development of thepredictive methods chapters. The research willdevelop crash frequency and severity models for seg-ments and intersections, as well as new chapters forthe HSM.22
An Evolving ScienceThe HSM was created to reduce fatal and serious crasheson the nation’s highways. The manual has changed thefield of highway safety by introducing the means toquantify the impacts of safety in common practice andto use this information throughout the program andproject development process. The HSM cannot be a static document; changes will occur as new informationdevelops and the science of safety evolves.
The development and implementation of the HSMhas spurred many research efforts and partnerships.The dedication of AASHTO, FHWA, and TRB in thepublication of the first edition is evident. Volunteersand staff have devoted and contributed countlesshours to the development of the HSM and to the effortto improve the science of highway safety. The TRBHighway Safety Performance Committee encouragesparticipation in HSM activities and in the nation’sToward Zero Deaths vision.23
Several NCHRP projectsare under way to informfuture editions of theHSM; one is examiningthe relationship betweenpedestrian crashes andcrossing treatments.
Training Materials for Applying the Highway Safety Manual
R eleased by the American Association for State Highway andTransportation Officials in 2010, the Highway Safety Manual
(HSM) presents tools to facilitate decisions on roadway planning,design, operations, and maintenance that focus on their safety con-sequences. Because a key component to the manual’s success iswide dissemination and comprehensive understanding of its tech-niques, the National Cooperative Highway Research Program(NCHRP) initiated an outreach and training project, summarized inNCHRP Report 715, Highway Safety Manual Training Materials.
Training materials include basic introductory information, aswell as specific content for advanced procedures. The report fea-tures 12 informational modules addressing the development ofthe HSM, safety fundamentals, network screening, predictivemethods, human factors, economic appraisal and prioritization,specialized procedures, diagnosis and countermeasure selection,safety effectiveness evaluation methods, crash prediction proce-dures, candidate crash modification factors, and introductoryinformation on safety assessment procedures. Training materialsinclude presentation slides with speaker notes, participant hand-outs, interactive sample problems, and smart spreadsheets.
The report emphasizes the importance of clear, concise, andobjective language in safety instruction and documentation.Phrases and terms used to describe transportation scenariosoften have connotations beyond their intended meaning—thereport also highlights phrases that can be perceived incorrectly.
For more information on NCHRP Report 715 or to downloadthe included CD-ROM, visit www.trb.org/Publications/Blurbs/167185.aspx.
19http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3420.20http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3421.21http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=2512.22http://apps.trb.org/cmsfeed/trbnetprojectdisplay.asp?projectid=2759.23www.safetyperformance.org.
PHOTO
: WASH
INGTO
NSTA
TEDOT
18http://apps.trb.org/cmsfeed/trbnetprojectdisplay.asp?projectid=3181.
TR NEW
S 282 SEPTEMBER–O
CTOBER 2012
15
T he new highway bill signed in July increases funding forhighway safety and emphasizes the value of state strategic
highway safety plans for all public roads. This will generatedemand for new and innovative safety management tools, suchas usRAP—the U.S. Road Assessment Program.a
The AAA Foundation for Traffic Safety completed an eight-state pilot test in 2010 to explore the benefits of usRAP. Mod-eled on programs in Europe and Australia and in effect in morethan 70 countries, usRAP uses crash history data or roadwayinventory data to assess and benchmark the relative safety ofroads. State and county engineers and other key stakeholdersare represented on the technical advisory panel that has guidedthe program.
Implementation of usRAP is under way across the country. InMichigan and Illinois, the usRAP team is building on a success-ful project in Kane County, Illinois, and is working with thestates’ departments of transportation and several counties toassist in the development of county-level strategic highwaysafety plans. These efforts will demonstrate the utility andapplicability of usRAP to county and local jurisdictions, whichoften do not have adequate crash data to deploy more tradi-tional analytical tools, such as the American Association of StateHighway and Transportation Officials’ Safety Analyst.
Through the Road Protection Score, usRAP identifies roadsegments with higher crash potential, analyzing road inven-tory data for the absence or presence of design features thatcorrelate strongly with the risk of serious crashes. By generat-ing a safety investment plan, usRAP offers cost-effective
options—that is, road safety improvements—for engineers toconsider for lowering the identified risks.
In Utah, the usRAP team has trained state safety engineersto perform analyses and produce color-coded risk maps fromhistorical crash data. The team plans additional training sessionsfor state and county engineers, as well as for consultants inter-ested in integrating the usRAP tools into their procedures.b
These efforts complement and supplement other highwaysafety management practices and help state and local jurisdic-tions respond to the new emphases on performance metrics andtransparency, with the move nationwide to enhance trafficsafety culture and to pursue the Toward Zero Deaths agenda.
The author is President and CEO, AAA Foundation for TrafficSafety, Washington, D.C.
Road Safety Management Tool Assists State and Local Engineers P E T E R K I S S I N G E R
awww.usrap.us.
bTo collaborate with usRAP or to participate in the training, contact the AAAFoundation at [email protected] or 202-638-5944.
The usRAP risk map for Utah shows historical crash data in the state.
An introductory video on the U.S. Road Assessment Program (usRAP)website provides an overview of the program. usRAP was pilot-testedin eight states and currently is being implemented across the country.
IMAGE: USRAP
IMAGE: U
SRAP
Blueprints to Improve
HIGHWAYSAFETY
