Handbook of Industrial Automationedited - Chapter 9 pdf

Due to the wide variety of available tech-nologies, and due to the fact that most ergonomicsrecommendations for the design of displays and con-trols remain the same regardless of the tec

1.1.1 Importance and Relevance of Human

Factors Considerations in Manufacturing

Systems Design

The design and operation of manufacturing systems

continue to have great signi®cance in countries with

large and moderate manufacturing base, such as the

United States, Germany, Japan, South Korea,

Taiwan, and Singapore It was widely believed in the

1980s that complete automation of manufacturing

activities through design concepts such as ``lights-out

factories,'' would completely eliminate human

in¯u-ence from manufacturing, and make manufacturing

more productive [1] However, we now see that

com-plete automation of manufacturing activities has not

happened, except in a few isolated cases We see three

basic types of manufacturing systems present and

emergingÐthe still somewhat prevalent traditional

manual manufacturing mode with heavy human

in-volvement in physical tasks, the predominant hybrid

manufacturing scenario (also referred to traditionally

as the mechanical or the semiautomatic systems) with

powered machinery sharing tasks with humans, and

the few isolated cases of what are called

computer-integrated manufacturing (CIM) systems with very

little human involvement, primarily in supervisory

capacities Indeed, human operators are playing, and

will continue to play, important roles in manufacturingoperations [2]

Another important factor that prompts due eration of human factors in a manufacturing system,during its design, is the recent and continuous upwardtrend in nonfatal occupational injuries that has beenobserved in the manufacturing industry in the UnitedStates [3] While these injuries may not be as severe andgrave as the ones due to accidents such as theChernobyl Nuclear Reactor accident (the Three MileIsland nuclear accident prompted an upswing inhuman factors research, especially in nuclear powerplants and in process industry settings), the increasingtrend in injuries leaves the claim that ``automation'' ofmanufacturing has resulted in softer jobs for manufac-turing workers questionable In fact, many manufac-turing researchers and practitioners believe that anincrease in severe injuries in manufacturing is primarilydue to the automation of simpler tasks, leaving thedicult ones for the humans to perform This belief

consid-is logical as the technology to automate dicult tasks

is either unavailable or expensive

The factors discussed suggest that manufacturingsystems (our de®nition of a system is broad; a systemmay thus be a combination of a number of equipment/machines and/or humans) be designed with humanlimitations and capabilities in mind, if the system is

to be productive, error-free, and safe, and result in749

quality goods and services, all vital goals for

manufac-turing organizations

1.1.2 The Human±Machine System Framework

for Interface Design

Traditionally, system designers have accounted for

human limitations and capabilities by considering the

human operator as an information processor having

sensory and motor capabilities and limitations

(Fig 1) It can be readily seen from Fig 1 that the

key elements to the ecient and error-free functioning

of a human±machine system are the provision of

infor-mation to human operators in the system, and the

provision for control of the system by humans

Displays provide information about the machine or

the system to human operators, and controls enable

human operators to take actions and change machine

or system states (conditions) Operator feedback isobtained through interaction with the controls (tactilesensing, for instance) Thus, in the classical view,human interaction with automation is mediatedthrough displays and controls for a two-way exchange

of information

The recent view of the human±machine system,resulting out of advances in computerized informa-tion systems, sees the human operator as a super-visory controller [4] responsible for supervisoryfunctions such as planning, teaching, monitoring,intervening, learning, etc (Fig 2) Even though, insuch a view, the human operator has a changed role,displays and controls still provide the fundamentalmedium for human interaction with the system.Indeed, properly designed displays and controls arefundamental to the ecient and error-free functioning

Figure 1 Traditional representation of human interaction with machine

of manufacturing systems Ergonomics, which we

de®ne as the study of issues involved in the application

of technology to an appropriate degree to assist the

human element in work and in the workplace, provides

recommendations for interface design based on

research in human sensory and motor capabilities

and limitations

1.1.3 Scope of This Chapter

Even though displays and controls, and their e€ective

design, are fundamental to the ecient and error-free

operation of the system, a number of important

activ-ities need to be carried out before one can think of

displays and controls These activities stem from the

central need to build systems to suit human

limita-tions and capabilities Some of these activities, such

as ``user needs analysis,'' are relatively new conceptsand form the core of what is called the ``usabilityengineering approach'' to design Techniques asso-ciated with other activities, such as task analysisand function allocation between humans and auto-mated equipment, are an integral part of designing

``good'' jobs, and have been in existence for sometime We present some of these techniques and meth-ods

Inherent throughout our presentation is the essence

of the ``human-centered interface design approach.''

We ®rst present elements of this approach and trast it with the ``system-centered interface designapproach.'' It is recommended that this concept ofhuman-centered design guide the designer at boththe system, as well as at the nuts-and-bolts, designlevels

con-Displays and controls, the selection, design, andevaluation of which will be the theme for the remainder

of the chapter, form a part of aids, equipment, tools,devices, etc., that are necessary for a system to operatesatisfactorily Due to the wide variety of available tech-nologies, and due to the fact that most ergonomicsrecommendations for the design of displays and con-trols remain the same regardless of the technology used(e.g., recommendations on the design of letteringremain the same whether the recommendation is for

a conventional hand-held meter, a visual display unit,

or printed material), we provide only general mendations for di€erent types of displays and controls,without reference to commercial products andequipment

recom-A few other notes about the scope of this chapter:due to the vast literature available in the area ofdesign of human±machine systems, our emphasis inthis chapter is on the breadth of coverage ratherthan depth in any area This emphasis is deliberate,and is motivated, in addition, by our intention toprovide the reader a taste of the process of designand evaluation of a modern human±machine system.Readers interested in more detail in any one area ortechnique should refer to our recommended readinglist Also, even though the recommendations andguidelines summarized in this chapter come fromresearch in human±machine settings other than hard-core manufacturing settings, they are equally applic-able to manufacturing systemsÐthe generalframework and the speci®c recommendations wehave collected and provided in this chapter fordesign of human±machine systems are applicableacross systems

Figure 2 The latest notion of human as a supervisory

controller

1.2 APPROACHES TO DESIGNING

SYSTEMS FOR HUMAN±MACHINE

INTERFACE

1.2.1 The System-Centered Design Approach

The system-centered design approach, as the name

sug-gests, analyzes the system currently in use, designs and

speci®es the new system based on this analysis, builds

and tests the new system, and delivers the system and

makes minor changes to the system (Fig 3) The focus

is on the goals of the system and the goals of the

orga-nization within which the system is to perform

Designers following this approach fail to consider the

users before designing the system As a result, users of

such systems are required to remember too much

infor-mation Also, typically, these systems are intolerant of

minor user errors, and are confusing to new users

More often than not, such systems do not provide

the functions users want, and force the users to

per-form tasks in undesirable ways New systems designed

the system-centered way have also been shown to cause

unacceptable changes to the structure and practices in

entire organizations [5]

1.2.2 The Human-Centered Design Approach

The human-centered design approach to human±

machine interaction, unlike the system-centered

approach, puts the human attributes in the system

ahead of system goals In other words, the entire

system is built around the user of the systemÐthehuman in the system This approach has been var-iously called the ``usability engineering approach,''the ``user-centered approach'' or the ``anthropocentricapproach to production systems,'' etc Figure 4 pro-vides our conception of the human-centered approach

to interface design The ®rst step in this designapproach is information collection Informationabout user needs, information about user cognitiveand mental models, information on task demands,information on the environment in which the usershave to perform, information on the existing interfacebetween the human operator (the user of the system)and the machine(s), requirements of the design, etc.,are some of the more important variables aboutwhich information is collected This information isthen used in the detailed design of the new interface.The design is then evaluated Prototype developmentand testing of the prototype are then performed just as

in any other design process User testing and tion of the prototype, the other important characteris-tic of this design process which calls for input from theend user, is then carried out This results in new input

evalua-to the design of the interface, making the entire designprocess iterative in nature

Even though the human-centered design approach

is intended to take human capabilities and limitationsinto account in system design and make the systemusable, there are a number of diculties with thisapproach The usability of the system is only as good

as its usability goals Thus, if the input from the usersabout the usability goals of the system are inappropri-ate, the system will be unusable One approach to over-come this problem is to include users when settingusability goals; not just when measuring the usabilitygoals Another common diculty with this approach isthe lack of provision to take into account qualitativedata for designing and re®ning the design This is due

to the de®ciency inherent in the de®nition of usabilitywhich calls for quantitative data to accurately assessthe usability of a system There is also the drawbackthat this approach is best suited for designing newsystems, and that it is not as e€ective for redesign ofexisting systems

Despite these limitations, the human-centereddesign approach merits consideration from designersbecause it proactively takes the user of the product(displays and controls with which we are concerned,and which make up the interfaces for human±machineinteraction, are products) into the system designprocess, and as a result, engineers usability, into theproduct

Figure 3 System-centered approach to design

1.3 THE PROCESS OF SOLVING HUMAN±

MACHINE INTERFACE PROBLEMS

Even though displays and controls are the ®nal means

of information exchange between humans and

machines in a system, the actual design of the

hard-ware and softhard-ware for displays and controls comes

only last in order, in the process of solving human±

machine interface problems The other key steps in

this process include user-needs analysis, task analysis,

situation analysis, and function allocation decisions,

after which the modes of information presentation

and control can be decided In the following sections,

we discuss each of these steps

1.3.1 User-Needs AnalysisThe goal of user-needs analysis is to collect informa-tion about users and incorporate it into the designprocess for better design of the human±machineinterface User-needs analysis typically involves thefollowing activities: characterization of the user,characterization of the task the user performs, andcharacterization of the situation under which the userFigure 4 Human-centered approach

has to perform the task What follows are

guide-lines and methods for performing each of these three

activities prior to designing the system

1.3.1.1 Characterization of the User

Table 1 provides a user characterization checklist

Included in this checklist are questions to elicit

infor-mation about the users, inforinfor-mation about users' jobs,

information about users' backgrounds, information

about usage constraints, and information about the

personal preferences and traits of the users

As is obvious from the nature of the questions in

the checklist, the goal of collecting such information

is to use the information in designing a usable

system

1.3.1.2 Characterization of the Task

Characterization of the tasks users have to perform to

attain system goals is done through task analysis Task

analysis is defned as the formal study of what a human

operator (or a team of operators) is required to do to

achieve a system goal [6] This study is conducted in

terms of the actions and/or the cognitive processes

involved in achieving the system goal Task analysis

is a methodology supported by a number of techniques

to help the analyst collect information about a system,organize this information, and use this information tomake system design decisions Task analysis is anessential part of system design to ensure ecient ande€ective integration of the human element into thesystem by taking into account the limitations and cap-abilities in human performance and behavior Thisintegration is key to the safe and productive operation

of the system

The key questions to ask when performing task lysis activities are shown inTable 2 The task analysismethodology ®nds use at all stages in the life cycle of asystemÐfrom initial conception through the prelimin-ary and detailed design phases, to the prototype andactual product development, to the storage and demo-lition stage Task analysis is also useful for systemevaluation, especially in situations involving systemsafety issues, and in solving speci®c problems thatmay arise during the daily operations of a system.Task analysis can be carried out by system designers

ana-or by the operations managers who run the system on aday-to-day basis

Table 1 User Characteristics ChecklistData about users What is the target user group?

What proportion of users are male and what proportion are female?

What is average age/age range of users?

What are the cultural characteristics of users?

Data about job What is the role of the user (job description)?

What are the main activities in the job?

What are the main responsibilities of the user?

What is the reporting structure for the user?

What is the reward structure for the user?

What are the user schedules?

What is the quality of output from the user?

What is the turnover rate of the user?

Data about user What is the education/knowledge/experience of the user relevant to the job?

background What are the relevant skills possessed by the user?

What relevant training have the users undergone?

Data about usage Is the current equipment use by users voluntary or mandatory?

constrains What are the motivators and demotivators for use?

Data about user What is the learning style of the user?

personal What is the interaction style of the user?

preferences and What is the aesthetic preference of the user?

traits What are the personality traits of the user?

What are the physical traits of the user?

Adapted from Ref 5.

While many di€erent task analysis techniques exist

to suit the di€erent design requirements in systems, our

primary focus here is on techniques that help in

design-ing the interface The key issues involved in designdesign-ing a

human interface with automated equipment are

asses-sing what will be needed to do a job (the types of

information that human operators will need to

under-stand the current system status and requirements; the

types of output that human operators will have to

make to control the system), and deciding how this

will be provided Table 3provides a summary of the

important activities involved in the process of interface

design and the corresponding task analysis technique

to use in designing this activity We present brief maries of each of these techniques in the followingsections The reader should refer to Kirwan andAinsworth [6], or other articles on task analysis, for adetailed discussion of the di€erent task analysis tech-niques

sum-Hierarchical Task Analysis This enables the analyst

to describe tasks in terms of operations performed bythe human operator to attain speci®c goals, and

``plans'' or ``statements of conditions'' when each of

a set of operations has to be carried out to attain anoperating goal Goals are de®ned as ``desired states of

Table 2 Checklist for Task Analysis Activities

Goals What are the important goals and supporting tasks?

For every important task:

Intrinsics of the task What is the task?

What are the inputs and outputs for the task?

What is the transformation process (inputs to outputs)?

What are the operational procedures?

What are the operational patterns?

What are the decision points?

What problems need solving?

What planning is needed?

What is the terminology used for task speci®cation?

What is the equipment used?

Task dependency and What are the dependency relationships between the current task and the other tasks and systems?criticality What are the concurrently occurring e€ects?

What is the criticality of the task?

Current user problems What are the current user problems in performing this task?

Performance criteria What is the speed?

What is the accuracy?

What is the qualityTask criteria What is the sequence of actions?

What is the frequency of actions?

What is the importance of actions?

What are the functional relationships between actions?

What is the availability of functions?

What is the ¯exibility of operations?

User discretion Can the user control or determine pace?

Can the user control or determine priority?

Can the user control or determine procedure?

Task demands What are the physical demands?

What are the perceptual demands?

What are the cognitive demands?

What are the envirornmental demands?

What are the health and safety requirements?

Adapted from Ref 5.

systems under control or supervision'' (e.g., maximum

system productivity) Tasks are the elements in the

method to obtain the goals in the presence of

con-straints (e.g., material availability) Operations are

what humans actually do to attain the goals Thus,

hierarchical task analysis is ``the process of critically

examining the task factors, i.e., the human operator's

resources, constraints and preferencesÐin order to

establish how these in¯uence human operations in

the attainment of system goals.'' System goals can be

described at various levels of detail (or subgoals), and

hence the term ``hierarchical.'' The hierarchical task

analysis process begins with the statement of overall

goal, followed by statements of the subordinate

opera-tions, and the plans to achieve the goal The

subordi-nate operations and the plans are then checked for

adequacy of redescription (of the goal into

subopera-tions and plans) The level of detail necessary to

ade-quately describe a goal in terms of its task elements

determines the ``stopping rule'' to use when

redescrib-ing A possible stopping rule could be when the

prob-ability of inadequate performance multiplied by the

costs involved if further redescription is not carried

out, is acceptable to the analyst

Activity Sampling This is another commonly used

task analysis method for collecting information about

the type and the frequency of activities making up

a task Figure 5 shows the steps involved in activitysampling

Samples of the human operator's behavior at

speci-®ed intervals are collected to determine the proportion

of time the operator spends performing the identi®edactivities Two key factors for the activity samplingmethod to work include the requirements that thetask elements be observable and distinct from oneanother, and that the sampling keep pace with theperformance of the task Typically, the analyst per-forming activity sampling, classi®es the activitiesinvolved, develops a sampling schedule (these twoaspects form the core of the design of activity samp-ling), collects and records information about activities,and analyzes the collected activity samples Activitysampling has its advantages and disadvantages.Objectivity in data recording and collection, ease

of administering the technique, and the ability of thetechnique to reveal task-unrelated activities that needanalysis, are some of the advantages of the method.Requirements of a skilled analyst (for proper identi®-cation and description of the task elements), and theinability of the technique to provide for analysis ofcognitive activities are the main disadvantages of thetechnique

Task Decomposition This is a method used to exactlystate the tasks involved in terms of information con-

Table 3 Summary of Task Analysis Activities and Methods Involved inInterface Design

Gathering task information Hierarchical task analysisrepresenting the activities within the task Activity samplingStating required information, actions, Work study

Decision/action diagramsChecking adequacy of provisions for Table-top analysisinformation ¯ows for successful Simulationcompletion of the task Walk-through/talk-through

Operator modi®cations surveysCoding consistency surveysIdentifying links between attributes Link analysis

(total system check) to ensure system Petri nets

Simulator trialsProvide detailed design Person speci®cation

Modi®ed from Ref 6.

tent, and actions and feedback required of the

opera-tor Once a broad list of activities and the tasks

involved have been generated using either hierarchical

task analysis or activity sampling, task decomposition

can be used to systematically expand on the task

descriptions The various steps involved in task

decom-position are presented inFig 6

Decision±Action Diagram This is one of the most

commonly used tools for decision making Figure 7

is an example of a decision±action diagram [7] The

decision±action diagram sequentially proceeds through

a series of questions (representing decisions) and

pos-sible yes/no answers (representing actions that can be

taken) The questions are represented as diamonds,

and the possible alternatives are labeled on the exit

lines from the diamond A thorough knowledge of

the system components, and the possible outcomes of

making decisions about system components is essential

for constructing complete and representative decision±

action diagrams

Table-Top Analysis As the name implies, this is a

technique through which experts knowledgeable

about a system discuss speci®c system characteristics

In the context of interface design, this task analysis

methodology is used for checking if the information

¯ows identi®ed during the initial task analysis andtask description, is adequate for successful task com-pletion Table-top analysis, hence, typically follows theinitial hierarchical or other forms of task analysiswhich yield task descriptions, and provides informa-tion input for the decomposition of the tasks A num-ber of group discussion techniques exist in practice,including the Delphi method, the group consensusapproach, the nominal group technique, etc., for con-ducting table-top analysis, each with its own meritsand demerits

Walk-Through/Talk-Through Analysis These lyses involve operators and other individuals havingoperational experience with the system, walking andtalking the analyst through observable task com-ponents of a system in real time Walk-through isnormally achieved in a completely operational system

ana-or in a simulated setting ana-or even in a mock-up setting.Talk-through can be performed even without a simula-tion of the systemÐthe only requirements are drawingand other system speci®c documentation to enable theanalysts to set system and task boundaries while per-forming the talk-through analysis For more informa-tion on walk-through and talk-through analyses, refer

to Meister [8]

Figure 5 Activities involved in activity sampling

Operator Modi®cation Surveys These surveys are

performed to gather input from the actual users,

(i.e., the operators) of the system, to check if there

will be diculties in using the system, and of what

types This checking of the adequacy of the interface

design of the system from the users' perspective is

done through surveys conducted on similar already

operational systems In general, operators and other

users of systems maintain and provide information on

design inadequacies through memory aids, such as

their own labels on displays to mark safe limits,

per-ceptual cues, such as makeshift pointers, and

organi-zational cues, such as grouping instruments through

the use of lines These makeshift modi®cations done

by the operators indicate design de®ciencies in thesystem, and can be planned for and included in theredesign of the existing system or in the design of anew system

Coding Consistency Surveys These surveys are used

to determine if the coding schemes in use in thesystem are consistent with the associated meanings,and if and where additional coding is needed Therecommendation when performing coding consis-tency surveys is to record the description of the loca-tion of the item, a description of the coding used forthat item (intermittent siren sound), a description ofany other coding schemes used for that item (inter-Figure 6 The task decomposition process

mittent siren sound accompanied by a yellow

¯ash-ing light), and a complete description of the function

being coded

Link Analysis This is a technique used to identify

and represent the nature, frequency, and/or the

impor-tance of relationships or links existing between

indivi-dual operators and some portion of the system [9]

Link analysis has been found to be particularly useful

in applications where the physical layout of equipment,instruments, etc., is important to optimize the inter-action of the human operator with the system Linkanalysis does not require extensive resources toperform (in fact, paper and pencil are the onlyresources required to perform a link analysis) Linkanalysis proceeds by ®rst collecting informationabout the system components used during task perfor-mance This information is then used to develop acomplete list of links between individual system ele-ments The links thus established are then diagramedand ranked for importance The order of importancemay be determined based on the frequency of activitybetween two links, or based on other appropriate mea-sures decided by the system expert The nature of thelinks to be studied (is it a movement of attention orposition between parts of the system?), and the level ofdetail to include in de®ning each link are importantfactors that determine the overall structure and useful-ness of the links established Link analysis does notneed observational data collection; a mere description

of the procedures in the form of a technical manual issucient for identifying and establishing the links Theextensive graphical and tabular representationsinvolved in link analysis, however, limits the use ofthis technique for large systems with involved linkages

in the system

Simulator Analysis The goal of simulation studies is

to replicate, and observe, system (including operatorand operating environment) performance while mak-ing the performance environment as representative andclose to the real-time environment as possible.Di€erent forms of simulations exist depending on theplatform or the simulator used for the simulation: asimple paper-and-pencil simulation, to a mock-up of

a system that may or may not be dynamic, to adynamic simulation which will respond in real time.Whatever the method of simulation used, the key con-sideration in simulation studies is the trade-o€ betweenthe ®delity of simulation (deciding the features of thesystem that need ®delity is an issue too), and the cost ofinvolved in building high-®delity simulations Despitethis limitation, simulation analysis can be useful whendesigning task situations that are dangerous forhumans to perform, or dicult to observe

Person Speci®cation The goal of person speci®cation

is to detail the key physical and mental capabilities, thekey qualifcations and personality traits, and experi-ence, required of the operator to perform specif edtasks Person speci®cation is similar to the user char-Figure 7 Generic function allocation analysis ¯owchart

Figure 7 (continued)

Figure 7 (continued)

acterization exercise described in Sec 1.3.1.1; the

checklist used for user characterization can be used

for person speci®cation also One of the widely used

techniques for person speci®cation is the position

ana-lysis questionnaire Broadly, position anaana-lysis

ques-tionnaires require the operator to identify for their

speci®ed tasks andjobs, the information input, the

mental processes, the work output, the context of the

job, the relationship with other personnel in the

sys-tem, and any other relevant job characteristics Using

the responses from the operators, the skill content of

tasks and jobs can be determined, and can help in

designing personnel selection and training programs

to ensure optimal human±machine interaction

Ergonomics Checklists These checklists are generally

used to ascertain if a particular system meets

ergo-nomic principles and criteria Ergoergo-nomics checklists

can check for subjective or objective information and

can cover issues ranging from overall system design to

the design of individual equipment Checklists can also

range in detail from the broad ergonomic aspects to

the minute detail Table 4 provides an example of a

checklist for equipment operation A number of

other standard checklists have also been developed

by the ergonomics community Important among

these are the widely used and comprehensive set of

checklists for di€erent ergonomics issues by

Woodson [10,11], MIL-STD 1472C [12] which covers

equipment design (written primarily for military

equip-ment, but can be used as a guide to develop checklists),

EPRI NP-2360 [13] which is a checklist for

mainte-nance activities in any large-scale system,

NUREG-0700 [14] which is a comprehensive checklist for

con-trol room design, the HSE checklist [15] which deals

with industrial safety and human error, and the

numer-ous checklists for CRT displays and VDUs [16,17]

1.3.1.3 Characterization of the Situation

Apart from the user and the task variables that could

a€ect system performance, the external environment in

which the system functions can also in¯uence the

human±system interaction performance.Table 5

pro-vides a representative checklist for the most commonly

encountered situations for which the system analyst

must obtain answers, and attempt to provide for

these situations in design

1.3.2 Allocation of Functions

In designing the human±machine interface, once

com-prehensive information about the users and the

activ-ities/tasks these users will perform is known (throughthe use of tools presented in the earlier sections), thespeci®c activities and tasks need to be assigned either

to the humans or to the machines The allocation offunctions is a necessary ®rst step before any furtherdesign of the interface in the human±machine systemcan be carried out

The need for solving the function allocation blem directly stems from the need to decide the extent

pro-of automation pro-of manufacturing activities This is sobecause, in the present day manufacturing scenario,the decision to make is no longer whether or not toautomate functions in manufacturing, but to whatextent and how

The function allocation problem is perhaps as old asthe industrial revolution itself Fitts' list, conceived in

1951 (Table 6), was the ®rst major e€ort to resolve thefunction allocation problem

However, while Fitts' list provided fundamental andgeneric principles that researchers still follow forstudying function allocation problems, its failure toprovide quantitative criteria for function allocationresulted in its having little impact on engineeringdesign practices The development of practical andquantitative criteria for allocating functions is com-pounded by an important issue: unless one candescribe functions in engineering terms, it is impossible

to ascertain if a machine can perform the function;and, if one can describe human behavior in engineeringterms, it may be possible to design a machine to do thejob better (than the human) But many functions can-not be completely speci®ed in engineering (numerical)terms This implies that those functions that cannot bespeci®ed in engineering terms should be allocated tohumans, with the rest allocated to the machines Inaddition, for the practitioner, function allocation con-siderations have been limited due to the lack of [19]:

1 Systematic and step-by-step approaches to sion making during function allocation

deci-2 Systematic and concise data for addressingissues such as the capability and limitations ofhumans and automated equipment, and underwhat circumstances one option is preferableover the other

3 Methodology for symbiotic agents such as ufacturing engineers and ergonomists, to inte-grate human and machine behaviors

man-4 Uni®ed theory addressing domain issues such asroles, authorities, etc

5 Integration of other decision-making criteria(such as the economics of the situation) so

Table 4 Example of an Ergonomics Checklist for Equipment Operation

Compromise butCharacteristic Satisfactory acceptable UnsatisfactoryConsole shape/size

Desk height, areaControl reachDisplay viewBody, limbclearancePanel locationFrequency of useSequence of useEmergency responseMultioperator usePanel layoutFunctional groupingSequential

organizationIdenti®cationClearance spacingDisplays

Functionalcompatibility forintended purposesIntelligibility ofinformation contentControl interactionLegibility; ®gures,pointers, scalesVisibility;

illumination,parallaxLocationIdenti®cationControlsFunctionalcompatibility forintended purposesLocation, motionexcursion, and forceDisplay interactionSpacing, clearance,size

Identi®cationAdapted from Ref 10.

that the function allocation decision is not made

in isolation

6 Easily usable tools to simulate di€erent

con®g-urations of humans and machines

In spite of these shortcomings, research on function

allocation has permitted the following general

infer-ences for the practitioner:

1 Function allocation cannot be accomplished

by a formulaÐor example, rules which may

apply in one situation may be irrelevant inanother

2 Function allocation is not a one-shot sionÐthe ®nal assignment depends on activ-ities at the levels of the tasks, the con¯ation

deci-of tasks into jobs, the relationships deci-of jobswithin a larger workgroup, and the likelychanges in the higher level manufacturing pro-cesses themselves

3 Function allocation can be systematizedÐit isclear that there are a number of sequentialsteps that can be taken to best allocate func-tions

4 Both humans and machines can be good orbad at certain tasks

5 Using analogies can facilitate the functionallocation process

6 Function allocation can be targeted to a ci®c time frame

spe-7 Function allocation depends on the nature ofthe taskÐit varies based on whether the task isperceptual, cognitive, or psychomotor

8 Function allocation decisions must be based

on sound economic analyses of options aswell as the capabilities and limitations ofhumans and machines

9 Human and machine performances are notalways antithetical

10 Function allocation decisions must considertechnology advances within a given timeframe

11 In cases where both humans and machines canperform a function, the system should bedesigned in such a way so that humans candelegate the function to machines, or can

Table 5 Checklist for Situation Analysis

What are the likely situations thatcould arise during system use and howwill these a€ect use of the system?

Equipment Falls short of target performance

Falls short of speci®cationFails

Availability Data is missing

Materials are missingPersonnel are missingSupport is missingOverloads Of people/machines

Of data, information, materials, etc

Interruptions The process breaks down

Complete restart of process requiredEnvironment Changes: in physical or social

environmentPolicy changes Changes in laws, rules, standards and

guidelinesAdapted from Ref 5.

Table 6 Fitts' List

Humans appear to surpass present-day machines with respect to the following:

Ability to detect small amounts of visual or acoustic energy

Ability to perceive patterns of light or sound

Ability to improvise and use ¯exible procedures

Ability to store very large amounts of information for long periods and to recall relevant facts at the appropriate timeAbility to reason inductively

Ability to exercise judgment

Present-day machines appear to surpass humans with respect to the following:

Ability to respond quickly to control signals, and to apply great force smoothly and precisely

Ability to perform repetitive, routine tasks

Ability to store information brie¯y and then to erase it completely

Ability to reason inductively, including computational ability

Ability to handle complex operations, i.e., to do many di€erent things at once

Adapted from Ref 18.

take over the function when circumstances

demand it

A number of approaches have been suggested in the

literature for solving the function allocation problem

Some of the promising approaches include function

allocation criteria based on speci®c performance

mea-sures (time required to complete tasks, for example)

[20±24], criteria based on comparison of capabilities

and limitations of humans with particular attention

given to knowledge, skills, and information sources

and channels [25±34] criteria based on economics

(allo-cate the function to the less expensive option),

[21,35,36], and criteria based on safety (to the human

operator in the system) [37±39]

Experiments with these approaches suggest that

functions that are well-proceduralized permitting

algo-rithmic analysis, and requiring little creative input, are

prime candidates for automation On the other hand,

functions requiring cognitive skills of a higher order,

such as design, planning, monitoring, exception

hand-ling, etc., are functions that are better performed by

humans The prime requirements for automating any

function are the availability of a model of the activities

necessary for that function, the ability to quantify that

model, and a clear understanding of the associated

control and information requirements Clearly, there

are some functions that should be performed by

machines because of:

1 Design accuracy and tolerance requirements

2 The nature of the activity is such that it cannot

be performed by humans

3 Speed and high production volume

require-ments

4 Size, force, weight, and volume requirement

5 Hazardous nature of the activity

Equally clearly, there are some activities that should be

performed by humans because of:

1 Information-acquisition and decision-making

needs

2 Higher level skill needs such as programming

3 Specialized manipulation, dexterity, and sensing

needs

4 Space limitations (e.g., work that must be done

in narrow and con®ned spaces)

5 Situations involving poor equipment reliability

or where equipment failure could prove

catastrophic

6 Activities for which technology is lacking

Mital et al [7] provide a generic methodology in theform of decision-making ¯owcharts for the systematicallocation of functions between humans and machines

Figure 7, presented earlier is a part of these ¯owcharts.These ¯owcharts are based on the requirements ofcomplex decision making, on a detailed safety analysis,and on a comprehensive economic analysis of the alter-natives These function allocation ¯owcharts are avail-able for di€erent manufacturing functions such asassembly, inspection, packaging, shipping, etc., andshould be consulted for a detailed analysis of the ques-tion of manufacturing function allocation

1.3.3 Information Presentation and Control1.3.3.1 The Scienti®c Basis for Information

Input and ProcessingReduced to a fundamental level, human interactionwith automation can be said to be dependent uponthe information processing ability of the human, andupon the exchange of information among the di€erentelements in a system Over the years, behavioral scien-tists have attempted to explain human informationprocessing through various conceptual models andtheories One such theory is the information theory[40] Information, according to information theory, isde®ned as the reduction of uncertainty Implicit inthis de®nition is the tenet that events that are highlycertain to occur provide little information; events thatare highly unlikely to occur, on the other hand, pro-vide more information Rather than emphasize theimportance of a message in de®ning information,information theory considers the probability of occur-rence of a certain event in determining if there is infor-mation worth considering For instance, the ``no-smoking'' sign that appears in airplanes before takeo€,while being an important message, does not conveymuch information due to the high likelihood of itsappearance every time an aircraft takes o€ On theother hand, according to information theory, messagesfrom the crew about emergency landing procedureswhen the plane is about to perform an emergency land-ing convey more information due to the small like-lihood of such an event Information is measured inbits (denoted by H) One bit is de®ned as the amount ofinformation required to decide between two equallylikely alternatives

When the di€erent alternatives all have the sameprobability, the amount of information (H) is given by

H ˆ log2N

where N is the number of alternatives For example,

when an event only has two alternatives associated

with it, and when the two alternatives are equally

likely, by the above equation, the amount of

informa-tion, in bits, is 1.0

When the alternatives are not equally likely (i.e., the

alternatives have di€erent probabilities of occurrence),

the information conveyed by an event is given by

hiˆ log2…1=pi†

where hiis the information associated with event i, and

pi is the probability of occurrence of event i

The average information (Hav† conveyed by a series

of events having di€erent probabilities is given by

HavˆXpi…log2…1=pi††

where pi is the probability of the event i

Just as a bit is the amount of information,

redun-dancy is the amount of reduction in information from

the maximum due to the unequal probabilities of

occurrence of events Redundancy is expressed as a

percentage, and is given by

% Redundancy ˆ …1 Hav=Hmax†  100

Information theory, while providing insight into

measuring information, has major limitations when

applied to human beings It is valid only for simple

situations which can split into units of information

and coded signals [41] It does not fully explain the

stimulus-carrying information in situations where

there are more than two alternatives, with di€erent

probabilities

The channel capacity theory, another theory

explain-ing information uptake by humans, is based on the

premise that human sense organs deliver a certain

quantity of information to the input end of a channel,

and that the output from the channel depends upon the

capacity of the channel It has been determined that if

the input is small, there is very little absorption of it by

the channel, but that if the input rises, it reaches the

threshold channel capacity, beyond which the output

from the channel is no longer a linear function of the

input [41] Experimental investigations have shown

that humans have a large channel capacity for

infor-mation conveyed to them through the spoken word

than through any other medium A vocabulary of

2500 words requires a channel capacity of 34 to 42

bits per second [42] Designers must keep in mind

that in this day and age of information technology,

the central nervous system of humans is subjected to

more information than the information channel can

handle, and that a considerable reduction in theamount of information must be carried out beforehumans process the information

In addition to theories such as the informationtheory and the channel capacity theory that explaininformation uptake, many conceptual models ofhuman information processing have been proposed

by researchers over the last four decades Figure 8

shows one such fundamental model (most othermodels contain elements of this basic model) depictingthe stages involved in information processing [43] Thekey elements of the model are perception, memory,decision making, attention, response execution, andfeedback The following is a brief discussion of each

of these elements

Perception may involve detection (determiningwhether or not a signal is present), or identi®cationand detection (involving detection and classi®cation).The theory of signal detection [43±45] through the con-cept of noise in signals, attempts to explain the process

of perception and response to the perceived signals.Four possible outcomes are recognized in signal detec-tion theory: (1) hit (correctly concluding that there is asignal when there is one), (2) false alarm (concludingthat there is a signal when, in actuality, there is none),(3) miss (concluding that there is no signal when, inactuality, there is one and (4) correction rejection (cor-rectly concluding that there is no signal when there isnone) The fundamental postulate of signal detectiontheory is that humans tend to make decisions based oncriteria whose probabilities depend upon the probabil-ities of the outcomes above The probability of observ-ing a signal, and the costs and bene®ts associated withthe four possible outcomes above, determine theresponses of the human to the signal The resolution

of the human sensory activities (ability to separate thenoise distribution from the distribution of the signal)has also been found to a€ect the signal detection cap-ability of the human

Memory, in humans, has been conceptualized asconsisting of three processes, namely, sensory storage,working memory, and long-term memory [43].According to this conception, information from sen-sory storage must pass through working memorybefore it can be stored in long-term memory Sensorystorage refers to the short-term memory of the stimu-lus Two types of short-term memory storage are wellknownÐiconic storage associated with visual senses,and echoic storage associated with the auditory senses[46] Sensory storage or short-term memory has beenshown to be nearly automatic requiring no sustainedattention on the part of the human to retain it

Information transfer from sensory storage to working

memory is brought about through attention (to the

process) Information from stimuli is believed to be

stored in the working memory primarily in the form

of either visual, phonetic, or semantic codes It is also

believed that the capacity of working memory is ®ve to

nine chunks of information (similar units regardless of

the size) [47] Researchers recommend presenting ®ve

to nine meaningful and distinct chunks of information

for improved recall It has also been determined that

there is a linear relationship between the number of

items in a memorized list and the time required to

search the list of items in the working memory [48]

Also, all items in the working memory are searched

one at a time, even if a match is found early in the

search process The transfer of information from

work-ing memory to the long-term memory is believed to

take place through semantic coding, i.e., by analyzing,

comparing, and relating information in the working

memory to past stores of knowledge in the long-term

memory [46] The extent to which information can be

retrieved from long-term memory depends on the

extent of organization of the information in the

long-term memory

Rational decision making is de®ned as the process

that involves seeking information relevant to the

decision at hand, estimating the probabilities of

various alternatives, and attaching values to theanticipated alternatives A number of biases, however,have been identi®ed to exist among humans that oftenmakes decision making irrational.Table 7lists some ofthese biases

Attention is another key factor in¯uencing humaninformation input and processing Research hasidenti®ed four types of tasks or situations requiringattention These are selective attention, focusedattention, divided attention, and sustained attention.When several information sources are to be monitored

to perform a single task, attention is said to be selective(e.g., a process control operator scanning severalinstrument panels before detecting a deviant value)

Table 8 provides guidelines for improving ances in tasks requiring selective attention When ahuman has to focus attention on one source of infor-mation and exclude all other sources of informationfor task performance, attention is said to be focused.Task performance under focused attention is a€ected

perform-by the physical proximity of the sources of tion While physical proximity enhances performance

informa-in tasks requirinforma-ing selective attention, it impedesperformance in tasks requiring focused attention

Table 9 provides guidelines for improving ances in tasks requiring focused attention Whenhumans do more than one task at a time, their atten-Figure 8 Fundamental model of human information processing

perform-tion is said to be divided (among the tasks) While

much of the theoretical base for explaining

perform-ance of tasks requiring divided attention is still

evol-ving [43,49], some guidelines for designing tasks that

require divided attention are available, and are

pro-vided in Table 10 When humans maintain attention

and remain vigilant to external stimuli over prolonged

periods of time, attention is said to be sustained

Nearly four decades of research in vigilance and

vigi-lance decrement [50±53] has provided guidelines for

improving performance in tasks requiring sustained

attention (Table 11)

In addition to the factors discussed above,

consider-able attention is being paid to the concept of mental

workload (which is but an extension of divided

atten-tion) Reviews of mental workload measurement

tech-niques are available [54±56], and should be consulted

for discussions of the methodologies involved in

men-tal workload assessment

1.3.3.2 The Scienti®c Basis for Human Control

of SystemsHumans respond to information and take controllingactions The controlling actions of the human aremediated through the motor system in the humanbody The human skeletal system, the muscles, andthe nervous system help bring into play motor skillsthat enable the human to respond to stimuli Motorskill is defned as ``ability to use the correct muscleswith the exact force necessary to perform the desiredresponse with proper sequence and timing'' [57] Inaddition, skilled performance requires adjusting tochanging environmental conditions, and acting con-sistently from situation to situation [58] A number

of di€erent types of human movements have beenrecognized in the literature [46] These include discretemovements (involving a single reaching movement to atarget that is stationary), repetitive movements (asingle movement is repeated), sequential movements

Table 7 Common Human Biases

Humans attach more importance to early information than subsequent information

Humans generally do not optimally extract information from sources

Humans do not optimally assess subjective odds of alternative scenarios

Humans have a tendency to become more con®dent in their decisions with more information, but do not necessarily becomemore accurate

Humans tend to seek more information than they can absorb

Humans generally treat all information as equally reliable

Humans seem to have a limited ability to evaluate a maximum of more than three or four hypotheses at a time

Humans tend to focus only on a few critical factors at a time and consider only a few possible choices related to these criticalfactors

Humans tend to seek information that con®rms their choice of action than information that contradicts or discon®rms theiraction

Human view a potential loss more seriously than a potential gain

Humans tend to believe that mildly positive outcomes are more likely than mildly negative or highly positive outcomes.Humans tend to believe that highly negative outcomes are less likely than mildly negative outcomes

Adapted from Ref 43.

Table 8 Recommendations for Designing Tasks Requiring Selective Attention

Use as few signal channels as possible, even if it means increasing the signal rate per channel

Inform the human the relative importance of various channels for e€ective direction of attention

Reduce stress levels on human so more channels can be monitored

Inform the human beforehand where signals will occur in future

Train the human to develop optimal scan patterns

Reduce scanning requirements on the human by putting multiple visual information sources close to each other, and by makingsure that multiple sources of auditory information do not mask each other

Provide signal for a sucient length of time for individual to respond; where possible, provide for human control of signal rate.Adapted from Ref 46.

(a number of discrete movements to stationary

targets), continuous movements (involving muscular

control adjustments during movement), and static

positioning (maintaining a speci®c position of a body

member for a speci®ed period of time) In addition,

certain theoretical models of human motor responses

explain the control aspects of human responses based

on only two fundamental types of movementsÐfast

and slow Closed-loop theories [59,60], whether the

movement be fast or slow, use the concept of sensory

feedback (sensory information available during or

after the motor response) to explain motor responses

(to correct/reduce errors obtained through feedback)

The sensory receptors for feedback and feedforward

(sensory information available prior to the action

that regulates and triggers responses), are believed to

be located in the muscle spindles (for sensing the

muscle length and the rate of change of length)

[58,61], tendons (the Golgi tendons inhibit muscle

contraction and regulate muscle action), joints (the

tension in the joints in¯uences the generation of

nerve impulses), cutaneous tissue (skin is believed to

have receptors that a€ect joint movement), and the

eyes (important for timing of responses) [62]

Open-loop theories, on the other hand, are based on the

belief that there are higher-level structured motor

programs containing information necessary for

patterning the di€erent movements [63,64] Di€erentde®ciencies, such as the error of selection (where aperson calls the wrong motor program for a control-ling action) and the error of execution (where thecorrect motor program fails during execution ofcontrolling actions) have been identi®ed with motorprograms [65] Much of the development in under-standing human controlling actions in response tostimuli is still in its infancy, but has important practicalconsequences (how to improve skilled performance,for example)

The time it takes for the human to respond to stimuli

is another critical factor that has been studied sively in the literature [46] An understanding ofresponse time of the human is essential for good design

exten-of the tasks involved in human interaction with mated systems Response time is, in general, composed

auto-of reaction time, and movement time Reaction time isde®ned as the time from the signal onset calling for aresponse, to the beginning of the response Simple reac-tion time (reaction time in the presence of a singlesource of stimulus) has been shown to be between0.15 sec and 0.20 sec The mode through which thesingle stimulus occurs (visual, auditory etc.,) thedetectability of the stimulus (intensity, duration, andsize), the frequency, the preparedness (of the humanfor the stimulus), the age, and the location of thestimulus (location in the peripheral ®eld of view, forinstance) are among the factors that have been shown

to a€ect simple reaction time Choice reaction time(reaction time in the presence of one of several possiblestimuli each with di€erent possible responses), is afunction of the probability of a stimulus occurring,i.e., the reaction time is faster for events with greaterprobability It has been shown to increase by about0.15 sec for each doubling of the number of possible

Table 9 Recommendations for Designing Tasks Requiring

Focused Attention

Make the di€erent channels of information as distinct as

possible from the channel to which the human must

attend

Physically separate the channel of interest from the other

channels

Reduce the number of competing channels

Make the channel of interest prominent by making it larger

in size, or brighter, or louder, or by locating it centrally

Adapted from Ref 46.

Table 10 Recommendations for Designing Tasks Requiring

Divided Attention

Minimize the potential sources of information

Provide human with a relative priority of tasks to optimize

the strategy of divided attention

Keep the level of diculty of tasks low

Make tasks as dissimilar as possible in terms of task demands

on the human

Adapted from Ref 46.

Table 11 Recommendations for Designing Tasks RequiringSustained Attention

Provide appropriate work±rest schedules

Provide task variation by interpolating di€erent activities.Make the signal larger, and/or more intense, and/or longer induration, and/or distinct

Reduce uncertainty in time and place of occurrence of signal.Use arti®cial signals and provide feedback to humans ontheir performance

Reduce the rate of presentation of stimuli if it is high.Provide optimal environmental conditions such as lighting,noise level, etc

Provide adequate training to humans to clarify the nature ofsignals to be identi®ed

alternative stimuli [66] Choice reaction time has been

shown to be in¯uenced by a numerous factors,

includ-ing the degree of compatibility between stimuli and

responses, practice, presence or absence of a warning

signal, the type and complexity of the movement

involved in the responses, and whether or not more

than one stimulus is present in the signal Movement

time is defned as the time from the beginning of the

response to its completion It is the time required to

physically make the response to the stimulus

Movements based on pivoting about the elbow have

been shown to take less time, and have more accuracy,

than movements based on upper-arm and shoulder

action Also, it has been determined that movement

time is a logarithmic function of distance of movement,

when target size is a constant, and further that

move-ment time is a logarithmic function of target size, when

the distance of movement is constant This ®nding is

popularly known as Fitts' law [67], and is represented

as

MT ˆ a ‡ b log2…2D=W†

where MT is the movement time, a and b are empirical

constants dependent upon the type of movement, D is

the distance of movement from start to the center of

the target, and W is the width of the target

Human response to stimuli is not only dependent

upon the speed of the response, but also on the

accuracy of the response The accuracy of the human

response assumes special importance when the

response has to be made in situations where there is

no visual feedback (a situation referred to as ``blind

positioning'') Movements that take place in a blind

positioning situation have been determined to be

more accurate when the target is located dead-ahead

than when located on the sides Also, targets below the

shoulder height and the waist level are more readily

reachable than targets located above the shoulder or

the head [68] The distance and speed of movement

have also been found to in¯uence the accuracy of the

response [69,70]

1.3.3.3 Displays

Types of Displays A display is de®ned as any indirect

means of presenting information Displays are

gener-ally one of the following four types: visual, auditory,

tactual, and olfactory The visual and the auditory

modes of displaying information are the most

common Displays based on tactile and olfactory

senses are mostly used for special task or user

situations (e.g., for the hearing impaired)

Selecting the mode of display whether it should bevisual or auditory in nature) is an important factor due

to the relative advantages and disadvantages certainmodes of display may have over other modes, for spe-ci®c types of task situations (auditory mode is betterthan visual displays in vigilance), environment (light-ing conditions), or user characteristics (person's infor-mation handling capacity) Table 12 provides generalguidelines for deciding between two common modes ofinformation presentation, namely, auditory and visual.The types of displays to use to present informationalso depend on the type of information to present.Di€erent types of information can be presented usingdisplays when the sensing mode is indirect.Information can either be dynamic or static.Dynamic information is categorized by changesoccurnng in time (e.g., fuel gage) Static information,

Table 12 Guidelines for Deciding When to Use VisualDisplays and When to Use Auditory Displays

Visual AuditoryCharacteristics displays displaysMessage characteristics

Immediacy of actionrequirement of message

events in timeMessage deals with plocations in space

too dark requiringsigni®cant adaptationLocation too noisy pAdapted for Ref 71.

on the other hand, does not change with time (e.g.,

printed safety signs) A number of other types of

infor-mation are also recognized in the literature Table 13

provides a list of these types along with a brief

descrip-tion of the characteristics of these types of informadescrip-tion

In the following sections, we discuss

recommenda-tions for the design of di€erent types of visual and

auditory displays (we restrict our attention in this

chapter only to these two common modes) We ®rst

provide a brief discussion of the di€erent factors

a€ect-ing human visual and auditory capabilities We then

present speci®c display design issues and

recommenda-tions for these two broad types of displays

Visual displays: factors affecting design

Accommo-dation refers to the ability of the lens in the eye to focus

the light rays on the retina The distance (of the target

object from the eye) at which the image of the object

becomes blurred, and the eye is not able to focus the

image any further, is called the near point There is

also a far point (in®nity, in normal vision) beyond

which the eye cannot clearly focus Focal distances

are measured in diopters One diopter is 1/(distance

of the target in meters) Inadequate accommodation

capacity of the eyes result either in nearsightedness

(the far point is too close) or in farsightedness (the

near point is too close) Literature recommends an

average focusing distance of 800 mm at the resting

position of the eye (also known as the resting

accom-modation) [72] Due to changes in the iris (which

con-trols the shape of the lens), aging results in substantial

receding of the near point, the far point remaining

unchanged or becoming shorter Figure 9 shows how

the mean near point recedes with age It is

recom-mended that the designer use this information whendesigning visual displays

Visual acuity is de®ned as the ability of the eye toseparate ®ne detail The minimum separable acuityrefers to the smallest feature that the eye can detect.Visual acuity is measured by the reciprocal of thevisual angle subtended at the eye by the smallest detailthat the eye can distinguish Visual angle (for anglesless than 108) is given by

Visual angle …in minutes† ˆ …3438H†=Dwhere H is the height of the stimulus detail, and D isthe distance from the eye, both H and D measured inthe same units of distance Besides minimum separablevisual acuity, there are other types of visual acuitymeasure, such as vernier acuity (ability to di€erentiatelateral displacements), minimum perceptible acuity(ability to detect a spot from its background), andstereoscopic acuity (ability to di€erentiate depth in asingle object) In general, an individual is considered tohave normal visual acuity if he or she is able to resolve

a separation between two signs 10 of arc wide Visualacuity has been found to increase with increasing levels

of illumination Luckiesh and Moss [73] showed thatincreasing the illumination level from approximately

10 l to 100 l increased the visual acuity from 100

to 130%, and increasing the illumination level fromapproximately 10 l to 1000 l increased the visualacuity from 100 to 170% For provision of maximumvisual acuity, it is recommended that the illuminationlevel in the work area be 1000 l Providing adequatecontrast between the object being viewed and theimmediate background, and making the signs and

Table 13 Commonly Found Types of Information and Their Characteristics

Quantitative information Information on the quantitative value of a variable

Qualitative information Information on the approximate value, trend, rate of change, direction of change, or other

similar aspects of a changeable variableStatus information Information on the status of a system, information on a one of a limited number of conditions,

and information on independent conditions of some classWarning and signal Information on emergency or unsafe conditions, information on presence or absence of some

Representational Pictorial or graphic representations of objects, areas, or other con®gurations

information

Identi®cation information Information in coded form to identify static condition, situation, or object

Alphanumeric and Information of verbal, numerical, and related coded information in other forms such asSymbolic information signs, labels, placards, instructions, etc

Time-phased information Information about pulsed or time-phased signals

Adapted from Ref 46.

characters (in the object being viewed) sharp, will also

increase visual acuity The general recommendation is

to use dark symbols and characters on a bright

back-ground than vice versa, as the former increases the

visual acuity Visual acuity has also been shown to

decrease with age [74] Figure 10 illustrates how visual

acuity decreases with age

Contrast sensitivity is another factor that has

impli-cations for design of the interface It is the ability of the

eye to di€erentiate lightness between black and white

Contrast sensitivity is generally expressed as the

reciprocal of the threshold contrast, where the

threshold contrast is the level of contrast that just

stops short of making the colors appear homogeneous

Other measures for contrast sensitivity include

modulation contrast computed as

C ˆ …Lmax Lmin†=…Lmax‡ Lmin†

where Lmax and Lmin are the maximum and the

mini-mum luminances in the pattern The literature provides

certain general rules to follow when designing displays

in order to provide the best possible contrast

sensitivity Since contrast sensitivity is greater forlarger areas, it is recommended that the viewing area

be made as large as possible Also, making the objectboundaries sharper will increase contrast sensitivity.The surrounding luminance, and the intensity of light(or the level of illumination), have been shown to have

an e€ect on contrast sensitivity Contrast sensitivityFigure 9 Effect of age on near point for visual accomodation

Figure 10 Effect of age on visual acuity

has been determined to be the largest when

the surrounding luminance is within the range of

70 cd/m2, and more than 1000 cd/m2 [75] Also,

Luckiesh and Moss [73] showed that increasing the

illumination level from approximately 10 l to

100 1 increased the contrast sensitivity from 100 to

280%, and increasing the illumination level from

approximately 10 l to 1000 l increased the contrast

sensitivity from 100 to 450% The literature [41] also

recommends that the background be at least 2%

brighter or darker than the target for optimal contrast

sensitivity As brie¯y described above, visual acuity

and contrast sensitivity are a€ected by a number of

factor, such as luminance level (in general, the higher

the luminance, the more the visual acuity and contrast

sensitivity), contrast, exposure time, motion of the

target, age (there is a decline in both visual acuity

and contrast sensitivity with age), and training

(through surgery of the eye or through corrective

lenses, etc.)

Adaptation is another factor that a€ects the visual

capability of the human eye It is de®ned as the

changes in the sensitivity of the eye to light A

mea-sure of adaptation is the time it takes for the eye to

adapt to light or dark It has been found that, in

general, adaptation to light occurs more quickly than

adaptation to the dark Darkness adaptation has

been found to be quick in the ®rst 5 min of

expo-sure; nearly 80% of the adaptation to darkness has

been shown to take about 25 min with full

adapta-tion taking as much as one full hour [41]

Adaptation can also be partial (depending on

whether the visual ®eld contains a dark or a bright

area), and can a€ect the sensitivity of the retina and

the vision For optimal adaptation, the overall

recommendation is to provide the same order of

brightness on all important surfaces, and provide a

stable and non¯uctuating levels of illumination It is

also important to avoid the e€ects of glare (which is

a process of overloading the adaptation processes of

the eye) This can be achieved by avoiding excessive

brightness contrasts, avoiding excessive brightness in

the light source, and providing for transient

adaptation

The ability of the eye to discriminate between

dif-ferent colors is called color discrimination Color

dis-crimination de®ciency is due to the reduced sensitivity

of the particular (to a color) cone receptors While it is

dicult to measure precisely the type and degree of a

person's color de®ciency, it is important from the

perspective of designing tasks which require perception

of colored targets for task performance

The ability to read, and the ability to perceive ing, are the other key factors that have to be accountedfor when designing visual displays

mean-Design recommendations for visual displays Asalready mentioned, visual displays are classi®ed onthe basis of the type of information they present tothe user Information presented to the user can bestatic or dynamic in nature Display of dynamic infor-mation will require capture of the changing nature ofthe information (for example, continuous changes inspeed indicated by the tachometer in the car) Staticdisplays do not display, in real time, the changes in theinformation content in time (Note that, in static dis-plays, the displays themselves do not change with time.However, static displays can be used to present, in theform of graphs, for example, changes in informationcontent over time, after the event has occurred; staticdisplays do not provide information in real time.)Almost all dynamic visual displays contain elements

of one of the more fundamental forms of static mation displays, namely, textual information, informa-tion in the form of graphical displays, information insome coded form, or symbolic information In the fol-lowing sections, we ®rst brie¯y present recommenda-tions on design of these four forms of static visualdisplays We then provide guidelines on designingdynamic information displays

infor-Static visual displays The literature distinguishesbetween two forms of textual displaysÐtextualdisplays in hardcopy format, and textual displays invisual display terminals or computer screens [46].While there are differences in performance based onwhether the display is in hardcopy form or in a visualdisplay unit, there are three essential characteristics ofany display in the form of text; the textual displayshould be visible, legible, and readable Visibility ofthe text refers to the characteristic that makes acharacter or a symbol distinguishable and separatefrom its surroundings Legibility of the text refers tothe characteristic of alphanumeric characters thatmakes it possible to identify one character from theother The stroke width, the character format, con-trast, illumination etc., in¯uence the legibility of thetext Readability of the text refers to the characteristic

of alphanumeric characters that enables organization

of the content into meaningful groups (of information)such as words and sentences

Various factors in¯uence the visibility, the legibility,and the readability of textual information presented inhardcopy form They are typography, size, case,layout, and reading ease

The typography has been found to be especially

important when the viewing conditions are

unfavor-able, or when the information is critical (such as a

sign warning of danger) Typography depends on

factors such as the stroke width of the alphanumeric

character (ratio of thickness of the stroke to height of

the character), the width-to-height ratio of the

character, and the type style

Table 14 also provides accepted guidelines, based on

research, for size of characters, case, layout of

characters, and for reading ease of alphanumeric

characters Some examples of type style and other

aspects in typography of text are given inFig 11

Numerical text can also be represented in graphicalforms Graphs can be in di€erent forms such as linegraphs, bar and column graphs, pie charts, etc.Pictorial information such as in the form of graphsimproves the speed of reading, but the general recom-mendation in the literature is to combine pictorialinformation with information in the form of plaintext, improve the accuracy of the information pre-sented [76,77]

The visibility, readability, and legibility of display-terminal-based text has been found to dependupon the typography, the reading distance, the size ofcharacters, and hardware considerations, such as the

visual-Table 14 Recommendations for Design of Hardcopy Text

Typography

Stroke width When the illumination is reasonable, use 1 : 6 to 1 : 8 for black on white and 1 : 8 to 1 : 10 for

white on blackWhen the illumination is reduced, use thick letters than thin letters for greater readabilityWhen illumination is low or with low background contrast, use boldface characters with a lowstroke width±height ratio

When letters are highly luminous, use 1 : 12 to 1 : 20 ratioWhen letters are black on a highly luminous background, use thick strokesWidth±height ratio Use a 3 : 5 ratio for most practical applications; for transluminated or engraved legends, use 1:1Size of character

For close reading When the reading distance is 12 to 16 in.:

Use 0.09±0.11 in or 22 to 270 of visual angle for normal use of alphanumeric charactersWhen the viewing distance is 28 in.:

For critical use under 0.03 fL luminance, and variable position of character, use 020±0.30 in.height

For critical use over 1.0 fL luminance, and variable position of character, use 0.12±0.20 in.height

For critical use under 0.03 fL luminance, and ®xed position of character, use 0.15 to 0.30 in.height

For critical use over 1.0 fL luminance, and ®xed position of character, use 0.10±0.20 in.height

For noncritical use 0.05±0.20 in heightFor distant reading Use Wsˆ 1:45  10 5 S  d, and HLˆ Ws=R, where Wsis the stroke width, S is the

denominator of the Snellen acuity score (20/20, 20/40 etc.), d is the reading distance, HListhe height of the letter, and R is the stroke width-to-height ratio of the font

Case In general, use lowercase letters than uppercase letters for better readability

Use initial uppercase for search tasksLayout

Interletter spacing Provide close-set type than regular-set type where possible for easier readability

Interline spacing Increase spacing between lines for better clarity

Reading ease

Type of sentence Use simple, armative, active sentences where possible

Order of words Match order of words in sentence to the order of actions to be taken

Adapted from Ref 46.

polarity of the screen, and screen color It is generally

recommended that the size of the dot matrix for

alpha-numeric characters used in visual display terminals be

at least 7  9 for continuous reading of the text The

ANSI recommendation for reading distance is 18±

20 in This distance denotes the distance from the eye

to the screen, and is based on the assumption that the

user is seated in an upright position The ANSI

speci-®cation for the minimum character height for capital

letters is 160 of visual angle for reading tasks where

legibility is important The maximum character height

according to ANSI should be 240of visual angle, with

the preferred character height set at 20±220 of visual

angle As regards polarity of the screen, since the

sen-sitivity to ¯icker is greater when the screen background

is brighter, the literature recommends that display

units with light backgrounds have a higher refresh

rate than display units with dark backgrounds

Information, in the form of stimuli, can be sensedeither through direct observation of the object, orthrough the use of a indirect mediating device.During indirect sensing, the stimuli themselves mostlycome in a coded form (such as a visual or an auditorydisplay), and sometimes in the form of exact ormodi®ed (in size) reproductions of the originalstimulus (such as a picture on a television screen).Coding of information can be along di€erent stimulusdimensions; for example, coding can be done based oncolor, shape, size, etc Research has shown that thesuccess of coding in conveying the necessary informa-tion depends on people's ability to distinguish betweentwo or more stimuli which vary along a dimension(e.g., which of the two stimuli is smaller in size), and

on the ability to identify a single stimulus based on themeasure of that stimulus on the dimension scale (e.g.,whether the target is bright or dim) [46] These abilities,Figure 11 Examples of different type styles and type size

respectively, are said to be dependent on the relative

(comparing more than one stimulus) and the absolute

judgments (identi®cation of stimulus without the

opportunity to compare) of people It has also been

shown that humans, in general, have the ability to

make better relative judgments than absolute

judg-ments [47,78] This being the case, the orthogonality

or independence of the coding schemes determines how

unique the information provided by a code is, and

results in an increase in the number of stimuli that

can be identifed on an absolute basis [46]; for example,

if size (large and small) and color (black and white)

were orthogonal dimensions of coding, then each of

the possible codes namely, large-black, large-white,

small-black, and small-white, would provide unique

information A number of di€erent guidelines also

exist in the literature that can help good coding

prac-tices Table 15 summarizes general guidelines for

designing a good visual coding system In addition,

di€erent visual coding methods have their own speci®c

design features that can be exploited by the designer

for speci®c tasks and work situations Using

alpha-numeric characters (which are 0, 1, 2, 3; ; 9

and a; b; c; ; z in the English language), singly and

in combination, for instance, has been found to be

useful for identi®cation purposes, and for situations

with space constraints Color coding of surfaces (24

or more combinations of hues, saturation, and

brightness are possible, though research recommends

use of no more than nine combinations) are useful

for industrial tasks requiring searching and counting

Color-coding surfaces can, however, be ine€ective if

the worker population is color de®cient [79,80] Color

coding any lights used in the workplace has been

shown to be e€ective for qualitative reading [81]

The recommendation is to limit the number of lights

coded to three Coding using geomerical shapes

(there are a total of 15 or more geometrical shapes),has been found to be useful in situations using sym-bolic representation of an action or an event Theliterature recommends the use of no more than ®vegeometrical shapes, as using more than ®ve will lead

to diculty in discrimination of the di€erent shapes[81] While a total of 24 di€erent angles of inclination(of characters) are available if coding is to be done

by using angles of inclination, the recommended limit

is 12 [82] Using this form of coding has been found

to be useful for indicating direction, angle, or tion on round instruments Other commonly usedforms of visual coding include di€ering brightness

posi-of lights (recommended limit is two levels) [81], anddi€ering ¯ash rates of lights (recommended limit istwo levels)

Using symbols for coding information is anotherimportant means of representing visual information.The e€ectiveness of symbolic coding depends on howstrongly the symbol is associated with the concept orobjects it is intended to represent The strength of thisassociation has been shown to depend on any existingand established association [83], and on the ease ofleaming any new associations The normal procedure

in setting up symbolic coding systems in the workplaceshould involve considerable experimentation withexisting and any new proposed symbolic codes Theexperimentation should involve the worker populationfor which the symbolic coding system is intended, andcoding system should be evaluated on the basis of theease of recognition, on matching symbols with whatthey represent (based on reaction time of participants),and based on the preferences and opinions of the users.Figure 12 provides examples of good and bad symboldesigns The symbol labeled ``bad design'' in the ®gurehas too much detail and is not simple in design Thesymbol labeled ``good design'' in the ®gure has all the

Table 15 General Recommendations for Designing a Good

Coding System

Make codes detectable by the human sensory mechanisms

under the given environmental conditions,

Make codes discriminable from each other by providing for a

di€erence threshold or a just-noticeable di€erence

Make codes meaningful to the user by providing for

conceptual compatibility

Where possible, standardize codes from situation to

situation

Use multidimensional codes to increase the number and

discriminability of coding stimuli used

Adapted from Ref 46. Figure 12 Examples of good and bad symbol design

details identifying the symbol within the boundary of

the symbol, and not outside the symbol

Dynamic information displays These displays are

used to present information about variables that are

subject to change in time Depending on the type of

information presented by the display, dynamic displays

can provide quantitative information, qualitative

information, check readings, and information on

situa-tion awareness measures

Quantitative visual displays provide informationabout the quantitative value of a variable of interest.The conventional types of displays used to conveyquantitative information include analog displays(®xed scale and moving pointer, moving scale and

®xed pointer) and digital displays (mechanical typecounters) Figure 13 provides some examples of thethree conventional types of displays Research in ana-log displays has provided certain general guidelines for

Figure 13 Commonly used quantitative displays

designing such displays [84] Fixed scale and moving

pointer displays are preferable to moving scale and

®xed pointer displays in most cases This is more so

especially when manual control is used to control the

moving element in the display (since it is better to

control the pointer rather than the scale) Also, any

small variations are better apparent when using a

moving pointer, ®xed scale device However, when

the range of numerical values is too large to be

accom-modated within the scale, the recommendation is to

use a ®xed pointer, moving scale display with

rectan-gular open windows in the scale for easier reference In

general, it has been determined that digital displays

perform better than analog displays where precise

numerical values are needed, and when the presented

numerical values are not continuously changing

In addition to these guidelines for the design of

quantitative displays, research has identi®ed numerous

characteristics that contribute towards making design

of quantitative displays e€ective and ecient Some of

these characteristics include the design of the scale

range (di€erence between the largest and the smallest

scale values), the design of the numbered interval in the

scale (numerical di€erence between adj acent scale

numbers), the design of the graduation interval (the

di€erence between the smallest scale points), the design

of the scale unit (smallest unit to which the scale can be

read), the numerical progressions used in scales, the

design of scale markers, and the design of scale

pointers [46] The numerical progression by 1's

(0; 1; 2; 3; † has been found to be the easiest to use

Decimals in scales, and scales with unusual numerical

progressions such as by 6's and 7's are discouraged

The most common recommendation for the length of

the scale unit is to use values ranging from 0.05 to

0.07 in The key factor in deciding the length of the

scale unit is that the values should be as distinct as

possible to permit easy human reading

Recommendations [81] are also available for design

of scale markers (see Fig 14 for a summary of these

recommendations) Some common recommendations

for design of pointers include having a pointed

(about 208 tip angle) pointers, and having the tip of

the pointer meet the smallest of the scale markers in the

scale Also, to avoid parallax between the scale and the

pointer, it is recommended to have the pointer as close

as possible to the surface of the scale [46]

Qualitative visual displays are used to present

infor-mation on a changing variable based on quantitative

information about a variable The information

presented could be indicative of a trend in the variable,

or a rate of change of the variable Also, qualitative

displays can be used to determine the status of avariable in terms of predetermined ranges (whetherthe fuel tank is empty, full, or half-full), or for main-taining a desirable range of values of a variable (such

as speed) The most common forms of presentingqualitative information through displays is by colorcoding or by using shapes (or areas to representvariables of speci®c interest, such as ``danger'') tocode the information Figure 15 provides an example

of a color- and area-coded qualitative display.Research [85] on check-reading displays (used todetermine if a particular reading is normal or not)has provided the following conclusions about thedesign of such displays:

1 In general, males make fewer errors in reading tasks than females

check-2 The accuracy of check reading is a function ofviewing time; fewer errors will be made if theexposure time is relatively long (greater than0.5 sec); also, check-reading performance di€er-ences between males and females become in-signi®cant when exposure time is increased to0.75 sec

3 The selection of background color is importantfor check-reading tasks; for exposure time lessthan 0.75 sec, black dials and pointers with awhite background lead to fewer errors incheck reading than with white dials andpointers on a black background; however, forexposure times greater than 0.75 sec, fewererrors result with a black dial background; the

®nal selection of the background color should

be based on the time routinely available forcheck reading (Fig 16)

4 Both the 9 o'clock and the 12 o'clock pointerpositions in the dial yield acceptableperformances; the actual design has then to bebased on user preferences

5 Check-reading performance is not a€ected bythe presence of between 1% and 3% deviantdials

6 The normal reading must be coded clearly; ifmany instruments are used in concert, the dis-plays must be con®gured clearly so that thedeviant reading stands out Figure 16 providesexamples of good and bad check reading dis-plays

One other type of qualitative display is statusindicator displays These indicators are usuallyrepresentative of discrete pieces of information such

as whether the condition is normal or dangerous, or

if the working surface is hot or cold Colored lights are

the most commonly used form of status indicators

Signal and warning lights are the types of dynamic

information displays that have relevance in the context

of industrial settings Detectability (of such lights) is

the most important design issue related to signal andwarning lights The detectability of signal and warninglights is in¯uenced by factors such as the size of thelight, the luminance, the exposure time, the color oflights, and the ¯ash rate of lights.Table 16 provideswidely accepted guidelines for the design of signal andwarning lights

Auditory displays: factors affecting design.Literature identi®es four different types of tasksinvolved in detection of auditory signals [46]: detection(determining whether or not a signal is present), rela-tive discrimination (differentiating between two ormore close signals), absolute identi®cation (identifying

a particular signal when only one signal is present),Figure 14 Recommendations on scale marker design for normal and low illumination viewing conditions

Figure 15 Example of area coded display

and localization (determining the direction of the signal

source) These functions are based upon fundamental

attributes in sound energy propagation, namely, the

frequency of sound, and the intensity of sound The

number of cycles of sound waves produced in one

second is called frequency Frequency of sound is

expressed in hertz (Hz) It is generally true that the

human ear can detect frequencies ranging from 20 to

20,000 Hz A related concept is the pitch of the sound

(pitch denotes the highness or lowness of a sound;

high frequencies result in high pitched tones, and low

frequencies result in low-pitched tones) The intensity

of sound is de®ned as the sound power in one square

meter of area (W/m2) Since it is dif®cult to measure

sound power level directly, the intensity of sound is

measured in terms of the sound pressure level Sound

pressure level, in decibels, is given by 20 log…P1=P0†;

where P1 is the sound power level corresponding tothe sound to be measured, and P0 is the sound powerlevel corresponding to 0 dB The sound pressure levelscan be directly measured using commercially availablesound level meters

The detectability of auditory signals depends uponthe environmental in¯uences (noise) present in thesignal In the presence of noise in the surroundings,the threshold of detectability of the signal is increased,i.e., the signal intensity must exceed this threshold if it

is to be detected A rule of thumb pertaining toauditory signal detection in the presence of noise ormasking states that the signal intensity (at the outerear) should be midway between the masked threshold

of the signal in the presence of noise and 110 dB Inquiet surroundings, the detectability of the signaldepends upon the frequency and the duration of theFigure 16 Check reading displays

signal The standard recommendation is that these

signals should be at least 500 ms in duration; if they

are shorter than this, the recommendation is to

increase the intensity of the signal Di€erent

recom-mendations have been made by researchers to improve

the detectability of auditory signals A summary of

these recommendations is provided in Table 17

The relative discriminability of auditory signals also

depends upon the intensity and the frequency of

sound, and the interaction between these two factors

Relative discriminability is usually measured in terms

of the just-noticeable di€erence, which is the smallest

change in the intensity or frequency that can be noticed

by humans 50% of the time The smaller the noticeable di€erence, the easier it is to detect thedi€erences in either intensity or frequency of sound.Research has shown that it is easier to detect thesmallest di€erences when the intensity of sound ishigher (at least 60 dB above the threshold level).Also, with respect to frequency, it is recommendedthat signals use lower frequencies for higher discrimin-ability However, since ambient noise is also a low-frequency sound, it is advisable to use signals in the500±1000 Hz range Also, it is good to keep signals

just-30 dB or more above the threshold level for ecientfrequency discrimination

It has also been determined that, on an absolutebasis (identi®cation of an individual stimulus presented

by itself), it is possible for the human ear to identifyfour to ®ve levels of intensity, four to seven levels offrequency, two to three levels of duration, and aboutnine levels of intensity and frequency combined.Sound localization is the ability to determine andlocalize the direction of the sound The di€erences inthe intensity of sounds, and the di€erences in the phase

of sounds are the primary measures by which thehuman auditory system determines the direction ofthe sound source It has been shown that for frequen-cies below 1500 Hz, if the source of the auditory signal

is directly to one side of the head, the signal reaches thenearer ear approximately 0.8 msec before it reaches theother ear Also, localization is dicult at lowfrequencies, since there is very little di€erence in thetime it takes for the signal to reach both earssimultaneously However, at high frequencies(generally above 3000 Hz), the presence of the headbetween the ears makes intensity di€erences morepronounced resulting in e€ective localization of thesound source

Design recommendations for auditory displays Asummary of recommendations for the design of audi-tory displays is provided in Table 18 This is inaddition to the recommendations in the table onwhen to use auditory displays, as opposed to visualdisplays

1.3.3.4 ControlsGeneral Considerations in Control Design Controlsare the primary means of transmitting the controllingaction to devices and systems Numerous factors affectthe design of control devices These factors include theease of identi®cation, the size of the control, control±response ratio, resistance of the control, lag, backlash,deadspace, and location In the following paragraphs,

Table 16 Recommendations for Design of Signal and

Warning Lights

Use signal and warning lights to warn of an actual or

potential danger

Use only one light in normal circumstances; if several lights

are used, have a master warning light to indicate speci®c

danger

For commonly encountered danger or warning situations, do

not use a ¯ashing light; use only a steady light For

situations that are new or occasional, use ¯ashing warning

lights

Use four ¯ashes per second when using ¯ashing warning

lights When using di€erent ¯ashing rates to indicate

di€erent levels of some variable, do not use more than

three such rates with one light

Have the signal or warning light at least twice as bright as the

background

Use red color for these lights and di€erentiate danger lights

from other signal lights in the immediate environment

Ensure that the warning lights subtend at least a visual angle

of 18

Adapted from Ref 84.

Table 17 Recommendations for Increasing the Detectability

of Auditory Signals

Reduce the intensity of noise near the frequency of the signal

of interest

Increase the intensity of the signal

Present the signal for at least 0.5±1 sec

Determine the frequency where noise is low, and change the

signal frequency to correspond this frequency

Present noise to both ears and the signal to one ear only

Introduce a phase shift in the signal and present the unshifted

signal to one ear and the shifted signal to the other

Adapted from Ref 86.

we will summarize and present recommendations from

research for each of these factors

The ease of identi®cation of controls depends upon

how well the controls have been coded The e€cacy of

the coding used can be determined using measures

mentioned in an earlier section, namely, using

detect-ability, discrimindetect-ability, compatibility, meaningfulness,

and the extent of standardization Controls can be

coded using shape, texture, size, location, operational

methods, color, and labels Shape coding uses tactual

sensitivity of the human for discriminating between the

di€erent shapes of the controls Figure 17 provides

examples of di€erent shapes that are commonly used

in controls Three di€erent types of textures have been

identi®ed as being suitable for coding control devices:

smooth surface, ¯uted surface, and knurled surface

The most important consideration when coding by

size is to provide adequate discriminability between the

di€erent sizes used For coding based on location of

controls, the recommendation is to use at least 2.5 in

between adjacent vertical controls, and at least 4 in

between adjacent horizontal controls In addition, it

is recommended that the general guidelines provided

in table be followed when coding controls based on

location There are instances when coding is based

on the method of operation of the control

(push-button controls, for example) Table 19 provides the

recommended minimum separation distances when

this is the case Such operational coding, is

undesir-able, however, when operation time or potential

opera-tor errors are considerations Another way to code

controls is by color Meaningful colors (such as redfor a danger button), combined with other codingdimensions such as shape and size, have been shown

to be e€ective in enhancing the discriminability of thecontrols Color coding, however, cannot e€ective insituations with poor illumination or in dirty environ-ments One of the most commonly used methods ofcoding controls is by labels In fact, labels areconsidered a minimum requirement in many situations

as they do not place extensive learning demands on theoperators Labels, however, have the disadvantage inthat they take time to read and are not useful as acoding method in situations that have a high operationspeed Also, the placement of the label on the controlhas been shown to pose accessibility problems to thereader Control devices can have unique combinations

of codes, or even redundant codes Considerationssuch as the illumination and the potential visualhandicaps of the operator, maintenance of mechanicalcontrols, and the speed and the accuracy with whichthe controls have to be operated, are other factors toconsider in designing controls for ease of identi®cation.Control±response ratio (denoted by C/R) is the ratio

of the movement of the control device to the ment of the system response By this de®nition, asensitive control will have a low C/R ratio (i.e., theresponse will be large even for a slight change in thecontrol) It is believed that human motor actions takeplace at two levelsÐat a gross-adjustment level, and at

move-a ®ne-move-adjustment level Hence the optimmove-al level of C/Rratio to use in a control device, is generally decided as a

Table 18 Checklist for Designing Auditory Displays

Compatibility Are the signal dimensions and the coded displays compatible with user excectations?

Approximation If the information presented is complex, are the signals attention-getting, and providing precise

information as well?

Dissociability Are the auditory signals of interest clearly discernible from other signals?

Parsimony Do the signals provide the correct amount of information?

Invariance Is a particular signal used for providing the same information every time?

Presentation Are the signals moderate and not extreme?

Is the signal intensity level such that it is not masked by noise?

Has care been taken not to overload the auditory system of operator by presenting too many signals atthe same time?

Installation Has the signal been tested with the target user group?

Are the new signals really new (are they noncontradictory to the existing signals?

If auditory displays are entirely new to the setting, have the operators been given enough time to adjust tothe new type of display?

Note: Answering ``Yes'' to all the above questions in the checklist is the desirable scenario.

Adapted from Refs 46, 87, and 88.

tradeo€ between the time it takes to accomplish the

gross movement and the time it takes for the ®ne

adjustment involved in a controlling action It has

been shown that an optimum C/R ratio is dependent

upon factors including the type of control (lever,

crank, wheel, etc.), the size of the display, and the

tolerance permitted in setting the control

Resistance in a control is responsible for providing

feedback about the controlling action to the operator

In essence, the resistance o€ered by the control is made

up of two fundamental elements: the force applied to

the control, and the distance to which this force is

applied (or the distance to which the control moves)

Free-position or isotonic controls o€er no resistance to

movement, and feedback to the operator is based on

the displacement that occurs Isometric controls, on

the other hand, provide feedback, based only on theforce or the pressure applied to the control Mostcontrols use a combination of both pure displacementand pure force mechanisms for providing operatorfeedback Control resistance can signi®cantly a€ectoperator performance by a€ecting the speed andprecision of control operations, by changing the feel

in the control, by changing the smoothness of the trol movement, and by subjecting the control to thee€ect of shock and vibration It is therefore vital toconsider control resistance when designing or selectingcontrols for a speci®c task Some design guidelinesregarding control resistance are provided inTable 20.Deadspace is de®ned as the amount of movementnear the null position of the control The amount ofdeadspace in a control device has been shown to a€ectFigure 17 Different shapes that have been commonly used and demonstrated to be effective for coding controls

con-the sensitivity, and hence con-the performance, of con-thecontrol system It has been shown by researchers thatdeadspace in a control device can be compensated to acertain extent by making the control-ratio relation-ships less sensitive.

Backlash in a control device is de®ned as the space at any control position Research on backlashshows that systems with high control gain need tohave minimum backlash to reduce system errors Ifthe control system design makes it impossible to reducethe backlash, the recommendation is to make the con-trol gain as low as possible, since humans have beenshown to cope badly with backlash errors

dead-Types of Control Devices Controls can be classi®ed

as being discrete or continuous controls based onwhether they transmit discrete (on and o€) orcontinuous (machine speed increase from 0 to

100 km/hr) information Controls are also classi®edbased on the amount of force required to operatethem (small or large) The most common types of con-trol devices used to transmit discrete informationand requiring a small force to operate include pushbuttons, keyboards, toggle switches, rotary selector

Table 19 Recommended Minimum and Maximum Separation for Di€erent Control Devices

Recommended separation (in inches)

Sequentially with di€erent ®ngers

Sequentially with di€erent ®ngers

Pedal Randomly with one foot 4 (between the inner 6 (between the inner

sides of the pedal) sides of the pedal)Randomly with one foot 8 (between the outer 10 (between the outer

sides of the pedal) sides of the pedal)Sequentially with one foot 2 (between the inner 4 (between the inner

sides of the pedal) sides of the pedal)Sequentially with one foot 6 (between the outer 8 (between the outer

sides of the pedal) sides of the pedal)Adapted from Ref 89.

Table 20 Recommendations on Control Resistance and

Control Operation

Control movements should be as short as possible

Positive indication of control activation must be provided to

the operator

Feedback on system response to control activation must be

provided to the operator

Control surfaces should be designed to prevent slippage when

activating

Arm or foot support should be provided to the operator

if precise, sustained positioning of the controls is

required

Controls must be provided with enough resistance to

avoid accidental activation due to the weight of hands or

feet

If a seated operator has to push a force more than 5 lbf on a

one-hand control, a backrest must be provided to the

operator

The operator has to be able to move the trunk and entire

body if both hands are required to exert more than 30 lbf

through more than 15 in in the fore-and-aft plan

The speed, force, and accuracy of controls should ®t most

people, not just the most capable

Adapted from Ref 90.

switches, and detent thumb wheels Common control

devices used to transmit discrete information and

requiring a large amount of force include detent levers,

large hand push buttons, and foot push buttons For

transmitting continuous information, the traditional

control devices such as rotary knobs, multirotational

knobs, thumb wheels, levers or joysticks, and small

cranks, require only a small amount force to operate

them On the other hand, other traditional control

devices used to impart continuous information, such

as handwheels, foot pedals, large levers, and large

cranks, need large amounts of force to manipulate

and operate In general, control selection for common

controls, such as toggle switches, rocker switches,

knobs, cranks, handwheels, etc., is based on

opera-tional factors such as speed, accuracy, space

require-ments, and ease of operation With the advent of

information technology, control devices such as

joy-sticks, trackballs, mice, touch tablets, light pens,

touch screens, etc., are becoming popular devices for

transmitting continuous information to the system

Technology has advanced to such an extent that

these modern devices demand only a small amount

of physical force from the human operator Given

the variety of both traditional and modern control

devices in use in industry (see Fig 18 for examples of

some of these control devices), it is beyond the scope of

this chapter to explain the design of each of thesedevices in detail Besides, many excellent design tablesand recommendations already exist in the literature fordesign and selection of control devices, and are widelyavailable The interested reader is referred to thesedesign guidelines Such guidelines can be found inSanders and McCormick [46], Woodson et al [11],Chapanis and Kinkade [91] Salvendy [92], EastmanKodak [90], etc

1.3.3.5 Other Design Considerations in

Information Presentation and ControlBesides the individual design factors a€ecting thedesign and operation of displays and controls, thereare other general considerations in display and controldesign that a€ect the overall e€ectiveness of theinformation presentation and control system as awhole We have chosen to present two such importantfactors They are compatibility, and grouping andlocation of controls

Compatibility This the relationship between theexpectations of the human and the input stimuli andresponses of the system with which the human is inter-acting Any system with human users should be com-patible with the human expectations In general, goodcompatibility will result in fewer user errors, and better

Figure 18 Examples of common control devices

human and overall system performance Literature

identi®es four types of compatibility [47] conceptual,

movement, spatial and modality compatibilities

Conceptual compatibility refers to the matching that

should exist between certain forms of stimuli such as

symbols, and the conceptual associations humans

make with such stimuli Movement compatibility (also

commonly referred to as population stereotypes)

denotes the relationship between the movement of

the displays and controls and the output response of

the system being controlled Numerous types of

move-ment compatibilities have been studied by researchers

The most important types of movement compatibilities

include the movement of a control to follow the

move-ment of a display, the movemove-ment of a control to

control the movement of a display, the movement of

a control to produce a speci®c system response, and

the movement of a display without any related

response The common principles of movement

compatibility for various types of displays and control

devices are presented inTable 21

Spatial compatibility refers to the relationship that

should exist between, the physical features, and

arrangement, of the controls and their associated

displays A good example of compatibility in physical

features between the displays and the controls is the

design of the function keys on a keyboard, and the

corresponding labels for these function keys In a

number of experiments with household stove tops,

human factors researchers have demonstrated

conclusively the need for physically arranging displays

and the associated controls in a corresponding and

compatible way

Modality compatibility is a fairly new addition to the

list, and refers to certain stimulus-response

combina-tions being more compatible with some tasks than with

others

Principles of Control-Display Arrangement in a

Workspace The physical location and arrangement

of the displays and controls in a given workspace also

has to be based on the human sensory capabilities, and

the anthropometric, biomechanical, and other

charac-teristics of the human user.Table 22 provides general

guidelines for locating controls in a workspace The

ideal goal of placing each and every display and control

at an optimal location and in an optimal arrangement

with respect to the human user, is dicult, if not

impos-sible, to achieve in practice A few general principles of

control-display location and arrangement are useful in

setting priorities and in determining tradeo€s for good

design, if not the optimal

According to the importance principle, componentsthat are vital to system goals should be placed inconvenient locations System experts determine whatthese vital goals are According to the frequency-of-use principle, components that are frequently usedshould be placed in convenient locations According

to the functional principle, components that are tionally related in the operation of the overall systemshould be grouped and placed together Figures 19a(before redesign) and 19b (after redesign) illustratethe use of the principle of functional grouping in theredesign of the machining controller of a DynamiteDM2400 bench-top programmable machining center.According to the sequence-of-use principle, componentsshould be arranged in the sequence in which they ®ndfrequent use in the operation of the system or in theperformance of a task Use of one or a combination ofthese principles requires that the system designercollect information about the human users involved(the user characterization step described in Sec 1.3.1

func-as the ®rst step in the process of solving human±machine interaction problems), the tasks involved(the task characterization step using task analysis tech-niques also described in Sec 1.3.1 as the second step inthe process), and the environment in which the userhas to perform the task (characterization of thesituation, again mentioned in Sec 1.3.1 as the thirdstep in the process) Based on extensive research, therecommendations that have been suggested fordesigning workspaces with various forms of displaysand controls are presented inTable 23

1.4 SUMMARYThis chapter presented the overall ``process'' of design-ing and evaluating systems involving humans andautomated devices The key elements involved in thisprocess were brie¯y described, and the essentials ofthese elements were presented in the form of guidelinesand recommendations for practice

Table 21 Common Principles and Recommendations for Movement Compatibility for Di€erent Displays and ControlsType of display±control relationship Principles of movement compatibility

Rotary displays and rotary controls in same plane For ®xed scale/rotary pointers, ensure that clockwise turn of the

pointer is associated with clockwise tum of the control.For ®xed scale/rotary pointers, clockwise rotation of pointer/display should indicate increase in value and vice versa.For moving scale/®xed pointer, ensure scale rotates in the samedirection as control knob

Ensure scale numbers increase from left to right

Ensure clockwise turn of control increases value

Linear displays and rotary controls in same plane When the control is located to the side of the display, the

common expectation is the display pointer will move in thesame direction of that side of the control which is nearest to it.The common expectation is pointer will move in the samedirection as the side of the control knob on the same side asthe scale markings on the display

The common expectation is a clockwise turn of a rotary controlwill increase the value on the display no matter where thecontrol is located relative to the display

Movement of displays and controls in di€ernt planes For rotary controls, the common expectation is a clockwise

rotation results in an increase in value

For rotary controls, the common expectation is a clockwiserotation results in movement away from individual and viceversa

For stick-type controls (both horizontally mounted onvertical plane and vertically mounted on horizontal plane), thecommon expectation is an upward movement of controlresults in an increase in value and an upward movement ofdisplay

Movement of power switches U.S system is switch-up is for on, and switch-down is for o€

British system is switch-up is for o€ switch-down is for on.Directional compatibility of operator movement The common expectation is that a control movement in a certain(when operator is not directly facing the control) direction produces a parallel movement of the indicator on the

display, irrespective of the position of the operator

The direction of movement of the display indicator when theindicator is in the visual ®eld of the subject, is the same as thedirection of movement of the controlling limb

The direction of movement of the display indicator when theindicator is in the visual ®eld of the subject, is the same as thedirection of movement of the control relative to the subject'strunk

Adapted from Refs 41, 89, 91, and 93±97