1 DEVELOPMENT OF HUMAN FACTORS DESIGN GUIDELINES John Campbell October 24, 2007

1

DEVELOPMENT OF HUMAN FACTORS DESIGN GUIDELINES

John Campbell

October 24, 2007

2

IntroductionsPresenter, John L. Campbell

• Research Leader, Battelle Human Factors Transportation Center

• 22 years of experience in the development of transportation systems and human factors design guidelines.

• 1985 to 1991, conducted research and developed design guidelines for advanced automotive displays for Hughes Aircraft Company’s Display Systems and Simulation Laboratory.

• 1991 to 1994, led the conduct of human factors research projects at Anacapa Sciences, with particular emphasis on advancing human factors methodology for development of design guidelines and applying these advances to the development of automotive displays, icons, symbols, and secondary controls.

• Since joining Battelle in 1994, has had primary responsibilities for design guideline development and transportation research projects in the areas of Crash Avoidance Systems (CAS), Advanced Traveler Information Systems (ATIS), Icons, Intelligent Transportation Systems, and Driver modeling.

3

IntroductionsPresenter, John L. Campbell (cont.)

• Past guideline development efforts include:

– Human factors guidelines for road systems (TRB, 2007)

– Human factors lessons learned for crash warning systems (NHTSA, 2006)

– A CD-ROM-based interactive software tool for symbol design, Icon IDEA (Icon Development and Evaluation Assistant; FHWA, 2003)

– Human factors guidelines for in-vehicle icons (FHWA, 2002)

– Human factors design guidelines for in-vehicle navigationsystems (FHWA, 1999)

– Interface guidelines for intersection collision avoidance devices (NHTSA, 1997)

4

Discussion Topics

• Developing Human Factors Guidelines

• Examples of Past Guidelines

• Lessons Learned

5

Developing Human Factors GuidelinesRole of Guidelines

FINAL DESIGN SPECIFICATIONS

• Objectives & Tasks• Available Design Data• Design Process• Existing Constraints• “Givens” in Design• Diverse Designers

Design Environment

Original Research

Research Compilations

SystemDesign

User-CenteredGuidelines

Designer needs for content,

organization, and format

Formulation ofIndividualGuidelines

Integrative review of

data sourcesDatabase of Human Factors Research

TechnicalReports

Books

LiteratureReviews

Handbooks

Standards

JournalArticles

ConferenceProceedings

EXPERIMENTS

6

Developing Human Factors GuidelinesHistorical Perspective

• Despite increasing demands for HF design guidance, it has been difficult for the HF community to develop information that designers judge to be valuable.

– designers didn’t use and had little interest in available HF information (Meister & Farr, 1967).

– designers found much of the HF research literature difficult to understand and that few data were generalizable to specific issues of concern (Rouse & Cody, 1988)

– designers rated information as very costly to obtain, with low value; they, would only expend small amounts of effort to obtain the information (Burns & Vicente, 1994)

– general design principles and HF heuristics were not deemed useful by non-human factors designers (Campbell, Carney, & Kantowitz, 1997)

7

Developing Human Factors GuidelinesKey Challenges

• Identifying appropriate content for the guidelines

• Lack of generalizeable research data

• Developing selection criteria for choosing data sources to be used to produce guidelines

• Variability across guideline users

• Developing effective guidelines without restricting innovative and effective design

8

Developing Human Factors GuidelinesKey Assumptions

• System design will proceed with or without human factors inputs to the design process.

• The “best-available” human factors information is better than no HF information at all.

• Users should be able to determine the relative contribution of expert judgment and experience data in design guidelines.

• HF design guidelines are intended to augment, not replace, designer experience, skill, and judgment.

9

Developing Human Factors GuidelinesKey Steps

STEP INPUTS / CRITERIA / CONSIDERATIONS

What is the most appropriate content, format, and organization of the guidelines?

Interviews with representative end users Resolutions of competing requirements

How will guidelines be applied?

Identify design phases Identify key constraints Identify needs for human factors information

What data sources should be used?

Identify generally relevant data sources Selection strategy should be overly inclusive, rather than overly exclusive

How useful are individual data sources?

Relative weight assigned to individual data sources Rate or summarize a range of criteria

What should the guideline be?

Integrative review of data sources Expert judgment and heuristics Needs of end users

Conduct User Requirements Analysis

Identify Relevant Components of the

Design Process

Select Data Sources

Determine the Quality and Applicability of Data

Sources

Formulate the Guidelines

10

Step 1: Conduct User Requirements Analysis

Demographics (education, work experience) Primary Design Responsibilities Design process Current HF Inputs to design design Difficulties in acquiring or using HF design

information (and suggested solutions) Priorities for system functions Priorities for HF topics and design data Review of candidate guideline formats

Developing Human Factors Guidelines

11

Step 2: Identify Relevant Components of the Design Process

• Key steps

• Starting points or “givens” in their design effort

• Manner in which requirements are expressed

• Design constraints

• Relative priorities among potential topics


12

Step 3: Select Data Sources

• Acquire a comprehensive and representative data base of relevant sources

• Employ a broad, inclusive search strategy

• Sources obtained from:– in-house libraries

– university collections

– on-line data bases, NTIS, DTIC, DIALOG, SAE

– foreign collections


13

Step 4: Determine the Quality and Applicability of Data Sources

• Criteria include:–source type–intended application area–for empirical sources:

- study setting- independent and dependent variables- sampling approach- internal and external validity

–consistency of findings/recommendations to other sources


14

Step 5: Formulate the Design Guideline

• Conduct integrative review and analysis of data sources

• Consider the design constraints and uncertainties associated with specific guideline topics

• Consider the use of transformed data

• Identify any caveats, exceptions, interactions with other guidelines, special performance issues


15

START HFG topicremaining?

Select topicfrom approved

outline

Conductintegrativereview of

relevant datasources

RevisionRequired?

Identify specificcontent

Preparesupporting text

Conduct analyticalactivities, including

use of heuristics andexpert judgment

Formulate theDraft Guideline

Conduct Projectteam review

and revise asnecessary

PrepareGraphics

ConductWorking Group

review

YES

NO

Delivercompletedmaterials to

TRB

YES

NO

Developing Human Factors GuidelinesProcess for Recent Road System Guidelines

16

Presentation Format

HFG SIGHT DISTANCE Version 0.01

KEY COMPONENTS OF SIGHT DISTANCE

Introduction

Sight Distance (SD) is the distance that a vehicle travels before completing a maneuver in response to some roadway element or condition that necessitates a change of speed and/or path. Sight Distance is based on two key components:

1) A Reaction Time (RT) required to initiate a maneuver (pre-maneuver phase), and

2) The time required to safely complete a maneuver (Maneuver Time; MT).

The reaction time includes the time needed to see/perceive the roadway element, time needed to complete relevant cognitive operations (e.g., recognize hazard, read sign, decide how to respond etc.), and time needed to initiate a maneuver (e.g., take foot off accelerator and step on brake pedal).

Maneuver Time includes actions and time required to safely coordinate and complete a required driving maneuver (e.g., stop at intersection, pass a vehicle, etc). Typically, a vehicle maintains its current speed and trajectory during the reaction time phase, while changing its speed and/or path during the maneuver time phase.

Design Guidelines

Sight Distance = Distance traveled while driver perceives,

makes decisions about, and initiates action in response to roadway element (RT)

+ Distance traveled while the driver completes an appropriate maneuver

(MT)

Based Prim arily on Expert Judgm ent

Based Equally on Expert Judgm ent and Em pirica l Data

Based Prim arily on Em pirical Data

Based Prim arily on Expert Judgm ent

Based Equally on Expert Judgm ent and Em pirica l Data

Based Prim arily on Em pirical Data

SCHEMATIC SHOWING THE REACTION TIME AND MANEUVER TIME COMPONENTS OF SIGHT DISTANCE

Sufficient Sight Distance

Insufficient Sight Distance

A

B

R eaction Tim e

M aneuverTim e

R eaction Tim e

M aneuverTim e

H azard(2ft h igh)

D river ’s Eye(3.5 ft H igh)

L ine of S ight

Diagram A: The hazard is visible to the driver far enough away that there is sufficient distance for the driver to recognize and react to the hazard and to complete the maneuver necessary to avoid it. Diagram B: Because of the steeper vertical crest, the driver’s sight distance is shorter than in Diagram A making it possible for a hazard to be hidden from sight until there is insufficient distance to avoid it. *Note: distances not to scale

5-1

HFG SIGHT DISTANCE Version 0.01 Discussion

Before drivers can execute a maneuver, they must first recognize there is a need for some action and decide what that action should be. Therefore, this mental activity–perception, cognition, and action planning–precedes an overt vehicle control action and takes some amount of time. The reaction time is typically defined as the period from the time the object or condition requiring a response becomes visible in the driver’s field to view to the moment of initiation of the vehicle maneuver (e.g., first contact with the brake pedal). Although a particular reaction time value (e.g., 2.5 s from AASHTO 2004) is used in deriving sight distance requirements for a given design situation, this “reaction time” value should not be viewed as a fixed human attribute, since it is influenced by many factors. Some the of the key factors that influence reaction time are shown in the table below.

FACTORS THAT AFFECT THE DIFFERENT COMPONENTS OF REACTION TIME

Factor Explanation

Low contrast (e.g., night) It takes longer to perceive low-contrast objects

Visual glare Objects are perceived less quickly in the presence of glare

Older Age Older drivers less sensitive to visual contrast and are more impaired by visual glare (e.g., oncoming headlights)

Object size /height Smaller objects/text require drivers to be closer to see them

Driver expectations It takes substantially longer to perceive unexpected objects

Seeing/ Perceiving

Visual complexity It takes longer to perceive objects “buried” in visual clutter

Older age Older drivers require more time to make decisions Cognitive elements Complexity

Drivers require more time to comprehend complex information or situations and to initiate more complex or calibrated maneuvers

Initiating Actions

Older age Older drivers require more time to make vehicle control movements and they may be limited their range of motion

In contrast to the reaction time, the maneuver time is primarily affected by the physics of the situation, including vehicle performance capabilities. In particular, tire-pavement friction, road-surface conditions (e.g. ice), downgrades, etc. can increase maneuver time or make some maneuvers unsafe at higher speeds. Maneuver time is also affected to a lesser extent by driver-related factors (e.g., deceleration profile), but these factors are highly situation specific since the maneuvers are very different (e.g., emergency stop, passing, left turn through traffic etc.). These factors are covered in more detail in the relevant guideline sections (see GL…).

Design Issues

It is important to note that although most design requirements are expressed as a design distance, from the driver’s perspective the critical aspect is time. It takes time to recognize a situation, understand its implications, decide on a reaction, and initiate the maneuver. While this process may seem almost instantaneous to us when driving, it can translate into hundreds of feet at highway speeds before a maneuver is even initiated. Speed selection is also critical, since the relative speed between the driver and the hazard determines how much distance is traversed in the time it takes the driver to initiate and complete the maneuver (see Speed GL).

Cross References

Specific types of sight distance (pg. 5-X, 5-X…); Greenbook section on calculating sight distance

Curves, Traffic engineering elements (signs), decision sight distance? (these are not currently included as HFG topics)

Key References

None

5-2

Right-hand page

References

CrossReferences

DesignIssues

Discussion

AbbreviatedHandbook Title(Both Pages)

AbbreviatedChapter Title(Both Pages)

Revision Version

(Both Pages)

Guideline TitleBar Scale Rating

Introduction

DesignGuideline

Left-hand pagePage Numbers

Figure,Table, orGraphic


17

ExamplesIcon Guidelines for the FHWA

Low Clearance Ahead

Effective Use of Text Label

Overheight vehicles takeanother route aroundrestricted clearance.

Ineffective Use of Text Label

Use both uppercase andlowercase letters.

Avoid using italicsto emphasize words.

The space between lines should beat least 1/30 the line length.

Make the type >0.27 degreesvisual angle.

Use a clear andsimple font.

Keep text labels brief - use no more than 2 to 3 words.

Use of Text Labels in Icon Design

18

ICON DESIGNPARAMETER

RECOMMENDATION DO THIS. . . . . .NOT THIS

Figure/GroundRelationship

Emphasize a clear, stable, and solidrelationship between a symbol andits background.

FigureBoundaries

Solid shapes are better than thin lineboundaries or dotted line boundariesunless the element in question isdepicting action or movement.

ClosureUse closed figures instead of figureswith discontinuous lines, outlines, ordisjointed elements.

Simplicity Icons should be simple with only thenecessary detail included.

Unity All parts of the symbol should beenclosed within a single boundary.


Perceptual Principles for Icon Design

19

DESCRIPTION EXAMPLE

Common boundary identifies agroup of icons as conveyingwarning information.

Icons identified by a common levelof detail and abstraction.

Icons identified by a commonstyle.

Icons grouped by border,background, color, and style.


Identifying Icons as Part of a Group

20

ExamplesCWS “Lessons Learned” for NHTSA

ON OFF

9mDistance

Forward SensorNeeds Calibration

Display Type

Explanation Example

Analog Display

Provides a graphical representation of continuous information.

Scale-based FCW

Discrete Display

Provides binary on/off information. “Vehicle detected” status indicator

Digital Display

Information is presented directly as a number.

Headway distance display (in meters)

Alphanumeric Display

Information is presented as messages in full or abbreviated form.

Complex system error message

Symbol/Icon Simple graphic signs that transmit message information. Note that some simple and familiar alphanumeric displays (e.g., Stop!; Brake!) may function as symbols.

FCW Icon

Selecting Display Types for FCW Devices

21

Examples CWS “Lessons Learned” for NHTSA

Recommended Detection Zones for LCW Devices

F ast A p p ro ach Z o n eP ro x im ity Z o n e

3 0 ft fro m rea r b u m p er(ad ju stab le to 2 0 ft)

11 ft

F ast A p p ro ach Z o n eP ro x im ity Z o n e

1 6 2 ft

A n y v eh ic le th e s ize o f a b icy c le o r la rg e r sh o u ld b e d e tec ted in e ith e r zo n e .

* R e fe ren ce 5 fo u n d th a t ex ten d in g th e p ro x im ity zo n e in fro n t o f th e v eh ic le w as u se fu l, h o w ev er, th e p rec ise d is tan ce is y e t to b e d e te rm in ed .

*

22

Examples CWS “Lessons Learned” for NHTSA

Design of CWS Controls

A. Well-designed Control Placement

B. Poorly-designed Control Placement

Controls are aligned with forward viewControls are within fingertip reachHigher priority controls are on the outside (easier to manipulate)Controls can be activated with both hands on the wheelControls are coded by location for easy identificationAdapted from Reference 6

Controls are not aligned with the forward view (driver must look away and down to see controls and display)Placement requires increased glance timeControls are partially obscured by steering wheelDriver must reach to operate controlControls are poorly or not labeledInappropriate control type for on/off switch

23

ExamplesHF Guidelines for Road Systems (TRB/NCHRP)

Schematic showing the Perception-Reaction Time and Maneuver Time Components of Sight Distance

Insufficient Sight DistanceB

Perception R eaction Tim e

M aneuverTim e

Sufficient Sight DistanceA

Perception R eaction Tim e

M aneuverTim e

Hazard(2ft h igh)

D river ’s Eye(3.5 ft H igh)

L ine of S ight

24


Time Gap (tg) (seconds) at design speed of major road

Design Vehicle Left Turn Right Turn

Passenger car 7.5 6.5

Single-unit truck 9.5 8.5

Combination truck 11.5 10.5

Acceptable Gap Distance

25


Vision-im paired pedestrians wa it longer for crossable gaps because they cannot extend gaps that are in itia llytoo short w ith eye gazes and m anual gestures in the sam e w ay that s ighted pedestrians can.

Traffic no ise from inside the circ lecan m ask sound cues from oncom ingvehic les, especia lly qu iet hybrid vehiclesor vehic les coasting dow nhill

Countermeasures for Improving Accessibility of Vision-impaired Pedestrians at Roundabouts

26

Lessons Learned

• Involving end-users in the guideline development and review process is crucial to success.

• Human factors design guidelines based on the "best-available" human factors research data can still provide useful information to designers.

• The human factors discipline has the tools needed to provide clear, relevant, and useful design guidelines to the system design community.

Documents

1 DEVELOPMENT OF HUMAN FACTORS DESIGN GUIDELINES John Campbell October 24, 2007