DOT.'FAA/RD-92126 Crew Resource Management:DOT-VNTSC-FAA-92-8Rosearch and Development Service An Introductory HandbookWashington. DC 20591

RSPA/VNTSC-- ---- --- -- ---- ---- ---

.. .. .. ... ... ... . .. ... ... ..

NX

COKI HF PROGRA

Richard.. Adams.

Advanced Aviation Concepts, Inc.

Final Report

~ August 1992

This document is available to the publicthrough the National Technical Information ServiceSpringfield, Virginia 22161

10U S. Department of TransportationFederal Aviation Administration

NOTICE

This document is disseminated under the sponsorshipof the Department of Transportation in the interest

of information exchange. The United States Governmentassumes ro liability for its contents or use thereof.

NOTICE

The Uniicd States Government does not endorseproducts or manufacturers. Trade or manufacturers'

names appear herein solely because they are consideredessential to the object of this report.

REPORT DOCUMENTATION PAGE ( Form ApprovedOMB No. ?704.0188

Pubtic reporting burden for this col iecion of informnation is estimated to average, 1hour per response, incLuding thetime for review ngi nttructions searching existing data sources, gathering and maitnaining the Oata needed, andcoapLeting and reylew ng the cotlection of informahion. Send comments regarding this burden es tiate or any otheraspect of this co Lectfon If nforqmtion, intruding suggestions for reducing this burden, to Washington HeWuartersSegices Directorate for nformation Operations aqd Reports, 1215 Jefferson Do is HIghway Suite 1204, Ar Lton A2 -1A..-UL.n41. -4 to the Officao- arrntdet. P r duc.n ret0704-01to.9Washintn C 2

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3 . REPORT TYPE AND DATES COVEREDAugust 1992 Final Report

4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Crew Resource Management: An Introductory Handbook A1070/FAlEZ

6. AUTHOR(S)James E. Driskell, Richard J, Adams

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATIONREPORT NUMBER

Florida Maxima Corporation Advanced Aviation Concepts, Inc.147 East Lyman Avenue 10356 Sandy Run Road DOT-VNTSC-FAA-92-8Winter Park, FL 32789 Jupiter, FL 33478

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORINGU.S. Department of Transportation AGENCY REPORT NUMBERFederal Aviation Administration DOT/FAA/RD-92/26Research and Development ServiceWashington, DC 20591

11. SUPPLEMENTARY NOTES

12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

This document is available to the public through the NationalTechnical Information Service, Springfield, VA 22161

13. ABSTRACT (Maximum 200 words)

Recent research findings suggest that crew resource management (CRM) training canresult in significant improvements in flightcrew performance. The objectives of thishandbook are to foster an understanding of the background and philosophy of CRM and top-,ovide an overview of the development, implementation and evaluation of CRM training.Currently, CRM programs have been implemented successfully at a number of airlines,large and small, civil and military. The variety of CRM training programs suggestthat there are a number of ways to achieve effective CRM.

14. SUBJECT TERMS 15. NUMBER OF PAGES62

Crew Resource Management (CRM). Air Carrier Training, Flight Crew Performance, Air CarrierAccident Prevention, Team Cooraination 16. PRICE CGOE

17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF REPORT OF THIS PAGE OF ABSTRACT

Unclassified Unclassified Unclassified

NSN 7540-01-280-5505 Standard Form 298 (Rev. 2-89)Prescribed by ANSI Std. 239-18298-102

PREFACE

Recent research findings suggest that crew resource management (CRM) training canresult in significant improvements in flightcrew performance. The objectives ofthis handbook are to foster an understanding of the background and philosophy ofCRM and to provide an overview of the development, implementation and evaluationof CRM training. Currently, CRM programs have been implemented successfully ata number of airlines, large and small, civil and military. The variety of CR1Mtraining programs suggest that there are a number of ways to achieve effectiveCRM.

-Accesion For--NTIS CAOTrC TAB3

IJ.:;t~fication

. . ..... ........ ..........

........................................... .......

Aval�bi~L Co: i ,

92-28443 ,

. ,.' ,,

iii ". ft.

METRIC/ENGLISH CONVERSION FACTORS

ENGLISH TO METRIC METRIC TO ENGLISH

LENGTH (APPROXIMATE) LENGTH (APPROXIMATE)

1 inch (in.) - 2.3 centimeters (cm) 1 millimeter (mm) w0.04 inch (in)

1 foot (ft) w 30 centimeters (cm) 1 centimeter (cm) - 0.4 inch (in)

1 yard (yd) w0.9 meter(in) 1 meter (in) a 3.3 feet (ft)

¶ mile(mi) - 1.6 kilomneters(km) 1 m eter (m) 1.1 yards(yd)1 kilometer (km) u0.6 mile (mi)

AREA (APPROXIMATE) AREA (APPROXIMATE)

I square inch (sq in, in2) -6.5 square centimeters (cm2) 1 square centi meter (cm 2) = 0.16 squ are i nch (sq i n,: n2)

1 square foot (sq ft. ft2) w0.09 square meter (in2) 1 square meter (in2) -1.2 square yards (sq yd. yd')

1 square yard (sq yd. yd2)) 0.8 square meter (m2) I square kilometer (kn2) = 0.4squaremile(sq ml,mn.i0)

1 square mile (sq mi, rni 2) . 2.6 square kilometers (kin2) I hectare (he) =10,000 square meters ,m2) = 2.5 acres

1 acre m0.4 hectares (he) u4,000 square meters (mn2)

MASS - WEIGHT (APPROXIMATE) MASS - WEIGHT (APPROXIMATE)

1 ounce (oz) -28 grams (gr) 1 gram (gr) =0.036 ounce (oz)

1 pound (Ib) =.43 kilogram (kg) 1 kilogram (kg) =2.2 pounds (II)

I short ton -2,000 pounds (lb) = 0-9tonne (t) 1 tonne (t) = 1,O000kilograms (kg) 1.1 short tons

VOLUME (APPROXIMATE) VOLUME (APPROXIMATE)

1 teaspoon (tsp) = 5 milliliters(mI) I milliliter (ml) w0.03 fluid ounce (fl oz)

1 tablespoon (tbsp) x 15 milliliters (ml) 1 liter (1) - 2.1 pints (pt)

1 fluid ounce (f I oz) - 30 milliliters (ml) 1 liter (1) = 1.06 quarts (qt)

1 cup (c) - 0.24 liter (1) 1 liter (1) - 0.26 gallon (gal)1 pint (pt) - 0.47 liter (1) 1 cubic meter (m]) .36 cubic feet (cu ft, ft3)

1 quart (qt) -0.96 liter (1) 1 cubic meter (in3) - 1.3 cubic yards (cu yd, yd])

1 gallon (gal) = 3.8 liters (1)1 cubic foot (cu ft, ft3) =0.03 cubiq meter (in3)

1 cubic yard (cu yd, yd3) a 0.76 cubic meter (in3)

TEMPE RATU RE (EXACT) TEMPERATURE (EXACT)

(N-c32)(5/9)J'F - yC (c(9/5) y + 3 2 1 C - x '

QUICK INC-H-CENTIMETER LENGTH CONVERSION

IN HE 0 2 4 5 6 7 89 10

CENTIMETERS 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21 3 24 2

2540

* QUICK FAHRENHEIlT-CELSIUS TEMPERATURE CONVERSION

-400 .22* .4 14' 32. so* 686 Be 100 1220 1400 Ise' 176' 190 212'

c .40* .30' .200 .100 00 1o0 200 300 40 so. 600 70 0 0 goo go 1000

For more exact and or other conversion factors, see NBS Miscellaneous Publication 286, Units of Weights andMeasures. Price S2.50. SD Catalog No. C1 3 10286.

iv

TABLE OF CONTENTS

Section Page

1. INTRODUCTION . . . . . . . . . . . . . . . . . . .1

Objectives . . . . . . . . . . . . . . . . . . . . . 1CRM Background and Philosophy . . . . . ....... 2Principles of Crew Resource Management ....... 8CRM Training . . . . . ................. . . 9Summary . . . . . . . . . . ...................... 10Overview of the Handbook . . o . . o ............ 11

2. CRM SKILLS . . o . . . . . . . . o. . o. . o . . o. 13

Communication Processes and Decision-Making . ... 16Communication . o . . . . . . . . . . . . 16Assertiveness . . . . . . . . .... .. 20Decision-Making . . . . ...... .. 23

Team Building and Maintenance ............. 24Leadership . . . . .- . .. . 24

Workload Management and Situational Awareness 28Workload Management ...... ............ . 28Stress M&nagement . . ............. 31

3. IMPLEMENTING CRM TRAINING ............... 33

The Systems Approach to Training . . . . . ... 33Step 1: Developing CRM Training . . o o . .. . . 33Step 2: Implementing CRM Training . . . . . . . . 45Step 3: Evaluating CRM Training o . o . .. .. . 48A Final Note on CRM Training . ...... ... .......... 51

4. SU 14ARY . . . . . . ............. . .................... 53

GLOSSARY ......... . .. . . . ... .. . ... ......................... 55REFERENCE MATERIALS ................ .................... .. 59

V

vi

Chapter 1: Introduction

Aviation Safety! This concept is readilyembraced by everyone in the aviationcommunity from flight crews to support staffto management. This was not always the casewith Crew Resource Management (CRM). Theconcept of crew resource management has beenboth blessed and cursed by those in aviation.It has been cursed because the emphasis oncrew resource management is relatively new,and people often have a healthy, skepticalreaction to new ways of doing things.

But CRM is also blessed by many because ofwhat it can accomplish. Recent researchfindings suggest that crew resourcemanagement training can result in significantimprovements in flightcrew performance. Notsurprisingly, a growing number of people inthe aviation community, from airlinemanagement to flight crews themselves, haveembraced crew resource management as aneffective approach to reducing flight errorsand increasiný, aviation safety. Currently,CRM programs have been implementedsuccessfully at a number of airlines, largeand small, civilian and military.

Objectives of This HandbookThe objectives of this handbook are to fosteran understanding of the background andphilosophy of Crew Resource Management, andto provide an overview of the development,implementation, and evaluation of CRMtraining. This handbook is written for Part135 and Part 121 carrier operators andmanagement, and is designed to serve as asupplement to Advisory Circular 120-51 asrevised, Crew Resource Management.

CRM Background and PhilosophyIt is useful to distinguish between thephilosophy of crew resource management andthe implementation of crew resourcemanagement (CRM training). There is generalagreement within the aviation communityregarding the principles underlying CRM.Most agree on key CRM concepts and the needto focus on crew skills and performance.However, there is less consensus regardinghow to implement CRM training. In fact,various training programs have appeared whichmeet the specific needs of individual users.The variety of CRM training programs suggeststhat there are a number of ways to achieveeffective crew resource management.What follows is a brief history of crewresource management, a discussion ofprinciples, and finally, an overview of CRMtraining.

We've been flying for over 90 years. Why CRMnow? The concept of crew resourcemanagement is not new. Anyone who thinksthat the Wright brothers did not makeeffective use of the resources at theirdisposal in 1903 at Kitty Hawk is certainlymistaken. Similarly, military and civilianpilot training programs have touched on CRMtopics for years. NASA's John Lauber recallsthe saying that if an idea is new, itprobably isn't good, and if it is good, itprobably isn't new. So, while the conceptsunderlying CRM are not new, what is new isthe heightened emphasis on crew resourcemanagement as one key to increased aviationsafety.From the 1950s to the 1990s we have witnesseda steady decline in aviation accidents (seeFigure 1). This decline in aviationaccidents has been attributed to betterequipment, better training, and betteroperating procedures. However, this happybig picture of system safety masks sometroubling data. As Figure 2 illustrates, asaccidents related to equipment weaknesseshave decreased, accidents attributed to humanweaknesses have increased. A comparison of

2

60-r.50

IW 1 40-- - ---

- 30z 2

< 20a)

60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90YEAR

Fiure 1. Total accident rate for commercial rmrA worldwide, 1959-1990.(Excludes sabotage, mltay actlion, tubulence and evax'aton Injuries.)(Boeing Commercial Airplane Group, 1991)

Figure 1 and Figure 2 suggests two points.First, Figure 1 indicates that after a sharpdrop in the 1960s, accident rates haveleveled off from 1970 through 1990. Second,the trends in causes of accidents illustratedin Figure 2 show that human error hasremained a major contributing factor inaviation accidents during these latter years.

I-HUMAN CAUSESzw0

L-0z0

00. MACHINE CAUSES0

W• TIME

5Figure 2. Changes in accident causal factors over time. (Internationalw Civil Aviation Organization, 1984)tY

3

Industry estimates of causal factors in aircarrier accidents are shown in Figure 3. Bya conservative estimate, well over 60% ofaircraft accidents have been attributed tocrew-related actions. in brief, it seemsthat the "human factors" contribution toaviation accidents may be a difficult problemto solve.

Pr A©?dacnt Percent of Total Accidents with Known Causes

F=ctor T o 10 20 30 40 50 6o 70--TTý; -r'_l

Flightorew 493 25i

Airplane 124 31

Maintenance 21 7

Weather 34 9

Airport/ATC 37 12 I

.,Also (Other) 47 23

Total withKnown Causes 56 207

Unknown or 1959 - 1990

Awaiting 117 70- eporsa Lust 10 years (1981 - 1990)

Total 873 277

Figure 3. Primary causal factors for commercial aircraft accidents, worldwide, 1959-1990.(Boeing Commercial Aiplane Group, 1991)

Concern with the factors underlying theseaccidents led NASA researchers in the 1970sto conduct a series of interviews with linepiLots to investigate their perceptions ofaviation mishaps. Charles Billings, GeorgeCooper, and John Lauber found that one mishapcomponent consistently mentioned by pilotswas inadequate training. Even moreinteresting, these researchers found that itwas not technical training that these pilotsfelt they lacked, but training in leadership,communication, and crew management. In otherwords, traditional training had done anexcellent job of imparting stick and rudderskills, but these pilots felt that they

4

needed more training in crew oaordination. Asubsequent analysis of jet transportaccidents between 1968 and 1976 revealed morethan 60 that involved problems with crewcoordination and decision making (Cooper,White, & Lauber, 1979).

These preliminary results, coupled with adogged determination to pursue answers toproblems that line pilots had identified,encouraged NASA researchers to conductfurther research and analysis. In oneclassic simulator study, B-747 flight crewswere observed in a highly realistic simulatedline trip from New York's Kennedy Airport toLondon (see inset: New York to London Minas.One Engine). During this tightly-scriptedscenario, an oil pressure problem forced thecrew to shut down an engine. The crew had todecide where to land the plane. Thisdecision was further complicated by ahydraulic system failure, bad weather, poorair traffic control, and a cabin crew memberwho demanded attention at the worst possiblemoments. Researchers found that t.here was awide variation in the performance of crewsduring this simulation. Most problems arose

New York to London Minus One Engine: The Ruffell indicator it 647,000 pounds, decided to make anSnih Simulator Study. over-gross-weight landing. A minute and a half later, theflight engineer rechecked the fuel as pat of the landingBecause the scenario Involved a high gross takeoff weight, checklist and became concerned about the gross weight.followed by an engine shutdown with a subsequent He spent a minute and a half rechecking calculations anddiversion, the crew needed to dump fuel to reduce the announced that the aircraft's gross weight computer must beaircrafts weight to maximize landing weight. As In actual in error. Two minites later, the simulator lands at 172 knotsline operations, this was a very busy period. In one case, with only 25 degrees of flap: a 1,000 foot-per-minute descentaftel the captain decided to dump fuel, the captain and the about 77,000 pounds over the correct weight, on a short, wetfirst officer togetner decided that 570,000 pounds was the runway.correct target landing weight. They reached the deciskliwithout consulting the flight engineer or any aircraft During the 32 minutes between the decision to dump fueldocumentation. The flight engineer then calculated a dump and the landing, the flight engineer was Interrupted 15 timestime of 4 minutes 30 seconds, which the captain accepted while performing specific tasks tailoring the amrount of fuel towithout comment even though it was approximately one-third be dumped in relation to the conditions and length of thethe actual time required. Without prompting, the flight landing runway. Nine of the interruptions came directJy orengineer recalculated the dump time to the nearly correct indirectly from the captain, four from the cabin crew member,figure of 12 minutes. and two from equipment problems. The flight engineer was

never able to complete and verify his fuel calculatons andInstead of dumping for 12 minutes, however, the flight dump times without Interruption, either by a routine pad ofengineer stepped after only 3 minutes, perhaps because he standard operating procedure, or by a request from thereverted to his original, erroneous estimate or because he captain orthe cabin crew member. He thus becamemisread the gross weight Indicator. Unsatisfied, he again overloodud and his work became fragmented. The captainstarted to recalculate, but the failure of the No. 3 hydraulic failed to recognize the situation and so did nothing to reso',esystem interrupted him. it.During the next eight minutes, the flight engineer was (Lauber, 1984)subjected to a high work load, but then noticed that the grossweight was too high and decided to refigure the fuel. Duringthat time, he was interrupted further and did nothing moreabout the fuel until the captain, noticing the gross weight

5m

not from a lack of technical knowledge orskills, but from poor resource management.Crews whose performance included a high rateof errors did a poor job of communicating,setting priorities, and sharing workload.Crews making few errors did a better job ofmanaging available resources.

In a subsequent analysis of the cockpit voicerecordings from this study, Foushee and Manos(1981) found that those crews whocommunicated more and who acknowledged theexchange of information made fewer errors.

Is CNM Training Necessary?

Factors related to faulty crew performance account for well over half of air carrier accidents.These Include:

* Eastern Airlines, Lockheed L-1011, Miami, Florida, December 29, 1972.* United Airlines, DC-8, Portland, Oregon, December 28, 1978.* Allegheny Airlines, Inc., BAC 1-11, Rochester, NY, July 9, 1978.* Air Florida, Boeing B-737, Washington, DC, January 13, 19R2.* Air Illinois, Hawker Siddley 748-2A, Pinckneyville, Illinois, October 11, 1983.• Galaxy Airlines, Lockheed Electra-L-188C, Reno, Nevada, January 21, 1985.* Air Ontano, Fokker F-28, Dryden, Ontario, March 10, 1989

This early work by Lauber, Cooper, Foushee,and many others culminated in the firstNASA/Industry Workshop on Resource Managementon the Flight Deock in 1979. This eventconverged the efforts begun by the militaryand by commercial carriers in this area.Subsequently, in the early 1980s, prograx•were developed and implemented by some aircarriers, including United Airlines, KLMa, PanAm, Trans Australia Airlines, and others.Other events, such as the 1986 NABA/XACWorkshop on Cockpit Resource ManagementTraining, and the biennial aviationpsychology symposia organized by Dick Jensenat Ohio State University, provided

6

opportunities to review progress in CR1Mprogram development.

The FAA officially recognized the value ofCRM types of training during this period byallowing a LOFT training period to be used asan approved period of training which could besubstituted for certain pilot's recurrentproficiency checks. More recently, SFAR 58,The Advanced Qualification Program (AQP),passed into law in 1990, has given greatlyexpanded latitude to air carriers with regardto training. One of the conditions of theAQP training option is that CRM training beincluded. It is projected that CRM may oneday be required in all formal aircrewcertification requirements.

The CRM concept has continued to evolve overthe last decade, guided by extensivefederal/'aniversity/industry research and bylessons learned from the implementation ofCRM programs at a growing number of airlines.FAA Advisory Circular 120-51 as revisedprovides a contemporary statement of CRMconcepts. This document underscores severalrecent developments in CRM:

* CRM has come to embody the entire flightoperations team, including the cabin crew,air traffic controllers, maintenance, andother groups that interact with thecockpit crew. A shift in terminologyreflects this emphasis: Cockpit ResourceManagement is now more appropriatelytermed Crew Resource Management.

0 A second recent initiative is theintegration of CRM skills with traditioraltechnical flying skills. Whereas CRMprograms stress the acquisition ofcrew-related skills, it is thought thatthese skills should ultimately beintegrated with technical skills in thenormal training and evaluation process.In other words, both technical skills andCRM skills interact to determineperformance on the flightdeck.Accordingly, these skills should betrained and evaluated together as part ofthe total training program.

7

* CRM programs have been in place for a timesufficient to allow a body of researchevaluating CR14 training effectiveness toaccumulate. The research results indicateclear evidence of positive changes inaircrew performance following theintroduction of CRM.

This brief look at the background of CRM hasnecessitated the omission of many importantcontributions by many people. However, it isnoteworthy that CRM program development hasbeen driven by input. from line pilots, notdreamed up in some ivory tower. Accordingly,CRM has become widely accepted within theaviation community.

Principles of Crew Resource ManagementCRM is defined as the effective utilizationof all available resources--equipment andpeople--to achieve safe, efficient flightoperations. Resources include autopilots andother avionics systems: operating manuals;and people, including crew members, airtraffic controllers, and others in the flightsystem. Therefore, the concept of effective

CRM is the effective CRM combines individual technical proficiencyutilization of all available with the broader goal of crew coordination,resources-hardware, thus integrating all available resources tonoftware, and achieve safe flight.personnel-to achievesafe, efficient flight The following principles are fundamental tooperations. the CRM concept:

"* Effective performance depends on bothtechnical proficiency and interpersonalskills.

" A primary focus of CRM is effective teamcoordination. The team encompasses theflight crew (cockpit and cabin),dispatchers, air traffic controllers,maintenance and others.

"* CR1 focuses on crew members' attitudes andbehaviors.

" Effective CRM involves the entire flightcrew. CRM is not simply a responsibilityof the captain, nor should CRM training beviewed as captain's training. Allcrewmembers are responsible for effective

8

management of the resources available tothem.

" The acquisition of effective CRM skillsrequires the active participation of allcrewmembers. Effective resourcemanagement skills are not gained bypassively listening to classroom lectures,but by active participation and practice,including the use of simulations such asLine-Oriented Flight Training (LOFT).

"* CRM training should be blended into thetotal training curriculum, includinginitial, transition, upgrade, andrecurrent training.

CRM TrainingCRM training programs come in many forms.Limited CRM training programs are nowavailable off-the-shelf from various sources.Specific organizations develop CRM programsto meet their own particular needs andcorporate culture. Therefore, someonereviewing current CRM training programs islikely to find a variety of programs andprogram acronyms. These include:

"• Flight Operations Resource Management(FORM),

"* Flight Deck Management (FDM),

"* Aircrew Resource Management (ARM),

"* Aircrew Coordination Training (ACT),

"* Flight Team Management (FTM).

This diversity reflects the difference insize, type of aircraft, mission, trainingfacilities, equipment, and financialresources of operators. Accordingly, nosingle training program is likely to meet therequirements of all operators.

All CRM training programs are built on theprinciples outlined above. FAA AdvisoryCircular 120-51 as revised may serve to builda consensus on program content by suggestingbasic CRM skills to be included in anyprogram of instruction. These skills aregrouped into three clusters:

9

1. Communications and Decision Skills. Thiscluster of skills includes behaviors relatedto communications and decisionmaking,including:

"* assertiveness

"* communications

"• decision making

"* conflict resolution

2. Team Building and Maintenance Skills.This cluster focuses on human interaction andteam management skills including:"* leadership

"* team management

3. Workload Management and SituationalAwareness. This cluster reflects skillsrelated to managing stress and workload,including:"* mission planning

"* stress management

"* workload distribution

These skills will be examined more closely inthe following chapter.

SummaryCrew resource management represents anapproach to improving aviation safety thatwas born of real life experiences of airlinepilots. They realized that technical skill

The overall goal of CRM alone was not enough to manage safely ais the blending of complex flight system. CRM emphasizes thetechnical skillsand effective utilization of all resourceshuman skillsto support available to the flight crew, includingsafe and efficient equipment and people. In addition tooperation of aircra. respecting the importance of traditionalstick and rudder skills, CRM focuses on those

other skills required for effective crewperformance. The overall goal of CR3 is theblending of technical skills and human skillsso as to support safe and efficient operationof aircraft.

10

Research and experience have both shown thatthe best CR1 training is like other effectivetraining - it will include three learningelements: awareness, practice, andreinforcement. CRM training should notfollow any single outline, however. It ismost effective when it is developed to meeteach user's unique set of needs.

Overview of the HandbookChapter 1 has introduced CRM. The followingthree chapters provide suggestions andexamples on what CRM is and how CRM trainingprograms can be provided.

Chapter 2 presents an explanation of thebasic CRM skills. This explanation citescases illustrating effective and ineffectiveutilization of these skills.

Chapter 3 provides guidance on developing,implementing and evaluating CRM training.

Chapter 4 provides a brief summary.

Following the text, a glossary is presentedas a key to the terms used in CR1 training.Finally, a bibliography is included toprovide supplementary reference material.

11

12

Chapter 2: CRM Skills

There are many skills required to fly anairplane safely. Some of these are referredto as technical or "stick and rudder" skills.Major categories of technical skills include:

Motor Skills: the physical control ofaircraft systems, aircraft attitude, andnavigation.

Procedural Skills: the execution ofstandard, abnormal, and emergency operatingprocedures.

Information Skills (Knowledge): the use ofinformation required to conduct safe airoperations in areas such as federalregulations, weather, and aircraft systems.

These skills constitute the technicalproficiency of crewmembers. As noted inChapter 1, these skills formed the primarybasis for the selection and training ofaviators for most of this century.

These skills are necessary for modernaircraft operations, but by themselves arenot sufficient to ensure safe flight. Inother words, these individual technicalskills must be paired with other crew-relatedskills to achieve safe flight operations.For example, it is not enough that acrewmember possesses the appropriatetechnical knowledge; each crewmember mustalso have the skills necessary to receive andto transmit information efficiently in thecrew setting--communication skills. The

The "Right Stuf'" for crewmember who tends to ignore input frommodern-day flight oper- others can be a hazard during normal flight,ations includes both and can be disastrous in emergencyindividual technical profi- conditions. Therefore, technical skills mustciency and crew resource be integrated with other crew-related skills,management skills, defined in Chapter 1 as CRM skills, to ensure

safe flight.CRM skills, those skills related to effectivecrew resource management, may be grouped intothe following categories:

13

communioation Prooesses and Decision Making:skills related to effective communicationsand crew decisions.

Team Building and Maintenance: skillsrelated to leadership/followership andmaintaining a supportive team environment.

Workload Management and SituationalAwareness: skills related to operationalawareness, planning, and managing stress andworkload.

Figure 4 provides an overview of the skillsthat determine flight performance. Note thatFigure 4 indicates that both technical skillsand CRM skills are necessary for effectiveflight performance. This view is consistentwith the recent initiative to integratetechnical and CRM skills in flight operationsand training.

Furthermore, each cluster of skills presentedin Figure 4 is broken down into basic orprimary-level skills. For example, specificskills that compose the CommunicationsProcesses and Decision Making cluster includecommunication skills, assertiveness skills,and decision making skills. The three majorCRM skill clusters provide one convenient wayto classify CRM skills. However, it is theprimary-level skills that form the basis forCRM training.

CRM skills within each of the three skillclusters are described in the followingsections. Each skill will be describedbriefly. Synopses of NTSB accident reportswill illustrate how skills can effect crewperformance. The purpose of this chapter isto provide an overview of selected CRMskills, and to demonstrate the importance ofthese skills to flight safety.

14

4, t

Tf

b .94

04 00 4) -0

v94R& ..... .~ '4

04

co 04

- 115

Communication Processes and Decision Making.The first cluster of CRM skills includesthose related to effective communication anddecision making. Three primary-level CRMskills within this cluster are described inthe following: Communication, Assertiveness,and Decision Making.

CommunicationOne of the most significant variablesrelevant to crew performance is theinformation flow within the cockpit andbetween the cockpit and other sources. Theeffective transfer of information is acomplex p- ess, and requires thatinformation be conveyed when needed,transferred clearly, attended to by thereceiver, understood and acknowledged by thereceiver, and clarified if needed. There arenumerous opportunities for breakdown in thisprocess.

The Aviation Safety Reporting System (ASRS)was created in 1976 by NASA and the FederalAviation Administration to provide a database for anonymous reports of aviationincidents. From the earliest months, itbecame obvious that common deficiencies inthe exchange of flight information werefrequently being noted in the reports to theASRS.

Billings & Reynard40 ............................................................................................................................... (1 9 8 1 ) a n a l y z e d a l a r g e

S......................................................................... group of the incidents30 ......................................................................... rep ort ed t o th e A SRS ,S......................................................................... find ing th at ov e r 70 % o f2D ....................................................................... the reports containedis .evidence of error in the10 .. .................... transfer of information.5 One of the most common

communication problemsraue c m.. (37% of the reported

,.* incidents) was failureFigure 5. CommunieWcene ermte repoid in the ASRS (filings & Re•wd, 1951). to initiate the

information transferprocess. In most of these cases, the neededinformation almost always existed, but it wasnot made available to those who needed it.Another common problem (37% of the incidents)was inaccurate, incomplete, ambiguous, orgarbled messages. Other problems included

16

the failure to transmit the message at theappropriate time (13%). In 11% of the cases,the message was either not received or wasmisunderstood. Only 3% of the informationtransfer problems were attributed toequipment failure.

Foushee and Manos (1981) also reviewed theASRS data base to examine incidents involvingcommunications problems. They observed thefollowing communication problems:

* 35% of the reports cited problems dealing with poor understanding anddivision of responsibilities. Often, the lack of appropriateacknowledgments and cross-checking was a factor.

0 16% were due to interference with pertinent cockpit communications byextraneous conversations between cockpit crewmembers or betweencockpit crewmembers and cabin crew.

* 15% of the incidents were due to information which one or morecrewmembers believed they had transferred, but due to interference orinadequacy of the message, was not transferred successfully.

* 12% reported a total lack of communication between crewmembers.Within this category, there were numerous examples of crewmembers notcommunicating regarding errors even when they had access to the correctinformation.

0 10% of the communication problems cited were due to overconfidence orcomplacency. Often, crewmembers assumed that everyone elseunderstood what was happening, when in fact, they did not.

This and related research suggests that:

"* Overall, there is a tendency for crews whocommunicate more often to perform betterthan crews who communicate less.

"* When more information regarding flightstatus is transferred, there are fewererrors related to system operation.

"* Crews who frequently acknowledge commands,inquiries and observations tend to makefewer errors.

In general, effective communication issupported by the following behaviors:

"* Convey information clearly, concisely, andin a timely manner.

"* Use standard terminology.

17

"* Advocate concerns and suggestions clearly

and assertively.

"* Acknowledge communications.

"* Provide information as required.

"• Repeat information.

"• Ask for clarification when needed.

"* Resolve conflicts constructively.

The following accident summary illustratesthe importance of these behaviors.

ACCIDENT SUMMARY - On July 19, 19. at about 3pm ailerons were slightly up, not damaged, and that the spoIlerslocal time, a DC-10 operated by United Airlines as flight 232, were locked down. There s no movement of the primaryexperienced a catastrophic fal'ure of the No. 2 tall mounted flight control surfaces.engine during cruise flight. Shortly after the engine failure, the 6. The captain directed the check airman to operate tWecrew noted that the hydraulic fluid pressure and quantity had throttles to free himself and the first officer to attempt tofallen to zero In all three redundant hydraulic systems. The maintain command of the flight controls. The check airmanengine failure precipitated damage that severed the three advised that the No. 1 and No. 3 engine thrust levers could nothydraulic systems, leaving the fllgnt control systems be used symmetrically, so he used two hands to manipulateinoperative. Approximately one minute after the engine failure, the throttles, Even so, he said that the airplane had athe flight data recorder Indicated no further movement of the continuous tendency to turn right and it was difficult to maintainflight control surfaces, a stable pitch attitude.The only means of control for the flight crew was from the 7. The captain reported to the approach controlier that theoperating wing mounted engines. The application of flight had no elevator control, they might have to make a forcedasymmetric power to these engine- changed the roll attitude, landing and asked the controller for the ILS frequencj, headinghence the heading. Increasing and decreasing power had a to the runway and length of the runway. He then Instructed thelimited effect on the pitch altitude. The airplane tended to second officer to start dumping fuel using the quick dump.oscillate about the center of gravity In the pitch axis. It was not 8. The captain asked the senior flight attendant If everyone Inpossible to control the pitch oscillations with any degree of the cabin was ready. She repo4ted In the affirmative and thatprecision. Moreover, becaue airspeed is primarily determined she observed damage on one wing. The captain sent theby pitch tnm configuration and power, there was no direct second officer back to inspect the empennage visually,control of airspeed. The crew foatnd that despite their bestefforts, the airplane would not maintain a stabilized flight 9. The second officer returned and reported damage to thecondition. The airplane subsequently crashed during an right and left horliz stabilizers. The captain replied thast'attempted landing at Sioux Gateway Airport, Iowa, There were what I thought." The captain then directed the fllghtcrew to285 passengers and 11 crewmernbers onboard. One flight lock their shoulder harnesses and to put everything away.attendant and 110 passengers were fatally injured. 10. Several seconds later, the controller alerted theEVENT HISTORY - About 1 hour and 7 minutes after takeoff, crewmembers to a 3,400 foot tower obstruction located 5 milesthe flight crew heard a loud bang or an explosion, followed by a to their right and asked how steep a right turn they could make.shuddering of the airframe. The following sequence of events The captain responded that they were trying to make a 30is In chronological order and is presented to summarize the degree bank, A crewmember commented that "1 can't handletype and variety of communications required. that steep of bank." The first officer stated, "were gonna have1. The flight crew determined that the No. 2 aft (tall mounted) to try it straight ahead Al..."engine had failed. The captain called for the engine shutdown II. The captain reported the runway in sight and thanked thechecklist. While shutting down the engine, the second officer controller for his help. The sontroller stated that the runway the(flight engineer) observed that the systems hydraulic pressure flight had lined up with was closed, but he added "thOtM workand quantity gauges indicated zero. sir, we're getting the equipment off the runway." The captain2. The first officer advised that he could not control the asked Its length and the controller reported 6,600 feet. Twelveairplane as it en:ered a right descending turn. The captain took seconds later the controller stated that Uwer was an open aidcontrol of the airplane and confirmed that it did not respond toflight control inputs. poblem.

3. The captain reduced thrust to the No. 1 engine and the 12. During the final 20 seconds befove touchdown, theairplane began to roll to a wings level attitude. airspeed averaged 215 knots, sink rate was 1.620 fet perminute and smooth oscillations in pitch and rolU continued. The4. A flight attendant advised the captain thai a UAL DC-10 captain recalled getting a high sink rate alarm from the groundtraining check airman was seated In the passenger proximity warning system and that at 100 feet above thecompartment and had volunteered his assistance. The captain ground, the nose of the airplane began to pitch downward.Immediately invited the airman into the cockpit. First contact was made by the right wing tip followed by the5. At the request of the captain, the check airman re-entered right main landing gear. The airplane skidded to the right ofthe passenger cabin and performed a visual inspection of the the runway, ignited, cartwheeled aind came to rest in anairplane's wings. He returned and reported that the inboerd inverted poietion. (NTSB, 1990.)

18

i i . . , , . . . . .

The quality and efficiency of the crewcommunications in UAL flight 232 is onefactor that minimized the loss of life inthis catastrophe. Specific communicationsbehaviors such as clarity (event #2),conciseness (event #5), timeliness (event #7)and acknowledgement (event #10) arerepresented in the event history.

A more intensive analysis of crewcommunication during this accident wasperformed by Predmore (1991), who broke downthe Cockpit Voice Recorder (CVR) transcriptinto the following communication categories:

"* Command-Advocacy (CMD-ADVOC)"* Inquiry

"* Incomplete-Interrupted (INCOMPL)

"* Reply-Acknowledge"• Observation

The following chart represents the cockpitand radio communications from flight 232during this emergency.

'70

10

020 28 30 32 34 36 38 40 42 44 4648 50 32 54 56 860 60

CVR TIME Pa minute.)E ]CMD/ADVOC GOBSERVANION [INQUIRY!

WN ICOMPL IS IEPLY/ACKNOWL

Figure 6. Communications ofthe UAL flight 232 crew. (Predmore, 1991)

Figure 6 reveals that the crew maintained aremarkably high level of communicationoverall. A second important factor is theconsistent level of acknowledgement evidentduring this event. It is important to notethat the crew of United 232 had receivedprior CRM training.

19

AssertivenessAccident reports reveal a number of instancesin which crewmembers failed to speak up evenwhen they had critical flight informationthat might have averted a disaster. In mostcases, this hesitancy involved a copilot orflight engineer who failed to question acaptain's actions or to express an opinionforcefully to the captain. These types ofincidents lead to the conclusion thatcrewmembers are often unwilling to state anopinion or to take a course of action, evenwhen the operation of the airplane is clearlyoutside acceptable parameters.

NASA's H. Clayton Foushee reported anincident from the ASRS data base whichillustrates this phenomenon. This reportdescribed a situation in which air trafficcontrol had instructed the aircraft to leveloff at 21,000 feet. As the aircraft reachedits assigned altitude, the copilot noticedthat the captain was allowing the airplane tocontinue climbing. The copilot alerted thecaptain, but not forcefully enough for thecaptain to hear. The copilot tried again andpointed to the altimeter, at which point thecaptain stopped the climb and began descentback to the assigned altitude. The copilotsummed up the reasons for his actions asfollows: t

The captain said he had misread hi altimeter and thought he was 1000feet lower than he was. I believe the main factor Involved hem was myreluctance to correct the captain. This captain Is very approachabM"and I had no reason to hold back. It Is just a bad habit that I think a lot ofcopilots have of double-checking everythvIng we say before we sayanything to the captain. (Foushee, 1982, p. 1063)

Assertiveness involves the ability to requestinformetion from others, make decisions, andcarry out a course of action in a consistentand forceful manner. Assertive behaviorincludes:"* Inquiry: inquiring about actions taken by

others and asking for clarification whenrequired.

"* Advocacy: the willingness to state what isbelieved to be a correc-. position and toadvocate a course of action consistentlyand forcefully.

20

* Assertion: stating and maintaining acourse of action until convinced otherwiseby further information.

Crewmembers are often hesitant to speak upfor several reasons:* Sometimes crewmembers fail to question

others' behavior because they are hesitantto point out incompetent behavior and toembarrass a captain or crewmember. Komich(1985) cites one copilot's hesitancy tocorrect a captain: "He's slow to catch hisairspeed and if I speak up every time he'sten knots low, it'll sound like aninstructional ride, so since he usuallycatches it at ten, I'll speak up atfifteen." Others have expressed fear ofcausing animosity and possibly creating areputation as that of someone who is"difficult to work with."

* Crewmembers sometimes hesitate to speak upbecause they perceive the captain as toointimidating. This kind of captain seeshimself as omnipotent, with the othercrewmembers there only to serve hiswishes, not to make any contributions tothe decision-making process. Foushee(1982) cites one extreme case:

Air traffic control had issued a speedrestriction. The copilot acknowledged andwaited for the captain to slow down.Assuming that the captain hadn't heard themescage, he repeated, "Approach said slow to180." The captain's reply was, "I'll do whatI want." Air traffic control inquired as towhy the aircraft had not been slowed, advisedthe crew that they had nearly collided withanother aircraft, and issued a new clearance,which the captain also disregarded. Followinga further advisory from the copilot, thecaptain responded by telling the copilot to"just look out the damn window."

* The status structure of the cockpit maycontribute to non-assertiveness amongcrewmembers. The captain has authority inthe cockpit and the responsibility forflight operations. People are naturallyhesitant to question those who have higher

21

status. Therefore, people tend to bedeferential to those in command. Second,crewmembers may simply assume that, sincethe captain is in charge, that the captain"knows what he is doing."

However, a lack of assertive behavior on thepart of crewmembers may have disastrousconsequences, as the following incidentillustrates.

ACCIDENT SUMMARY - About 1815 Pacific Standard the flight attendants ample time to prepare for the emergency,

Time on December 28,1978, United Airllnee Flight 173,. cockq"t procedures In the ewnt of an evauation after lending,

DC--61 aircraft crashed Into a wooded area during an rid the p ure On captain would be using during the

approach to the Portland International Airport. The aircraft = and lending.had delayed southeast of the airport for one hour while the 8. AT 1802:44, the flight engineer advised, "We got aboutflightcrew coped with a landing gear malfunction ani prepared three on the fel and thas It." The aircraft was about 5the passengers for an emergency landing. nautical miles south of the airport on a southwest heading.The NTSB determined the probable cause of the accident was 9. At 1802:44, Poitland Approach asked Flight 173 for a statusthe failure of the capl',ln to monitor propedy the aircrafet' fuel report. The first officer replied, "Yeah, we have indication ourstate and to properly respond to the crewmembers' advisories geer le abnormal. it'll be our Intention, In about five minutes, toregarding fuel. Contributing to the accl'ent was the failure of land on two eight left. We would like the equipment standingcrwnmembers either to fully comprehend the criticality of the by. Our Indications are the gear Is down and locked. We•vefuel state or to assertively communicate their concens to the got our people prepared for an evacuation In the event thatcaptain. should become neoesaary."

EVENT.J ISTORY - The first problem faced by the captain 10. At 1803:14, Portland Approach asked that Flight 173was the unsafe landing gear indication during the initial advlae them when the approach would begin. The captainapproach to Portland international Airport. This indication responded, "'heyve about finished In the cabin. I'd guessfollowed a loud thump, an abnormal vibraton, and an abnormal about another three, four, five rminutos." At this time, theaircraft yaw as the landing gear was lowered. aircraft was bout 8 nautical miles south of the airport.1. At 1712:20, Portland Approach requested, "United one 11. At 1806:19, the first light attendant entered the cockpit.seven three, contact the tower, one one eight point sven." The captain asked, "How you doIng?" She responded, "Well, IThe flight responded, "Negative, we'll stay with you. We'll tay think w're ready." At this time, the aircraft was about 17at five. We'll maintain about a hundred and seventy knots. We aeutical miles south ofe a•kport. Almost limultaneously thegot a gear problem. We'll let you know." first officer aid, "I think you just lost number Wour" followed

2. At 1746:52. the first officer asked the flight engineer. "How Immediately by advice to the flight engineer, *Batter get somemuch fuel we g0t...?" The flight engineer r , "FWe crosafeeds open there."thousand." The first officer acknowledged the response. 12. At 1806:45, the fis offir told the captain, "We're going to

3. At 1748:00, the first officer asked the captain,"... what' the lose an engine" The captain replied, "Why' at 1806, the firstfuel show now... r The captain replied, "FIve." At 1749, lfter officer again stated, "We're losing an engine." Again thea partlalry unintelligible comment by the flight engineer captain asked, 'Why'?" The firs officer responded, "Fuel."concerning fuel pump lights, the captain stated, "Tht's about 13. At 1807:12. the captain called Portland Approach andright, the feed pumps are starting to blink." At this point, requested, "Would like clearance for an approach into twoaccording to air traffic control data, the aircraft was about 13 eight left, now." The aircraft was about 19 nautical milesnautical miles south of the airport. south-southwest of the airport and turning left. This was the

4. About 1750:20, the captain asked the flight engineer to first request for an approach clearance from Flight 173."Give us a current card on weight. Figure about another fifteen 14. At 1813:21, the flight engineer stated, "We've lost twominutes." The first officer responded, "Fifteen minutes?" To engines, guys." At 1813:25, he stated. "We Just lost twowhich the captain replied, "Yeah, give us three or four engines- one and two."thousand pounds on top of zero fuel weight." The flight 15. At 1813:38. the captain sald, "They're all going. We can'tengineer then said, "Not enough. Fifteen minutes Is gonno make Trotudlle." The first officer said, 'We cant makereally mun us low on fuel here," anything."

5. From 1752:17 to about 1753:30, the flight engineer talked to 16. At 1813:46, the captain told the first officer, "Okay.Portland and discussed the aircraft's fuel state, the number of Dclare a mayday." At 1813:50, the first officer celled Portlendpersons on board the aircraft and the emergency lending International Airport Tower and declared, "Portland Tower,preopratlons at the airport. United one se vnty three heavy, Mayday. We're - the engines6. At 1756:53, the first officer asked, 'How much fuel you got are flaming out. We're going down. Wiere not going to be b•now?" The flight engineer responded tht 4,000 pounds to make the airport." This was the last radio transmission fromremained, 1,000 pounds in each toak Flight 173.

7. From 1757:30 until 1800:50, the captain and the first office (NTSB, 1979)engaged In conversatlon which Included discussions of gi0ng

22

In this accident, the flight engineer wasincreasingly concerned about the criticalfuel situation, making several observationsto the captain that were not acknowledged. Inits report, the NTSB stated:

Admittedly, the stature of a captain and his management style may exertsubtle pressure on his crew to conform to his way of thinking. it mayhinder interaction and adequate monitoring and force anothercrewmember to yield his right to express an opinion. (NTSB, 1979)

Decision MakingDecision making is a topic that may at firstglance seem to be an individual matter.After all, the captain is the final authorityand responsible for flight decisions.However, aircrew decision making is a groupprocess, and clearly illustrates thecollective nature of crew resourcemanagement. There are a number of hardware,software, and human resources available inthe cockpit, including other crewmembers,ATC, dispatch, and various sources ofinformation. The decision maker who does notrely on input from other crewmembers and fromother flight team members outside the cockpitis more likely to make poor decisions.

Although decisions are certainly founded onaeronautical knowledge, flying skills, andexperience, it is often difficult to describehow decisions are made in actual flyingsituations. It has generally been assumedthat learning to make good decisions could beattained only through experience. However,research has shown that aircrew decisionmaking skills can be shaped through training.

The decision making process may be brokendown into the following five steps:

1. Reoognizing or identifying the problem.Does a problem exist that requires action?

2. Gathering information to assess thesituation. This step requires determiningwhat information is needed, who has theneeded information, and whether theinformation is verified by othercrewmembers and resources.

3. Identifying and evaluating alternativesolutions. This step includes evaluating

23

the advantages as well as the risksassociated with each alternativeidentified, and selecting the optimumalternative.

4. Implementing the decision. This stepincludes executing the decision andproviding feedback on actions taken tocrewmembers.

5. Reviewing consequences of the decision.This step involves evaluating theconsequences of the decision and revisingthe decision if consequences are not asanticipated.

Some decisions, especially those that must bemade under extreme time pressure, must beseat-of-the-pants decisions. In these cases,there is very little time to gather allavailable information or to evaluatealternative solutions. These situations callfor intuitive decision making, which is basedon gut reactions, or more specifically, isbased on past experience and training.However, these emergency situations arerelatively rare. Most situations allowsufficient time to make a more deliberate oranalytical decision. This decision moreclosely follows the steps outlined above.Analytical decision making uses the resourcesavailable to the decision maker and resultsin more informed decisions.

Team Building and MaintenanceTeam Building and Maintenance skills includethose skills related to fostering effectiveteam performance.

LeadershipThe term "leadership" implies that this skillis relevant only to the captain. There aretwo reasons why this is not true. First, aflightcrew is a team with a clearlydesignated leader: the captain. The captainas designated leader retains the authorityand responsibility for flight operations.However, there are times when othercrewmembers must Flay functional leadershiproles. A functional leader may carry outleadership duties for a specialized task on atemporary basis, such as a takeoff or

24

landing. In this case, the crewmember mustdirect task activities and serve as afunctional leader to carry out that task.

Second, leadership would more properly becalled leadership/followership. Leadershipis a reciprocal process, and there arebehaviors that both a leader and a followermust apply to ensure effective performance.For example, one leader behavior might be toprovide direction for carrying out a task;correspondingly, one follower behavior mightbe to provide feedback on performance of thetask. In other words, leader behaviors areless effective without complementary followerbehaviors.

Leadership is not just "captain's" material.All crewmembers must perform leadershipduties in some situations. Furthermore,leadership is not a one-way process, butrequires both leader actions and effectivecrewmember responses.

Understanding the leadership role requires anunderstanding of what it is that leaders do.Effective leaders perform four primaryfunctions:

1. Regulating Information Flow. The leadermust regulate, manage, and direct the flow ofinformation, ideas, and suggestions withinthe cockpit crew and between the cockpit crewand outside sources. This function includesthe following behaviors:

* Communicating flight information

* Asking for opinions, suggestions

"* Giving opinions, suggestions

"* Clarifying communication

"* Providing feedback

"* Regulating participation

2. Directing and Coordinating CrewActivities. The leader must function as crewmanager to provide orientation, coordinationand direction for group performance. Thisfunction includes:

"* Directing and coordinating crew activities

"* Monitoring and assessing crew performance

"* Providing planning and orientation

25

e setting priorities

0 Delegating tasks

3. Motivating Crewmombers. The leader mustmaintain a positive climate to encourage goodcrewmember relations and to invite fullparticipation in crew activities. Thisfunction includes:

o Creating proper climate

* Maintaining an "open" cockpit atmosphere

* Resolving/preventing conflict

0 Maintaining positive relations

* Providing non-punitive critique andfeedback

4. Decision-making. The leader isultimately responsible for decisions. Thisfunction includes:

"* Assuming responsibility for decisionmaking

"* Gathering and evaluating information

"* Formulating decisions

"* Implementing decisions

"* Providing feedback on actions

The following excerpt from an NTSB accidentreport illustrates the errors that can occurwhen certain leadership and followershipbehaviors are applied poorly or not at all.

ACCIDENT SUMMARY - On October 11, 1963, Air Illinois genertor from the airplanes d.c. electrical system. ANFlight 710, a Hawker Siddley 745-2A was being operated subsequent attempts to restore the right generalor to thebetween Chicago, Illinois and the Southern Illinois Airport, airplanes d.c. distribution system were unsuccessful, and theCarbondale, Illinois, with on Intermediate stop at Spdngfield, airplane proceeded toward Carbondale relyng solely on ftIllinois. At 2020 central dayiigh time (CDT), Flight 710 betteft for d.o. electrical power.departed Springfield with seven peseengers and three EVET HISTR - The flight was about 45 minutes behindcrewmembere on board. About 1.5 minutes later, Flight 710 schedule when it arrived at Capitol Airport, Springfield, tIlinols.called Springfiekd departure control and reported that It had The fllghtcrew remained on board while the airplane wasexperienced a slight electrical problem but that it wan fueled. At 2011:44, when Flight 710 requested Its IFRcontinuing to its destination about 40 minutes way. clearance, N also requested 5,000 feet for No enroute altitude,The flight toward Carbondale was conducted in instrument as opposed to the IFR flight plan of 9,000 feet stored In themeteorological conditions. The cloud beses In the ares were at ARTCC computer. At 2019:40, Spdngfeld tow cleared2,000 feet MSL with tops at 10,000 feet. Visibility below the Flight 710 for takeoff.cloud bases was 1 mile in rain, and there were scattered 1. At 2021:14, Flight 710 contacted departure control andthunderstorms In the area. informed the controller that it was climbing through 1,500 fee.The Cockpit Voice Roorder Vaript showed that shortly The departure controller advised the flight that he had t inafter takeoff, Flight 71's left generator suffered a complete radar contact, cleared it to climb to and malntain 5,0)0 fest,mechanical failure and that in respondin to the failure of the and cleered it to proceed direct to Carbondale after it receieleft generator, the first officer mistakenly Isolated the right the Cartwi alate VOR.

26

2. At 2021:34, Flight 710 Informed the departure controller that Instructed her to brief the passengers that he had turned off theit had experienced a "slight electrical problem..." The controller excess lights because the airplane had experienced 'a bit of anasked the flight If it was going to return to Springfield, and the electrical problem..." but that they were going to continue toflight reported that it did not intend to do so. Carbondale.

3. At 2022:10, the flight told departure control that "We'd like 8. At 2038:41 (17 minutes after the Initial failure), the firstto stay as low as we can," and then requested and was cleared officer told the captain, "Well, when we... started losing the leftto maintain 3,000 feet. The controller asked the flight If he one I reached up and hit the right (isolate button) trying tocould provide any assistance, and the flight responded, "We're Isolate the right side because I assumed the problem was thedoing okay, thanks." right side but they (the generators) both still went off."

4. At 2023:54, the first officer told the captain that "the left 9. At 2044:59, in response to the captain's request, the first(generator) is totally dead, the right (generator) is putting out officer reported that the battery voltage was 20 volts. Atvoltage but I can't get a load on it." About 30 seconds later, he 2049:23, Kansas City center requested Flight 710 to changereported "zero voltage and amps on the left side, the right is radio frequencies. The flight acknowledged the request, whichputting out 27.5 volt; but I can't get it to come on the line." At was the last radio communication from Flight 710.2025:42, he told the captain that the battery power was going 10. At 2051:37, the first officer told the captain, "I don't know ifdown "pretty fast." we have enough juice to get out of this." At 2052:12, the5. At 2027.24, the captain called Kansas City center and captain asked the first officer to "watch my altitude, I'm going tostated that he had an "unusual request." He asked clearance go down to twenty-four hundred (feet)." He then asked the firstto descend to 2,000 feet "even If we have to go VFR." Healso officer If he had a flashlight and to have it ready. At 2053:18,asked the controller "to keep an eye on us If you can." The the first officer reported, "We're losing everything, ...down tocontroller told the flight that he could not clear ft to descend. about thirteen volts," and at 2053:28, he told the captain theThe captain thanked the controller and continued to maintain airplane was at 2,400 feet.3,000 feet. 11. At 2054:00, the captain asked the first officer if he had any6. At 2028:45, the captain said, "Beacons off... and Nav lights instruments. The first officer asked him to repeat, and atare off." At 2031:04, tVie first officer reminded the captain that 2054:16, the captain asked "Do you have any Instruments, doCarbondale had a 2.000 foot ceiling and that the visibility was 2 you have a horizon (attitude director indicator)'?" Flight 710miles with light rain and fog. There was no reply or crashed near Centralia, Illinois VORTAC located about 40acknowledgement from the captain, nautical miles n'orth of the Southern Illinois Airport. Threecrewmrembers and seven passengers were killed In the crash.7. At 2033:07, the flight attendant came forward and the

captain asked her if she could work with what she "had back (NTSB, 1984)there." The flight attendant reported that the only lightsoperating in the cabin were the reading lights, the lights by thelavatory, the baggage light and the entrance lights. The captain

This accident involved an HS-748-2A aircraftwhich experienced a generator failure atnight. Proper procedures were not followed,causing disconnection of the secondgenerator. A series of poorly managedactions followed, including an attempt tomake the destination on battery power alone.Several behaviors related toleadership/followership are identifiable:

* Poor monitoring and assessment of crewactivities (Event 8)

* Little feedback or acknowledgement ofactions (Event 6, 8)

* Utilization of critical informationunknown (Event 6)

Furthermore, a crucial error occurred whenthe co-pilot reminded the captain of IFRweather at the destination, got no responsefrom the captain, and did not press the issueuntil it was too late.

27

This accident shows that it is sometimesdifficult to apply vital team behaviorsrequired in the cockpit such as leadership,planning, problem solving, delegating,motivating, and setting priorities. Thisdifficulty reinforces the importance of CRMtraining.

Workload Management and Situational AwarenessThese skills reflect the extent to whichcrewmembers maintain awareness, prepare forcontingencies, and manage workload andstress.

Workload ManagementWorkload management includes preparation,vigilance and avoidance of distractions andcomplacency. Pilots interviewed aboutworkload management offered the followingtips:

Preparation - "Commit SOP' (Standard Operation Procedures), limitations andemergency procedures to memory, to free up mental capacity to deal withunforeseen events."

Planning - "Before each flight, I typically spend about one hour at homereviewing the route and airport information."

Vigilance - "Be especially vigilant when everything is going well." and "Neverassume anything, but verify and cross-check all critical Information."

Complacency - "Avoid complacency. The minute you think something won't hurtyou, it will."

Distractions - "Maintain a terrain awareness and a general knowledge of thetopography over which you are flying."

(from Kelly, 1991)

Workload varies according to the phase of anoperation, from the routine of preflightplanning and enroute cruise to the highworkload of a low visibility instrumentapproach. Either workload level can bedangerous.

28

Accidents often occur when workload demandsexceed crew capabilities. Figure 7illustrates the phases of flight in whichmost accidents occur. Note that takeoff,approach, and landing phases account for mostaircraft accidents. These phases are alsoperiods of high crew workload. If anydistraction or irregularity occurs duringthese phases, an accident is much more likelyto occur than at other times.

Percent of Accidents

4.2% 12.9% 8.5% 6.4% 5.7% 6.8% 7.0% 21.3% 27.2%

Takeoff Initial Climb Cruise Descent InitialClimb Approach

SHolding U

Nay Fix OuterMarker

1% 1 % 13% 80% 10 % 11% 3% 1%

Exposure - Percent of Flight Time

Figure 7. Percent of total commercial aircraft accidents by phase of flight,wortle, 1959-1990. (Boeing Commercial Airplane Group, 1991)

Paradoxically, low workload can also be ahazard to safety. Crews may be less alertduring long cruise segments. These lowworkload periods are times when complacency,forgetfulness and drowsiness are most common.Examination of the errors associated with lowand high workloads reveals that performancefollows a YERKES DODSON arousal curve likethe one depicted in Figure 8.

29

Higi (OPTIMAL AROUSAL)

CL

(UNDERSTIMULM ED) (OVERSTIMULATED)

LowArousal High

Figure 8. Very low o% very high arousal can degrade performanoe.

The following accident emphasizes therelationship between arousal and performance.Crucial elements identifiable in this excerptinclude the importance of unbroken attentionto instrument scan and the insidious role ofdistractions.

ACCIDENT SUMMARY - On December 29, 1972, an International Airport in Jamaica, New York to the Miami

Eastern Airlines Lockheed L-1 011 crashed In the Everglades Interntlonal Airport In Miami, Florida.

about 18 miles west-northwest of Miami International Airport 1. At 2334:05, EAL 401 called the MIA tower and stated, "Ah,(MIA). There were no major mechanical problems, severe tower, this Is Eastern. ah, four zero one, It looks like we'reweather phenomena or crew Incapacitation. The flight diverted gonna have to circle, we don't have a light on our nose gearfrom its approach because the nose landing gear position yet."Indicating system of the aircraft did not Indicato that the nose 2. At 2334:14, the tower advised, "Eastern four oh one heavy,gear was locked In the down position. The aircraft climbed to roger. pull up, climo straight ahead to two thusand, go back to2,000 feet MSL and followed a clearance to proceed st from approach control, one twenty eight sIx."the airport at that attitude. During that time the crew attemptedto cowrect the malfunction and to determine whether or not the 3. At 2335:09. EAL 401 contacted MIA approach control andnose landing gear was extended. Unfortiunately, during that reported, "All right, ah, approach control, Eastern four zeroperiod, woridoad management in terms of flying, navigating and one, we're right over the airport here and climbing to twocommunicating was totally Ignored due to fixation on the thousand fest. in fact. we've just reached two thousand feetrelatively minor failure. and we've got to get a green light an our nose gear."

The flight was conducted In clear weather conditions with 4. At 2336:04, the captain Instructed the first officer, who wasunrestricted visibility. However, the accident occurred In flyi the aircraft, to engage the autopilot. The firs officerdarkness with no moon. The flight was uneventful until the aclnowWedgd the oInstrction. Subsequently, the captain tookapproach to MIA. The landing gear handle was placed In the over the flying responsibilities. The first officer succeefully"down* position during the preparation for landing, and the removed the roms gear light lens assembly, but It jammedgreen light, which Indicates to the crew that the landing gear is when he attempted to replace It.fully extended and locked, failed to Illuminate. The captain 5. At 2337:01, the c4ptai Instructed the second officer torecycled the landing gear, but the green light still failed to enter the forward electronics bay, below the flight deck. toilluminate. check visually the alignment of the nose gear Indices.

The NatiotWl Transportation Safety Board determined that the 6. At 2337:48, approach omnol requested the flight to turn leftprobable cause of this accident was failure of the flight crew to to a heading of 270 degrees magnetic. EAL 401monitor the flight Instruments during the final four minut" of aocknowledged the request and turned to the new heading.flight, and to detect an unexpected descent soorn enough to Mearnwile, the flightcrew contiuod their attempts to free theprevent Impact with the ground. Preoucupation with a nose gear position light Is from Its reainer, without success.malfunction of the nose gear position Indicating system At 2338:34, th. captain again directed the second officer todistracted the crsw's attention from the Instruments and desc•nd Into the forward electronics bay and check theallowed the descent to go unnoticed, alignment of the nose gear Indices.

EVENT HJISTR - Eastern Airlines Flight 401 was ascheduled passenger flight from the John F. Kennedy

30

7. At 2340:3e, a half.second C-chord, which indiceled a uncommof n; and that more that one scan on the display woulddeviation of +/- 250 feet from the selected altitude, sounded In be required to ve.-fy a deviation requiring controller action.

the cockpit. No crewmember commented on the C-chord. No 9. At 2342:05, the first officer said, "We did something topitch change to correct the loss of satitude was recorded. A altitude," The captain's reply was "What?" The first officershort time later, the second officer raised his head into the asked, "Were still at two thousand, right?" The captaincockpit and stated, *1 cant see It. Ir pitch dark and I throw the Inmedletely exclaimed, *Hey, where happening here?'little light, I get, sh, nothing." The flightcrew and an EAL 10, At 2342:10, the first of six radio altimeter warning "beep"maintenance specialist who was occupying the forward sounds began; they ceased immediately before the sound ofobserver seat then discussed the operation of the nose initial ground Impact. The aircraft crashed while In a left bankwheelwell light. Afterward, the specialist went Into the of 28 degrees. The aircraft was destroyed. There were 163electronics bay to assist the second officer. passengers end s crew of 13 aboard the aircraft. Ninety-four

8. At 2341:40, MIA approach control asked, "Eastern, sh, four passengers and five crew members received ftal Injuries. Alloh one how are things comIng' slon% out there? This query other occupants received Injuries which varied from minor towas made a few seconds after the controller noted an altitude critical.reading of 900 feet In the EAL alphanumeric block on his radar (NTSB, 1973)display. The controller later testified that momentary deviationsIn altitude Information on the radar display were not

The distraction shown in this accident reportwas an operational one. Ironically, thedistraction itself was not serious. Thecrew's mismanagement of the distractioncaused it to be fatal.

Distractions can also come from outside thecockpit in the form of traffic, weather,unexpected rerouting, etc. The keys todealing with distractions in order to avoidcatastrophic consequences lie in the crew'sability to focus on aircraft control andhazard avoidance.

The crew should be prepared to avoiddistractions. The captain can fly theaircraft and delegate tasks that mightinterfere. Or vice-versa. While thedistraction in the L-1011 accident demandedimmediate attention, it was the captain'sresponsibility to set priorities and todelegate responsibilities or make workassignments. Tragically, the entire crewbecame absorbed in the distraction at theexpense of aircraft control.

Stress ManagementThere are two types of stress that candegrade flight performance. One type hasbeen called background stress. Backgroundstressors are chronic stress factors that arein the background of our everyday activities.They include job stress, stress to maintainschedules, fatigue, family stress, and the

31

stresses imposed by organizationalrequirements. Sometimes no single source ofstress may seem very serious, but thecombined or cumulative effects of stressorscan lead to degraded performance. In otherinstances, major life stressors such asdivorce or death cf a loved one can have illeffects on one's performance. Counseling oncoping skills has proven effective incontrolling background stressors.

A second type of stress is acute stress.Acute stress is the overload that occurs in ahigh intensity event, such as an unfamiliarflight irregularity. This type of stressoccurs all at once, and the results can becatastrophic. Acute stress results inseveral negative consequences:0 "Tunnel vision", or the restriction of

attention to only part of a task

0 Rigidity of response, or maintaining asingle course of action even thoughconditions have changed

* A tendency to scan alternatives less

effectively during decision making, and

0 "Ballistic" decision making; makingdecisions without thinking through theconsequences of a decision.

The intense time pressure, unfamiliarity, andoverload inherent in acute stress conditionsresult in a narrowed view. In turn,awareness is reduced regarding the hazards ofthe task, and the resources available to meetthose hazards.

32

Chapter 3: Implementing CRM Training

Early versions of CRM training were largelyattempts to adapt existing training materialsthat had been developed for other purposes,such as management training. But it becameclear that a more complete method was needed.

The Systems Approach to Training

The systems approach to training has beenused successfully in a variety of settings,including the military, airlines, and otherindustries. While the systems approach totraining exists in many forms, it generallyprovides guidelines for training in threesteps: DEVELOPING the training, IMPLEMENTINGthe training, and EVALUATING the training.It is not a rigid rule book but rather ageneral process for building trainingprograms.

What follows is a simplified scheme for CRMtraining, using the systems approach.

STEP 1: Developing CRM Training

Training program development activities maybe grouped in three areas: (1) NeedsAssessment; (2) Setting PerformanceObjectives; and (3) Preparing a TrainingPlan.

1. Needs AssessmentThe first step in the development of asuccessful CRM training program is to assessthe organization's training needs. There aremany ways to assess training needs, includingsurveys, reviews of incidents, and studies byspecial advisory committees.

The scope of the needs assessment depends onthe specific organization's size and

33

complexity. A comprehensive needs assessmentcan help an organization identify trainingneeds most vital to its operations and toarrange other needs by priority.

An organization may not be entirely receptiveto the new training that it needs. Oneuseful tool for assessing an organization'sattitudes toward CRM training is the "CockpitManagement Attitudes Questionnaire",developed by Helmreich (1984). (SeeAppendix) This questionnaire can provideuseful readings on embedded attitudes towardtraining. Further, this type of measurementencourages people to get involved in thedevelopment of the training program.

Training needs exist on at least threelevels: the individual, the group, and theorganization. We have mentioned theindividual in focusing on the group. But wehave said little about the organization. Is

Management support is the organizational culture supportive of CRMessential to effective training? Is management on board? A CR!'CRM training, program has to be supported from the top of

the organization to the bottom. Therefore,some training program may be needed formanagement. Management training may consistof distributing materials on crew-relatedincidents; posting information on CRMsuccesses such as the United Airlines flight232 story; or running a seminar formanagement describing the nature and value ofa CRM program. In any case, it is essentialto build management support early in programdevelopment.

To assist in the assessment of needs, a setof practical questions can be asked. Avariety of answers are appropriate. Thespecific answers depend on the uniquecharacteristics of the user organization.However, there have been some valuablogeneral lessons learned from the CRM trainingprograms that have been implemented atvarious airlines. Some of these lessonsfollow.

WHO?

Who should develop the training? Variouspeople can and Co develop CRM trainingprograms. Programs can be bought off theshelf or developed from scratch. They can be

34

developed by outside consultants or byin-house experts. Some sort of compromiseamong these options often provides the bestapproach.

Two hints have proven helpful in CRM programdevelopment. First, there are excellentmodels of CRM training available, includingprograms developed by several airlines and bythe military aviation community. And manypublications are available covering lessonslearned over the last decade on building asuccessful CRM training program.

Second, most organizations have found that ateam approach is effective in developing aCRM training program. Many organizations donot have training specialists on staff. Inthis case, the design and development teamshould include both outside trainingconsultants and in-house representatives.In-house representatives should comprise awide array of managers, flightcrew, checkairmen, and instructors. This team approachavoids the pitfall of having an outsideexpert design a program in isolation--aprogram that later proves not to fit anorganization's needs. Perhaps even moreimportant, this team approach ensures thatflightcrews are involved in the CRM trainingprogram from the beginning--a critical factorin securing acceptance and commitment byusers.

Who should be trained? The CRM concept hasevolved from an initial narrow focus on thecockpit crew. It has broadened to includeall other groups who interact with tha

CRM training Is cw cockpit crew and who are involved intraining, andmayinclude decisions that affect flight safety. Theseall groups that work with groups include cabin crewmembers, air trafficthe cockpit crew and are controllers, dispatchers, maintenance people,involved in decisions customer service agents, and even specializedthat impact flight safety. crisis teams such as bomb threat and hijack

teams. Although CRM training has focusedprimarily on cockpit crewmembers, someairlines have begun to develop CRM trainingprograms that include flight attendants,dispatchers, and maintenance people.

Central to the CRM concept is that CRMtraining is for the whole crew, not just forthe individual; that while individualexcellence is always desirable, teamwork is

35

the focus; and that a good blend of resources

is the best bet for a safe flight.

WHAT?

What should be trained? Early CRM trainingprograms borrowed heavily from managementtraining practices. Conveniently, managementtraining had traditionally dealt with topicslike teamwork and leadership. Managementconcepts were often applied broadly to theaircrew setting, with varying degrees ofsuccess. One characteristic of more recentCRM training programs is a shift from broadconcepts to specific aircrew skills andbehaviors. While no set curriculum isnecessarily appropriate for all airlines, theCRM primary-level skills identified in Figure4 provide one useful framework for CRMtraining. These primary-level skills can beused in developing performance objectiveslater in Step 2: Setting Performanceobjectives.

HOW?

How should training be introduced?Machiavelli wrote in The Prince that "Thereis nothing more difficult to arrange, moredoubtful of success, and more dangerous tocarry through than initiating changes."Resistance to change can take many forms,from passive disinterest to outrightsabotage. (The term sabotage, in fact, stemsfrom the French sabot or shoe, referring tothe wooden shoes thrown by workers into

The goal of successful machinery to jam gears). Because people tuneimplementation is that out what they are not motivated to hear, manythe new training a training program has beeA scuttled because"disappears" into the of a lack of user acceptance. This problemorganization: It becomes is sometimes called lack of "buy-in" oran accepted and routine "sign-up."part of the normaltraining program. Usually, introducing change in an

organization invites problems; but notalways. Sometimes the need for change is soglaringly obvious that buy-in is almostautomatic.In the early 1900s, for example,statisticians at American Telephone andTelegraph identified two powerful growthtrends: telephone use and population growth.Projecting these trends, they forecast that

36

by 1920 every female in the United Stateswould have to be employed as a switchboardoperator in order to meet demand. Within twoyears, AT&T had developed the automaticswitchboard. Everyone from overworkedswitchboard operators to impatient telephonecustomers welcomed the new system. Andtoday, of course, the automatic telephoneswitchboard is one of the key elements of ourso-called information age.

The trends in crew-related airline accidents(shown in Figure 3) are almost as obvious.And the consensus within the airlinecommunity strongly favors CRM training. Buta program known to be beneficial will notnecessarily be accepted. Even the best ofprograms may fail if the introduction of thetraining program is managed poorly. Peopleoften become accustomed to doing things in acertain way, and resistance to change isnormal. But some steps can be taken topromote acceptance of "new" training. Thosesteps are covered later in this chapter under"Implementing CR14 Training".

WHERE?

Where should training take place? InitialCRM training can take place in any settingthat is conducive to learning. Manyorganizations hold initial training inoff-site facilities in order to avoiddisruptions. However, the location ofinitial CRM training is less important thanthe process of bringing crew memberstogether.CRM focuses on orew interaction. And CRMtraining is most effective in groups largeenough to include entire crew units. Laterphases of training are most effective whentrainees are broken out into crew units andtrained in simulators or other cockpitmock-ups.

HOW MUCH?

How much training is required? Initial CRMCRM training must b* training can be accomplished in as little asintegrated Into the total two or three days. But continual CRMtraining program. training should become part of the total

training program. CRM skills should beconsidered a major element of the overall

37

skills package that produces safe flight.CRM training never really ends. More oncontinual training follows in this chapterunder "Preparing a Training Plan."

2. Setting Performance Objectives.The next step in the development of a CRMtraining program is to set performanceobjectives. Performance objeotives are thedesired outcomes of training. They answer

Performance objectives the question, What will trainees be able tospecify what Is to be do at the end of a training session that theylearned, could not do before training? Objectives

must be simple. It must be clear if theyhave been met at the end of training. Usableperformance objectives include the specificbehaviors desired and standards for measuringsatisfactory performance. A performanceobjective for classroom training might be:

* Given a written scenario describingaccident No. xxx, crewmembers will be ableto state in writing at least threebarriers to communication present in thissituation.

By expressing objectives in this manner,those people designing the training programare forced to identify exactly what eachlesson should accomplish. Later theseperformance objectives serve as guidelinesfor the evaluation of training. The basicCRM skills were discussed in Chapter 2, CRMSkills, and can be used to develop CRMperformance objectives.

There are different types of learningobjectives. Trainees may learn:

0 Intellectual skills, such as problemsolving

* Factual information (knowledge)

• Attitudes* Motor skills

Therefore, any one tral ng session ortraining module may have a mix of trainingobjectives.

38

CRM depends upon support from the entireorganization, not just the aircrew. Thephrase "organizational shell" has been coinedto describe the organizational culture inwhich flight crews operate. One aviationpsychologist writes:

Imagine, If you will, a beautifully designed and professionally executed CRM program that helpscrew members learn and practice precisely the skills that they need to operate well as a team in ademanding flight environment. Now place that program in an organization where lines of flight arebadly constructed and constantly changing at the last minute, crews are poorly composed andshort-lived, norms of conduct reinforce individual order giving and taking rather than team-levelplanning, excellent crew performance goes wholly unrecognized, and crews often are unable toobtain information, technical assistance, or material resources when they need them to proceedwith the work.

To complete a good CRM course in an organization that has... an unsupportive organizationalcontext is like getting all dressed up for a dance and having the ;ar break down halfway there.Cockpit resource management simply cannot take root and thrive unless organizational conditionsalso foster and support effective teamwork. (Hackman, 1987, p. 37)

The point is well made that performanceobjectives must be set for the organizationas well as the individual.

3. Preparing a Training Plan.The third step in the development of CRMtraining is preparing a training plan. Atraining plan provides a complete descriptionof a course of instruction including

A training plan provides sequenced lessons. Lesson details provide aa blueprint for CRM description of training objectives, content,training development, methods, training aids, and other elements

required for instruction. A sample of anabbreviated training plan is shown on thefollowing page.

A training plan identifies trainingobjectives, provides an outline of coursecontent, specifies training methods, and evenprovides an estimate of the time required foreach topic.

A variety of training methods may be chosento present course content. Since peoplelearn in a variety of ways--by listening, byseeing, by discussing, and by doing--itfollows that there are a number of trainingmethods available. Training methods includelectures, training tapes, seminars,

39

SAMPLE LESSON PLAN

Session Title: Communication Skills

Objectives:

1. Given a written scenario describing incident No. XXX, the crewmembers are ableto state In writing three barriers to communication present in this situation.

2. Crewmembers acknowledge communications from others during role-play.3. Crewmembers use standard terminology throughout a LOFT simulation.4..__

Topic Content Training method Time

Introduction A. Barriers to effective Videotape demonstration 10aircrew communication minutes

Lecture with viewgraphs 20minutes

Verbal and IA. Giving and receiving Classroom demonstration 10nonverbal feedback minutescommunica- IB. Acknowledging Lecture with viewgraphs 20tion communications minutes

C. Using standard terminology Role-play communications 20"D. Noiverbal communications exercise minutesI

Recognition A. How to recognize ineffective Role-play exercise with self 30communication skills in self and instructor critique. minutesand others

demonstrations, role-playing, and simulation.In general, it is preferable to use a varietyof training methods in order to enhancelearning and to sustain the trainees'interest.

Lesson developers may find the followingpoints helpful. First, education specialistshave identified nine critical instructionalevents that support learning:

40

1. Gain the trainee's attention.

2. Describe the objectives of training.

3. Stimulate recall of prerequisiteskills.

4. Present content to be learned.

5. Provide guidance and support.

6. Elicit the performance.

7. Provide feedback.

8. Assess performance.

9. Enhance retention and transfer.

(from Gagne, Briggs, & Wager, 1988)

Second, research shows that people rememberabout 10% of what they hear, 30% of what theysee, and 50% of what they see and hear.

WHAT YOU REMEMBER It follows thatthe more of thetrainee's senses

60 - are involved, thegreater the

50 learning

40 experience.Hands-on training

30- (hearing, seeingand feeling) is

20 better than10- watching a

training tape. Atraining videotapewith a soundtrack

What you What you What you is better than ahear see see and hear training audio

tape alone.

Third, it is always desirable to mixinstructional methods and actively involvethe learner. More than young people, adultsare known to respond to training that can bereadily applied in their work. Instructionshould not be limited to lectures andtraining tapes, for example. Role-playing,workshops and other practical exercises that

41

involve the trainee promote the sense thatthe training is relevant and useful.

Finally, all effective traininig, includingCRM training, comprises three essentialcomponents:

1. Indoctrination/Awareness

2. Practice and Feedback

3. Operational Reinforcement

1. Indootrination/Awareness. Initialtraining is conducted in a classroom settingwith the goal of laying a foundation forskills acquisition. One goal of training atthis stage is to convince the crewmember thatInitial training focuses on he or she can benefit personally from CRM

introducing CRM skills training. This goal can be met by reviewingand fostering awareness data on crew-related accidents, by describingof the need for CRM crew coordination problems, by viewingtraining, videotapes of good and poor crew performance,

and by assessing personal interaction styles.A second goal is to establish common conceptsand terms for CRM. Discussions,demonstrations, and role-playing in the CPR1skills are commonly used methods.2. Practice and Feedback. Following initialclassroom instruction, trainees are given theopportunity to apply the newly acquired CRMskills. The primary emphasis of CRM trainingat this stage is crew performance.

Changes in knowledge and attitudes sometimesare sufficient to effect changes inLearning new skills performance, but not always. Something more

requires active practice is usually required. Practicing skills andand feedback. receiving feedback on performance are proven

to be effective in completing the learning ofdesired behaviors.

Effective practice may be achieved throughhigh, medium, or low fidelity sa.mulations.Realistic high fidelity simulations may beachieved through blending CRM training intoLine Oriented Flight Training (LOFT). LOFTprovides the ideal setting for skillspractice, because it presents a full-crew,full-mission scenario in which CRM skills canbe evaluated along with technical skills.

42

CRM practice may range from full-missionsimulation to role-playing using cockpitprocedures trainers (CPTs) or even crudercockpit mock-ups.

Feedback on performance may also be providedby a number of means. Videotape provides aclear and vivid record of performance, and iswidely used. Replay, fast forward, slowmotion, and freeze frame are some of thevaluable videotaping functions that can beused during debriefings.

Under the guidance of a skilled instructor,the crew's critique of its own performanceduring debriefing can effectively build CRMskills. Other feedback valuable to theindividual comes from a pilot's fellowcrewmembers. And often the most lastingbenefit to the pilot comes from his ownself-assessment. Each of these assessmentsis vastly improved by the use of videotapingand other audio-visual aids.

3. Operational Reinforcement. All of ur havehit ourselves on the finger with a hammer.We experienced what is known as one-triallearning. We immediately grasped the conceptof hammer/finger interaction, and probablyvowed immediate]y never to do that again.However, the term "one-trial learning" isactually misleading. If we never practiceeven a simple skill like hammering, we aremore likely to hit our fingers again.

Sustaining the sophisticated skills of crewcoordination requires continual repetitionand reinforcement. Studies have shown that

Behavior that is not the initial benefits of CRM traininqrehearsed and reinforced dissipate within a relatively short period if

training is not reinforced over time. Thus,decays overtime. CRM training should be included as a regular

part of recurrent training. Since peopletend to forget both facts and skills uvertime, continual training should includerefresher seminars reinforced by practice andfeedback exercises such as LOFT.

•rA.jnWWthod selection may vary accordingto the component. of ClI training beingtaught. The purpoue of the indoctrination/awareness component of CRM training is toprovide concepts and terms for crewcoordination problems, and to address CRMskills. Indoctrination/awareness commonlybegins in a classroom setting. Methods of

43

instruction include lectures, seminars andguided discussions.

The purpose of the practice and feedbackcomponent of CRM training is to apply newlyacquired CRM skills. Accordingly, emphasisis placed on exercises, demonstrations,role-playing, and simulations. Trainees aregiven an opportunity to try out thesebehaviors and to receive feedback on theirperformanca.

The purpose of operational reinforcement isto prevent the loss of CRM skills once theyare acquired. CRM training must be includedas a regular part of the recurrent trainingrequirement. Continued reinforcement of CRMskills will require a mix of trainingmethods, including seminars, active practice,and feedback.

The mission of CRM training is to promotechanges in behaviors related to crewcoordination in order to prevent accidents.The bottom line is crew performance.Performance starts with awareness ofappropriate knowledge, attitudes, andbehaviors. But awareness is not enough.Like the athlete with great potential,crewmembers need an opportunity to hone theirperformance through realistic learningexperiences. The practice and feedbackcomponent of CRM training meets this need.

The best training method known for CRMpractice and feedback is simulated line

LOFT provides realistic, operations using flight simulators. Linefull-mission simulation to Oriented Flight Training (LOFT) is the bestenhance skill transfer. known of these programs. LOFT blended with

CRM training provides several advantages:

"* It embeds CRM into the total trainingprogram.

"* It provides a full mission scenario inwhich crews are evaluated for both CRMskills and traditional technical skills.

"* It allows crew training in a full crewcontext.

"* It invites the use of video recording andplayback during debriefing, providing anexcellent means of evaluating performance.

44

* It is accepted by most crewmembers andprovides an effective means of reinforcingCRM skills.

If LOFT is not available, CRM practice andfeedback can be accomplished with lower-levelsimulations. Realistic demonstrations,problem-solving scenarios, and role-playingexercises, complemented by audiovisual aids,can provide an effective means of reinforcingCRM skills. Keys to effective practiceinclude direct learner involvement, hands-ontraining experience, and immediate feedbackwith the opportunity for further practice.

STEP 2: Implementing CRM Training

It ir important to develop an implementationplan early in the training developmentprocess. Determine who should be involved inthe implementation; anticipate user reactionsand consider ways to overcome possiblenegative responses; and develop a timetablefor implementation. Finally, introduce thetraining carefully and deliberately. Someorganizations have tried out new trainingprograms on test groups, then asked forinput. They have then revised the programsbefore implementing them on a widespreadbasis.

Seek participation. When a change occurs ina corporate setting, it is crucial toencourage buy-in and discourage resistance.In the extreme, resistance can become

Getting key personnel outright rejection, sometimes called the Notinvolved in the design Invented Here syndrome. Not Invented Here isand development of CRM most likely when people do not feel involved.training is one way to Get as many key people as possible involvedenhance commitment. in the development of the CRM training. As

early as possible sign up the. opinion leaderswithin your organization and any othersignificant group such as the pilots'association.. Seek their participation early,and maintain their involvement while the CRMelement is worked into the total trainingproqram.

Demonstrate program support. In 1961, amajor American corporation established aschool for managers to teach a new approachto company operations. The new approach was

45

based on an innovative marketing philosophycentered on customer needs. After completingthe course, 85% of the management traineesleft the company within a short time. Thereason? The company had not changed, andmanagement had retained its habitual way ofdoing things. In this case, the corporateculture did not support the very change thatupper management had tried to introduce.This example illustrates that all levels inan organization must be on-board or a newprogram may quickly die for lack of support.

Thus, it is critical that CRM program supportbe conspicuous from top management through toline operations. Similarly, the support ofthe pilots' association can provide avaluable program endorsement.

Communicate. Commitment can't be assumed; itmust be earned. Therefore, it is vital tocommunicate the reasons for a CRM program.People need to know what CRM is and what itisn't, what it can accomplish, and why CRMtraining is important. Communication can beaccomplished through meetings, seminars,newsletters, and posters. The following is auseful guide to effective communications.

The Seven Cs of Effective Communication

"* Credibility: The messenger must be seen ascredible and competent by the receiver.

"* Context: The message should be delivered in acontext that is comfortable to the receiver andinvites his participation.

"* Content: The message must be relevant andimportant to the receiver.

"* Clarity: The message must be clear and must bedelivereci in terms the receiver will understand.Jargon is to be avoided.

"* Continuity and consistency: The message shouldbe reinfcrced often and should be :onsistentfrom all voices within the organization.

"* Channels: Channels of communication should befamiliar and relevant to the receiver.

"• Capability: The message should be suited to thecapabilities of the receiver.

46

Follow-up. The most valuable guidance comesfrom the student, not from the instructor.Monitor program implementation throughfeedback from CRM trainees. Be prepared tcaddress their concerns, and to revise theprogram as appropriate.

The Key Role of Chech Airmen and Instructors

The success of any CRM training program restson the skills of the people who administerthe training. Thus, it is vital that checkairmen and instructors be carefully chosen.Although check airmen and instructors areoften chosen for their proficiency intechnical matters, proficiency in teachingand evaluating CRM skills is an additionalcredential, not always easily earned.

Check airmen and instructors are especiallyCheck airmen and important for two reasons. First, they areinstructors must be directly responsible for instruction,highly skilled in all areas observation, feedback, and programof CRM performance. evaluation. Second, they are in a strategic

position to "sell" CRM and are likely to beopinion leaders within the organization.Because of their high profile, any attitudeshort of conspicuous support on their partcan undermine the entire CRM program, whichdepends upon voluntary buy-in at every levelof the organization.

There are at least three ways to encouragecheck airmen and instructors to support CRMand to become proficient in CRM skills. Forone, they can be given CRM training beforeanyone else. For another, they can be givenmore training than anyone else. Finally,check airmen and instructors can be signed upfor the CRM program at an early point. Theycan then provide input into the developmentof the CRM program and develop a sense ofownership of it. That sense of ownership, inturn, sets an example and encourages buy-inby others.

47

STEP 3: Evaluating CRM Training

Ex-Mayor Koch of New York was known forasking "How am I doing?" as he strolled NewYork's boroughs. This habit was not onlygood for public relations, but it alsoprovided Koch with immediate feedback on hisperformance.

There are several rnasons why the question"How are we doing?" needs to be askedcontinually during CRM training:

" To determine if the training meetsoLjectives. For example: Aftercompleting the training module onEffective Communications, are traineesactually communicating any better?

" To determine if results from courses meetoverall program goals. For example:After mastering the Communication lessonobjectives for individuals, are grewscommunicating any better?

"" To provide feedback to trainees."" To provide feedback to instructors."* To review and improve the training program

itself.

Preparations for training program (CRM)evaluaticn should be made during the trainingdesign phase, while training objectives are

The consideration of how being set. Objectives should be clear andtraining will be evaluated measurable. Objectives that are vague orshould begin in the eauly difficult to measure invite trouble instages of training design. evaluating the success of the program. A

clear plan for program evaluation should bedeveloped early in the training designprocess because evaluation activities shouldbegin before training is implemented.

As mentioned earlier, it is likely that avariety of outcomes will come from training,including changes in knowledge, attitudes,and behavior. It is appropriate to usemultiple measures of effectiveness inevaluating training. As an example,researchers at the Naval Training SystemsCenter have adopted the following measures ofeffectiveness for a Navy aircrew coordiijatio-training program:

48

Tvog of Evaluation Sm1MRe ItemIMeasure

Pro-Training Assessment 1. Aircrew coordination is critical toflight safety (agree/disagree).

2. List four barriers to effectibecommunication in the cockpit.

Trainee Reaction 1. This training was relevant to my flightperformance.

Learning 1. List four barriers to effectivecommunication in the cockpit.

Performance 1. Crew requests clarification of garbledcommunication during simulatedexercise.

Organizational Outcome 1. Data on incidents invu4ving crewcoordination.

Adapted from Cannon-Bowers et al. (1989).

Pro-training asmessment is done beforetraining begins. its purpose is to provide apre-training baseline measure of skills andattitudes. This pre-training baseline isused as a reference against whichpost-training improvements are measured.

Trainee reaction measurements can be taken toassess trainees' reactions regarding trainingprogram relevance, content, methods, or otherfeatures. "Tho Cockpit Management AttitudesQuestionnaire" is one evaluation tool that isused to assess pre-training and post-trainingattitudes toward crew coordinat i on (SeeAppendix).

Tracking various reactions to CRM training isvaluable for several reasons. First,negative reactions signal a lack ofacceptance by users. But they often point tospecific areas where small adjustments may

49

make a big difference immediately. Second,positive reactions probably point the way tofurther program improvemeats. Third, othergroups: reactions to the training program mayprovide valuable cues for improve!ments;groups like flight attendants, agents,mechanics, and dispatchers. And finally,reactions tracked over time can reveal muchabout changes within the corporation.Learning is measured against learningobjectives, which are just the flip-side oftraining objectives. The most familiar kindof measuremsnt is probably the question and

Evaluation can be used answer test. Answers may be oral or written.to assess reactions to As an example, one objective might be thattraining, learning, trainees be able to describe barriers tochanges in performance, communication in the cockpit. Measurementand the degree to which might take the form of a simple questionorganizational goals art-, requiring the trainee to describe fourmet. barriers.

Performance is measured against performanceobjectives. Typically, instructors receivespecial training in conducting performanceevaluations, and then observe and rate crewperformance against performance ubjectiveF.during simulation exercises.Organizational outcome measures includecomparisons of data on crew-relatedaccidents. Although increased safety is asimple enough ideal, attaining a validmeasure of the effect of CRM training onsafety is very difficult. One reason is thatit is difficult to analyze events that seldomoccur--aviation accidents. Another reason isthat it is often difficult to measure whetherchanges in the accident rate are related toCRM traininc or to other events, such asequipment upgrades or industry-wide changes.The difficulty in using such comparisons tomeasure training program effectiveness pointsto the need for multiple measures.Using multiple measures

of training effectiveness Multiple measures of effectiveness are usefulpenrits a comprehensive in documenting program successes andand informative training shortfalla. For example, trainees may failevaluation, because they have not learned the skills, do

not clearly understand CRM concepts, or failto see the t~i•,iiing as neaningful orrelevant.. Using multiple measures of aprogram'e effectivaness allows the evaluatorto isolate specific areas for programimprovement.

50

Finally, evaluation should not be punitive.The purpose of CRM training evaluation is tomeasure program effectiveness and to providespecific guidance for improvements to theprogram. Training evaluation should not beused to assess an individual's fitness forduty.

CRM skills are crew skills. CRM evaluationshould focus on crew or team performance.Individual crewmemoers should receive clearfeedback on their own performances, but theemphasis should remain on crew performance.

A Final Note on CRM Training

Setting up a CRM training pzogram is not asdifficult as it might seem to the CRMnewcomer. Using a systems approach theprocess c~an be broken into three steps:Development, Implementation, and Evaluation.Each of these steps can be further brokendown into manageable units such as NeedsAssessment, Setting Performance Objectives,and Preparing a Training Plan. And theseunits can be further broken down asnecessary.

rurthermore, good CRM training can beaccomplished with a conservative budget byblending it with existing training, usingexisting resources.

A well-conceived CRM training package canblernd nicely writh existing training and meetwith little resistance. While the CRMelement is perceived as new, it is oftentreated as a stand-alone training product.Over time, as the blending becomes complete,CRM should become indistinguishable fromother training. At that point it has become"seamless", or "invisible" to the trainees,the most desirable state.

Airlines that do not yet have CRM programsare in the enviable position of beiag able tobuild on the lessons learned from more than adecade of CRM program development. There isnow a wealth of materials available tosupport CRM program development. Eachdeveloping CRM program can be custom-fittedto the unique needs of the organization while

51

drawing on the abundant resources alreadyavailable.

52

Chapter 4: Summary

Commercial aviation is one of the safestforms of transportation, with a safety recordthat is excellent by any standards. Thenumber of commercial jet aircraft in serviceworldwide has climbed steadily over the pastthree decades to a total of 9,530 in 1990,while annual departures have increased to13,298,000 for 1990 (Boeing, 1991).Remarkably, over this same period, the totalaccident rate has declined from over 60accidents per million departures to about 2.5-- less than one twenty-fourth the accidentrate in 19591 (see Figure 1) This fortunatetrend can be attributed to advances inequipment technology, to a high level ofindividual technical proficiency, and toimproved operating procedures.

Despite the record, one challenge has notbeen met: Breakdowns in crew performancehave remained the primary factor incommercial jet accidents. Two out of threeaccidents are attributable to flightcrewerror.

In the early years, when equipmentreliability was the biggest problem, theaviation community responded with ingenuityand resolve. Engines and other aircraftcomponents became more reliable, and relatedaccidents declined. Today, with crewperformance the most significaat threat toaviation safety, the industry has respondedwith an ambitious program to supporteffective crew coordination and performance:Crew Resource Management training.

One aviation observer has projected that thisindustry focus on crew performance has thepotential to double system safety. Data onthe effectiveness of existing CRM programsindicate that this challenge is being met.

54

Glossary

ACTIVE LISTENING: the skill of hearing and understanding otherpeople and checking the accuracy of one's understanding bycommunicating with the sender.

ACTIVE LEARNING: physically performing the essential movementsor a skill or displaying behavior that has been taught (asopposed to passive learning, or simply listening to instruction).

ATTITUDE: a way of thinking or feeling; a mental disposition

towards. something that determines that person's response.

BEHAVIOR: a person's observable responses to a stimulus.

COMMUNICATION: the transfer of information and/or messagesbetween or among people by the use of words, letters, symbols, ornonverbal communication.

COMI*UNICATIONS PROCESS AND DECISION BEHAVIOR SKILLS: a clusterof CR1 behaviors related to effective communications and decisionmaking.

CREW RESOURCE MANAGEMENT: the effective utilization of allavailable resources--hardware, software and personnel--to achievesafe, efficient flight operations.

DECISION-MAKING: the process of selecting a course of actionfrom available options, based on information available at thetime.

FEEDBACK: response messages which clarify and ensure thatmeaning is transferred.

INDOCTRINATION/AWARENESS: the first phase in CRM training, whichprovides a conceptual framework for other phases. It typicallyconsists of classroom instruction focusing on identification ofCRM skills and concepts.

INQUIRY/ASSERTION: the skill of actively seeking out relevantinformation and showing a concern for both self and others'rights.

LEADERSHIP: the ability to utilize appropriate interpersonalskills to motivate, manage, and direct crew activities to achievea task.

LINE-ORIENTED FLIGHT TRAINING (LOFT%: a full miision simulationpresented in real-time which is usually videotaped for later crewself-critique.

55

NEEDS ASSESSMENT: the process of determining where training isneeded in an organization, what a trainee must learn in order toperform their Job effectively, and who needs training.

ORGANIZATIONAL CULTURE: the larger organizational environment inwhich the flightcrew operates. Structure, standards, and rewardpolicies are part of organizational culture. Management ororganizational support for CRM training is a critical componentof effective program implementation.

PASSIVE LEARNING: training that does not actively engage thetrainee in the instructional process; the trainee is expected topassively "absorb" inetructional material.

PERFORMANCE OBJECTIVESt statements which describe the desiredoutcomes of training in specific, behavioral terms.

PLANNIWG: the ability to establish an appropriate course ofaction for self and others to accomplish a specific goal.

PRACTICE AND FEEDBACK: the second phase of CRM training, inwhich participants actively employ newly acquired CRM skills andreceive feadback on their effectiveness.

REINFORCEMENT: the final phase of CRM training. This phase isongoing and involves ensuring that CRM becomes an inseparablepart of the organization's culture by garnering top managementsupport, identifying and reinforcing effective behavior in normalline operations, and instituting CRM training as a regular partof the recurrent training requirement.

ROLE-PLAYING: a training technique in which trainees are told toimagine themselves in the situations presented by the trainer.Trainees are free to act out different behaviors and reactions aslong as they stay "in role" throughout the session.

SITUATIONAL AWARENESS: an active awareness of internal andexternal conditions that affect flight safety. It includes arealization of current, past and future contingencies that mayaffect flight performance.

STRESS MANAGEMENT: any of a variety of techniques, methods, orgeneral strategies which have been developed to help people copewith the adverse consequences of stress.

SYSTEMS APPROACH TO TRAINING: a systematic approach to trainingdevelopment which provides guidelines for training design,development of instructional activities, implementation, andevaluation of training.

TEAM BUILDING AND MAINTENANCE SKILLS: a cluster of CRM skillsfocusing on interpersonal relationships and effective teampractices.

56

TEAM MANAGEMENT: command and leadership by the captain andsupportive behavior by crewmembers.

TEAM REVIEW: skills involved in pro-mission planning andanalysis, ongoing synthesis and evaluation of information, andpost-mission debriefing.

TRAINING PLAN: an outline of what will take place during atraining session, including training objectives, content,training methods, and training media.

USER ACCEPTANCE: the extent to which a training program isaccepted and endorsed by users.

WORKLOAD MANAGEMENT AND SITUATIONAL AWARENESS SKILLS: a clusterof CRM skills which reflect the extent to which crewmembersmaintain awareness of the operational environment; anticipatecontingencies; and plan and allocate activities that managestress and workload.

57

58

Reference Materials

General Articles

The following references include documents cited in the text,plus other relevant articles and documents of general interest.

Boeing Commercial Airplane Group (1991). Statistical sumiiary ofcommercial jet aircraft accidents, worldwide oDertions,1259-1%9Q. Seattle, WA: Product Safety Organization, Boeing.

Billings, C. E., & Reynard, W. D. (1981). Dimensions of theinformation transfer problem. In C.E. Billings & E.S.Cheaney (Eds.), Information transfer problems in theaviation system (NASA Technical Paper No. 1875). MoffettField, CA: NASA-Ames Research Center.

Caunon-Bowers, J. A., Prince, C., Salas, E., Owens, J. M.,Morgan, B. B., & Gonos, G. H. (1989, November). Datrminingaircrew coordination training effectiveness. Paper presentedat the l1th annual meeting of the Interservice/IndustryTraining Systems Conference, Ft. Worth, TX.

Foushee, H. C. (1984). Dyads and triads at 35,000 feet: Factorsaffecting group process and aircrew performance. AericanPsvchologist, 39, 885-893.

Foushee, H. C. (1982). The role of communications,socio-psychological, and personality factors in themaintenance of crew coordination. Ay.iation. Space. andEnvironmental Medicine, 53, 1062-1066.

Foushee, H. C., & Manos, K. L. (1981). Information transferwithin the cockpit: Problems in intracockpit communications.In C. E. Billings & E. S. Cheaney (Eds.), InfxMaiontransfer problems in the aviation system (NASA TP-1875).Moffett Field, CA: NASA-Ames Research Center.

Gagne, R. M., Briggs, L. J., & Wager, W. W. (1988). Principles ofinstructional desian. New York: Holt.

Hackman, J. R. (1987). Organizational influences. In H. W. Orlady& H. C. Fousbee (Eds.), Cockpit resource managementtraining: Proceedings of the NASA/MAC workshog (MSACP-2455) (pp. 23-39). Moffett Field, CA: NASA-Ames ResearchCenter.

Helmreich, R. L. (1984). Cockpit management attitudes. H=AunFactors, 26, 583-589.

59

Helmreich, R. L., Chidester, T. R., Foushee, H. C., Gregorich,S., & Wilhelm, J. A. (1989). How effective is cockpitresource management training? Issues in evaluating theimpact of programs to enhance crew coordination (NASA/UTTechnical Report 89-2).

Helmreich, R. L., Chidester, T. R., Foushee, H. C., Gregorich, S.E., & Wilhelm, J. A. (1989). Critical issues in implementingand reinforcing cockpit resource management training(NASA/UT Technical Report 89-5).

International Civil Aviation Organization (1984). Aprevention manual (1st. Ed.). ICAO Doc 9422-AN/923.

Jensen, R. S. (1989). Aeronautical decision making-___cp itrespurce management (DOT/FAA/PM-86/46). Arlington, VA:Systems Control Technology.

Kelley, W. (1991). Safety tips from the pros. Aviation Safety,11, (21).

Lauber, J. K. (1984). Resource management in the cockpit. AirlinePilot, 5_3_, 20-23, 38.

National Transportation Safety Board. (1973). Aircraft accidentreport: Eastern Airlines. Inc. L-1011. N310EA Miami Florida(NTSB/AAR-73/14). Washington, DC.

National Transportation Safety Board. (1979). Aircraft accidentreport: United Airlines Flight 173. DC-8-61. N8082U(NTSB/AAR-79/12). Washington, DC.

National Transportation Safety Board. (1984). Aircraft accident"report: Air Illinois HawkQe Siddley HS 748-2A. N748LL(NTSB/AAr-84/10). Washington, DC.

National Transportation Safety Board. (1989). Aircraft accidentreport: United Airlines B-737-200 structitral failure(NTSB/AAR-89/03). Washington, DC.

National Transportation Safety Board. (1990a). Aircraft accidentreport: United Airlines DC-10-10 engine explosion andlanding at Sioux City Iowa (NTSB/AAR-90/06). Washington, DC.

National Transportation Safety Board. (1990b). Aircraft accidentreoort: United Airlines B-747-122 cargo door failure andexplosive decompression (NTSB/AAR-90/01). Washington, DC.

Predmore, S. C. (1991). Xicrocoding of communications in accidentanalyse*: Crew coordination in United 811 and United 232. InR. S. Jensen (Ed.), Proceedings of the sixth annualinternational symposium on aviation Rsycholoav. Columbus,OH: Ohio State University.

Ruffell Smith, H. P. (1979). A simulator study of the interactionof pilot workload with errors. vigilance, and decisions(NASA Technical Memorandum 78482). Moffett Field, CA:NASA-Ames Research Center.

60

Books and Conference Proceedings

Cooper, G. E., White, M. D., & Laube.:, J. K. (Eds.). (1979).Resource management on the flig deck (NASA ConferencePublication 2120). Moffett Field, CA: NASA-Ames ResearchCenter.

Jensen, R. S. (Ed.). (1987). Proceedings of the fourth annualinternational sympos.ium on aviation Rsychology. Columbus,OH: Ohio State University.

Jensen, R. S. (Ed.). (1989). Proceedings of the fifth anIualinternational svrmosium on aviation DsychologV. Columbus,OH: Ohio State University.

Jensen, R. S. (Ed.). (1991). Proceedings of the sixth annualinternational symposium on aviation Dsvchology. Columbus,OH: Ohio State University.

Komich, J. N. (1985'. An analysis of the dearth of assertivenessby subordinate crewmembers. In R. S. Jensen and J. Adrion(Eds.), Proceedings of the third symposium on aviationpsychology (pp. 431-436). Ohio State University, Columbus,OH.

Orlady, H. W., & Foushee, H. C. (Eds.). (1987). Cockpit resourcemanagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Wiener, E. L., & Nagel, D. C. (Eds.). (1988). Human factors inaviation. San Diego: Academic Press.

CRM Tr -ninc Applications

Bruce, K. D., & Jensen, D. (1986). Cockpit management training at"Žople Express. In Orlady, H. V., & Foushee, H. C. (Eds.).Lockit resource managementot_raining (NASA ConferencePublication 2455). Moffett Fiell, CA: NASA-Ames ResearchCenter.

Christian, D., & Morgan, A. (1986). Crew coordination concepts:Continental Airlines CRM training. In Orlady, H. W., &Foushee, H. C. (Eds.). C_,ckpit resource management training(NASA Conference Publication 245!). Hoffett Field, CA:NASA-Ames Research Center.

Davidson, J. (1986). Introduction tc Trans Australia Airlines CRMtraining. In Orlady, H. W., & Foushee, H. C. (Eds.).Cockpit resource management traininjg (NASA ConferencePublication 2455). Moffett Field, CA: NASA-Ames ResearchCenter.

61

Scott-Milligan, F., & Wyness, B. (1986). CRM training and humanfactors training: what Air New Zealand is doing about it. InOrlady, H. W., & Foushee, H. C. (Eds.), CockDit resourcemanagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Sellards, R. (1986). Cockpit resource management at USAir. InOrlady, H. W., & Foushee, H. C. (Eds.), Cockpit resourcemnagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Vandermark, M. J. (1989). A comprehensive crew management programfor the America West Airlines pilots and attendants. In R.S. Jensen (Ed.), Progeedings of The Fifth Annual Symposiumon Aviation Psychology (pp. 640-645). Columbus, OH: OhioState University.

Wilkinson, D. (1989). Cockpit resource management training - TheQantas experience. In R. S. Jensen (Ed.), Proceedinas of TheFifth Annual SVmposium on Aviation Psychology (pp. 456-461).Columbus, OH: Ohio State University.

WiDson, T. (1989). Aircrew team management - The Australianexperience. In R. S. Jensen (Ed.), Procedinas of The FifthAnnual Symposium on Aviation PsycijpL:y (pp. 4o2-467).Columbus, OH: Ohio State University.

Yocum, M., & Monan, W. (1986). CRM training in corporate/regio,'.loperations (Group V). In Orlady, H. W., & Fcushce, H. C.(Eds.), gockpit resource management training (NASAConference Publication 2455). Moffett Fiold, CA: NASA-AmesResearch Center.

62