Root Cause Analysis

QUALITY PROGRESS I JULY 2004 I 45

Root Cause AnalysisFor Beginnersby James J. Rooney and Lee N. Vanden Heuvel

oot cause analysis (RCA) is a processdesigned for use in investigating and cate-gorizing the root causes of events with safe-

ty, health, environmental, quality, reliability andproduction impacts. The term “event” is used to

generically identify occurrences that produce orhave the potential to produce these types of conse-quences.

Simply stated, RCA is a tool designed to helpidentify not only what and how an event occurred,but also why it happened. Only when investiga-tors are able to determine why an event or failureoccurred will they be able to specify workablecorrective measures that prevent future events ofthe type observed.

Understanding why an event occurred is thekey to developing effective recommendations.Imagine an occurrence during which an opera-tor is instructed to close valve A; instead, theoperator closes valve B. The typical investiga-tion would probably conclude operator errorwas the cause.

This is an accurate description of what hap-pened and how it happened. However, if the ana-lysts stop here, they have not probed deeplyenough to understand the reasons for the mistake.Therefore, they do not know what to do to pre-vent it from occurring again.

In the case of the operator who turned thewrong valve, we are likely to see recommenda-tions such as retrain the operator on the proce-dure, remind all operators to be alert when

R

QUALITY BASICS

In 50 WordsOr Less

• Root cause analysis helps identify what, how

and why something happened, thus preventing

recurrence.

• Root causes are underlying, are reasonably

identifiable, can be controlled by management

and allow for generation of recommendations.

• The process involves data collection, cause

charting, root cause identification and recom-

mendation generation and implementation.

manipulating valves or emphasize to all personnelthat careful attention to the job should be main-tained at all times. Such recommendations do littleto prevent future occurrences.

Generally, mistakes do not just happen but canbe traced to some well-defined causes. In the caseof the valve error, we might ask, “Was the proce-dure confusing? Were the valves clearly labeled?Was the operator familiar with this particulartask?”

The answers to these and other questions willhelp determine why the error took place andwhat the organization can do to prevent recur-

rence. In the case of the valve error, example recommendations might include revising the procedure or performing procedure validation toensure references to valves match the valve labelsfound in the field.

Identifying root causes is the key to preventingsimilar recurrences. An added benefit of an effectiveRCA is that, over time, the root causes identifiedacross the population of occurrences can be used totarget major opportunities for improvement.

If, for example, a significant number of analysespoint to procurement inadequacies, then resourcescan be focused on improvement of this managementsystem. Trending of root causes allows developmentof systematic improvements and assessment of theimpact of corrective programs.

Definition Although there is substantial debate on the defi-

nition of root cause, we use the following: 1. Root causes are specific underlying causes.

2. Root causes are those that can reasonably beidentified.

3. Root causes are those management has controlto fix.

4. Root causes are those for which effective rec-ommendations for preventing recurrences canbe generated.

Root causes are underlying causes. The investi-gator’s goal should be to identify specific underly-ing causes. The more specific the investigator canbe about why an event occurred, the easier it willbe to arrive at recommendations that will preventrecurrence.

Root causes are those that can reasonably beidentified. Occurrence investigations must be costbeneficial. It is not practical to keep valuable man-power occupied indefinitely searching for the rootcauses of occurrences. Structured RCA helps ana-lysts get the most out of the time they have invest-ed in the investigation.

Root causes are those over which managementhas control. Analysts should avoid using generalcause classifications such as operator error, equip-ment failure or external factor. Such causes are notspecific enough to allow management to makeeffective changes. Management needs to knowexactly why a failure occurred before action can betaken to prevent recurrence.

We must also identify a root cause that manage-ment can influence. Identifying “severe weather”as the root cause of parts not being delivered ontime to customers is not appropriate. Severe weath-er is not controlled by management.

Root causes are those for which effective recom-mendations can be generated. Recommendationsshould directly address the root causes identifiedduring the investigation. If the analysts arrive atvague recommendations such as, “Improve adher-ence to written policies and procedures,” then they probably have not found a basic and specificenough cause and need to expend more effort in theanalysis process.

Four Major StepsThe RCA is a four-step process involving the fol-

lowing: 1. Data collection.2. Causal factor charting.

46 I JULY 2004 I www.asq.org

QUALITY BASICS

Identifying “severe weather”as the root cause of parts not

being delivered on time tocustomers is not appropriate.


Causal Factor ChartFIGURE 1

Aluminum melts,

forming hole in pan

Electricburner

shorts out

Grease igniteswhen it

contactsburner

Fire startson thestove

Mary meetswith Jane

Arcing heatsbottom ofaluminum

pan

Mary leavesthe fryingchicken

unattended

Jane ringsthe doorbell

Jane comesto the door

Mary beginsfrying

chicken

Mary uses an

aluminum pan

CF

CF

Mary

Pan

Jane

Jane, Mary

Mary

Burner

Pan

Pan

Conclusion

Mary

Mary

10 minutes

Firegenerates

smoke

Assumed

Mary runsinto thekitchen

Mary

Smokedetectoralarms

Jane, Mary

About 5:10 pm

Fire extinguisheris not

charged

Mary

Fire extinguisherdoes not

operate whenMary tries to use it

Mary

Mary pulls the plug

on the fire extinguisher

Mary

Mary seesthe fire

on the stove

Mary

Mary tries to use the fire

extinguisher

Mary

CF

How much oil is used? How

much chicken?Chicken, pan, oil

What exactly

did she see?Mary

Had it been

previously used?Inspection tag

Had it not been

originally charged?Fire

extinguisher

Had it leaked?

Fire extinguisher,floor

Does Mary know how

to use a fire extinguisher?

Mary

Is "plug" the sameas pin?

Mary

Part one

CF = Causal factor

5:00 pm

Figure 1 continued on next page


3. Root cause identification.4. Recommendation generation and implementa-

tion. Step one—data collection. The first step in the

analysis is to gather data. Without complete infor-mation and an understanding of the event, thecausal factors and root causes associated with theevent cannot be identified. The majority of timespent analyzing an event is spent in gatheringdata.

Step two—Causal factor charting. Causal factorcharting provides a structure for investigators to orga-nize and analyze the information gathered duringthe investigation and identify gaps and deficienciesin knowledge as the investigation progresses. Thecausal factor chart is simply a sequence diagramwith logic tests that describes the events leading upto an occurrence, plus the conditions surroundingthese events (see Figure 1, p. 47).

Preparation of the causal factor chart shouldbegin as soon as investigators start to collect infor-mation about the occurrence. They begin with askeleton chart that is modified as more relevantfacts are uncovered. The causal factor chart should

drive the data collection process by identifyingdata needs.

Data collection continues until the investigatorsare satisfied with the thoroughness of the chart(and hence are satisfied with the thoroughness ofthe investigation). When the entire occurrence hasbeen charted out, the investigators are in a goodposition to identify the major contributors to theincident, called causal factors. Causal factors arethose contributors (human errors and componentfailures) that, if eliminated, would have either pre-vented the occurrence or reduced its severity.

In many traditional analyses, the most visiblecausal factor is given all the attention. Rarely, how-ever, is there just one causal factor; events are usu-ally the result of a combination of contributors.When only one obvious causal factor is addressed,the list of recommendations will likely not be com-plete. Consequently, the occurrence may repeatitself because the organization did not learn all thatit could from the event.

Step three—root cause identification. After allthe causal factors have been identified, the investi-gators begin root cause identification. This step

QUALITY BASICS

Part two

Fire spreads throughout the kitchen

Kitchen, Mary

Mary throwswater onthe fire

Mary

Mary calls thefire department

Mary, FD

Fire departmentarrives

Observation

Fire departmentputs out fire

FD, observation

Kitchen destroyed

by fire

Other lossesfrom smoke andwater damage?

Time?Time?Time?CF

Fire was agrease fire

Mary, panDid she do

anything else?Mary

Was Mary trying to do this?

Mary

Did she know this was wrong? Lack of practice

fighting fires?Mary

What is Jane doing during

this time?Mary, Jane

How long did it take for the

FD to arrive?FD

dispatcher

Did the FD use the correct

techniques?FD


involves the use of a decision diagram called theRoot Cause Map (see Figure 2, p. 50) to identify theunderlying reason or reasons for each causal factor.

The map structures the reasoning process of theinvestigators by helping them answer questionsabout why particular causal factors exist oroccurred. The identification of root causes helpsthe investigator determine the reasons the eventoccurred so the problems surrounding the occur-rence can be addressed.

Step four—recommendation generation andimplementation. The next step is the generation ofrecommendations. Following identification of theroot causes for a particular causal factor, achievablerecommendations for preventing its recurrence arethen generated.

The root cause analyst is often not responsiblefor the implementation of recommendations gener-ated by the analysis. However, if the recommenda-tions are not implemented, the effort expended inperforming the analysis is wasted. In addition, theevents that triggered the analysis should be expect-ed to recur. Organizations need to ensure that rec-ommendations are tracked to completion.

Presentation of ResultsRoot cause summary tables (see Table 1, p. 52)

can organize the information compiled during dataanalysis, root cause identification and recommen-dation generation. Each column represents a majoraspect of the RCA process.

• In the first column, a general description of thecausal factor is presented along with sufficientbackground information for the reader to beable to understand the need to address thiscausal factor.

• The second column shows the Path or Pathsthrough the Root Cause Map associated withthe causal factor.

• The third column presents recommendationsto address each of the root causes identified.

Use of this three-column format aids the investi-gator in ensuring root causes and recommenda-tions are developed for each causal factor.

The end result of an RCA investigation is gener-ally an investigation report. The format of thereport is usually well defined by the administrativedocuments governing the particular reporting sys-

tem, but the completed causal factor chart andcausal factor summary tables provide most of theinformation required by most reporting systems.

Example Problem The following example is nontechnical, allowing

the reader to focus on the analysis process and notthe technical aspects of the situation. The followingnarrative is the account of the event according toMary:

It was 5 p.m. I was frying chicken. My friendJane stopped by on her way home from the doc-tor, and she was very upset. I invited her intothe living room so we could talk. After about 10minutes, the smoke detector near the kitchencame on. I ran into the kitchen and found a fireon the stove. I reached for the fire extinguisherand pulled the plug. Nothing happened. Thefire extinguisher was not charged. In despera-tion, I threw water on the fire. The fire spreadthroughout the kitchen. I called the fire depart-ment, but the kitchen was destroyed. The firedepartment arrived in time to save the rest ofthe house.

Data gathering began as soon as possible afterthe event to prevent loss or alteration of the data.The RCA team toured the area as soon as the fire

department declared it safe. Because data frompeople are the most fragile, Mary, Jane and the fire-fighters were interviewed immediately after thefire. Photographs were taken to record physicaland position data.

The analysts then developed the causal factorchart (see Figure 1, p. 47) to clearly define thesequence of events that led to the fire. The causalfactor chart begins with the event; Mary begins fry-ing chicken at 5 p.m. As the chart develops from

In many traditional analyses,the most visible causal factoris given all the attention.


QUALITY BASICS

Root Cause MapFIGURE 2

Section one1

2

Start here with each causal factor. 1

6

Equipment reliability program

problem 7

Installation/fabrication

8

Equipment misuse

2Equipment difficulty

Corrective maintenance LTA• Troubleshooting/corrective action LTA• Repair implementation LTAPreventive maintenance LTA• Frequency LTA• Scope LTA• Activity implementation LTAPredictive maintenance LTA• Detection LTA• Monitoring LTA• Troubleshooting/ corrective action LTA• Activity implementation LTA

29

32

36

30

31

33

34

35

3738

39

40

Proactive maintenance LTA• Event specification LTA• Monitoring LTA• Scope LTA• Activity implementation LTAFailure finding maintenance LTA• Frequency LTA• Scope LTA• Troubleshooting/ corrective action LTA• Repair implementationRoutine equipment rounds LTA• Frequency LTA• Scope LTA• Activity implementation LTA

41

4243

44

45

4748

49

50

5253

54

46

51

Equipment reliability program implementation

LTA 28Procedures

111

No programProgram LTA• Analysis/design procedure LTA• Inappropriate type of maintenance assigned• Risk acceptance criteria LTA• Allocation of resources LTA

22

23

24

25

26

27

Equipment reliability program design

less than adequate (LTA) 21

16Design input LTADesign output LTA 17

Design input/output

15

Equipment design records LTAEquipment operating/ maintenance history LTA

19

20

Equipmentrecords

18

Administrative/management

systems 55

Note: Node numbers correspond to matching page in Appendix A of the Root Cause Analysis Handbook.

Customer interface/ services• Customer requirements not identified• Customer needs not addressed• Implementation LTA

106

108

109

110

Document and configuration control• Change not identified• Verification of design/ field changes LTA (no PSSR*)• Documentation content not kept up to date• Control of official documents LTA

100

102

103

104

105

Procurement control• Purchasing specifications LTA• Control of changes to procurement specifications LTA• Material acceptance requirements LTA• Material inspections LTA• Contractor selection LTA

93

95

96

97

98

99

Product/material control• Handling LTA• Storage LTA• Packaging/ shipping LTA• Unauthorized material substitution• Product acceptance criteria LTA• Product inspections LTA

8587

88

89

90

91

92

Safety/hazard/ risk review • Review LTA or not performed• Recommendations not yet implemented• Risk acceptance criteria LTA• Review procedure LTA

72

74

75

76

77

Standards, policies or administrative controls (SPACs) LTA• No SPACs• Not strict enough• Confusing, contradictory or incomplete• Technical error• Responsibility for item/activity not adequately defined• Planning, scheduling or tracking of work activities LTA• Rewards/incentives LTA• Employee screening/ hiring LTA

5759

60

6162

63

64

65

66

SPACs not used• Communication of SPACs LTA• Recently changed• Enforcement LTA

67

6970

71

Problem identification control• Problem reporting LTA• Problem analysis LTA• Audits LTA• Corrective action LTA• Corrective actions not yet implemented

78

80

8182

83

84

Not used• Not available or inconvenient to obtain• Procedure difficult to use• Use not required but should be• No procedure for task

112

113

114

115

116

Misleading/confusing• Format confusing or LTA• More than one action per step• No checkoff space provided but should be• Inadequate checklist• Graphics LTA• Ambiguous or confusing instructions/ requirements• Data/computations wrong/incomplete• Insufficient or excessive references• Identification of revised steps LTA• Level of detail LTA• Difficult to identify

118

117

120

121122

123

124

125

126

127128

129

Wrong/incomplete• Typographical error• Sequence wrong• Facts wrong/ requirements not correct• Wrong revision or expired procedure revision used• Inconsistency between requirements• Incomplete/situation not covered• Overlap or gaps between procedures

130

131132

133

134

135

136

137

5

Equipment design problem

Figure 2 continued on next page


Section TwoStart here with each causal factor. 1

4Other difficulty

1

3Personal difficulty

9

Companyemployee

10

Contractemployee 11

Naturalphenomena

12

Sabotage/horseplay

13

External events

14Other

Training163

Human factorsengineering

138Communications

192

No training • Decision not to train• Training requirements not identified

164

165

166

Training recordssystem LTA • Training records incorrect• Training records not up to date

167

168

169

Training LTA • Job/task analysis LTA• Program design/ objectives LTA• Lesson content LTA • On-the-job training LTA • Qualification testing LTA • Continuing training LTA • Training resources LTA• Abnormal events/ emergency training LTA

170

171

172

174

175

176

177

178

179

Immediatesupervision

180

Preparation• No preparation• Job plan LTA• Instructions to workers LTA • Walkthrough LTA • Scheduling LTA • Worker selection/ assignment LTASupervision duringwork• Supervision LTA • Improper performance not corrected• Teamwork LTA

181182

188

183

184185

186

187

189

190191

Personalperformance

208

Problemdetection LTA*Sensory/perceptualcapabilities LTA*Reasoningcapabilities LTA*Motor/physicalcapabilities LTA*Attitude/attentionLTA*Rest/sleep LTA(fatigue)*Personal/medicationproblems

209

210

211

212

213

214

215

No communication ornot timely • Method unavailable or LTA• Communication between work groups LTA• Communication between shifts and management LTA • Communication with contractors LTA • Communication with customers LTA

194

195

196

197

198

199

Misunderstoodcommunication• Standard terminology not used• Verification/ repeat back not used• Long message

200

201

202203

Wronginstructions 204

Job turnover LTA• Communication within shifts LTA• Communication between shifts LTA

205

206

207

Workplace layout• Controls/displays LTA• Control/display integration/ arrangement LTA• Location of controls/displays LTA• Conflicting layouts• Equipment location LTA• Labeling of equipment or locations LTA

140

141

143

144

145

146

147

Work environment• Housekeeping LTA• Tools LTA• Protective clothing/ equipment LTA• Ambient conditions LTA• Other environmental stresses excessive

148149

150

151

152

154

Workload• Excessive control action requirements• Unrealistic monitoring requirements• Knowledge based decision required• Excessive calculation or data manipulation required

155

156

157

158

159

Intolerant system• Errors not detectable• Errors not correctable

160

162

161

2

© 1995, 1997, 1999, 2000 and 2001, ABSG Consulting Inc.

*Note: These nodes are for descriptive purposes only.

Shape Description

Primary difficulty source

Problem category

Root cause category

Near root cause

Root cause

*PSSR = Project scope summary report


QUALITY BASICS

Root Cause Summary TableTABLE 1

Event description: Kitchen is destroyed by fire and damaged by smoke and water. Event #: 2003-1

Description:Mary leaves the frying chicken unattended.

• Personnel difficulty.• Administrative/management systems.• Standards, policies or administrative controls (SPACs) less than adequate (LTA).• No SPACs.

• Implement a policy that hot oil is never left unattended on the stove.• Determine whether policies should be developed for other types of hazards in the facility to ensure they are not left unattended.• Modify the risk assessment process or procedure development process to address requirements for personnel attendance during process operations.

Paths Through Root Cause Map RecommendationsCausal factor # 1

Description:Electric burner element fails (shorts out).

• Equipment difficulty.• Equipment reliability program problem.• Equipment reliability program design LTA.• No program.

• Replace all burners on stove.• Develop a preventive maintenance strategy to periodically replace the burner elements. • Consider alternative methods for preparing chicken that may involve fewer hazards, such as baking the chicken or purchasing the finished product from a supplier.

Description:Fire extinguisher does not operate when Mary tries to use it.

• Equipment difficulty.• Equipment reliability program problem.• Equipment proactive maintenance LTA.• Activity implementation LTA.

• Equipment difficulty.• Equipment reliability program problem.• Administrative/management systems.• Problem identification and control LTA.

• Refill the fire extinguisher.• Inspect other fire extinguishers in the facility to ensure they are full. • Have incident reports describing the use of fire protection equipment routed to maintenance to trigger refilling of the fire extinguishers.

• Add this fire extinguisher to the audit list. • Verify that all fire extinguishers are on the quarterly fire extinguisher audit list. • Have all maintenance work requests that involve fire protection equipment routed to the safety engineer so the quarterly checklists can be modified as required.

Description:Mary throws water on fire.

• Personnel difficulty.• Company employee.• Training.• Training LTA.• Abnormal events/emergency training LTA.

• Provide practical (hands-on) training on the use of fire extinguishers. Classroom training may be insufficient to adequately learn this skill.• Review other skill based activities to ensure appropriate level of hands-on training is provided. • Review the training development process to ensure adequate guidance is provided for determining the proper training setting (for example,classroom, lab, simulator, on the job training, computer based training).

Paths Through Root Cause Map is a trademark of ABSG Consulting.




left to right, the sequences begin to unfold. The lossevents—kitchen destroyed by fire and other lossesfrom smoke and water damage—are the shadedrectangles in the causal factor chart.

Although we read the chart from left to right, itis developed from right to left (backwards).Development always starts at the end because thatis always a known fact. Logic and time tests areused to build the chart back to the beginning ofthe event. Numerous questions are usually gener-ated that identify additional necessary data.

After the causal factor chart was complete (addi-tional data were gathered to answer the questionsshown in Figure 1), the analysts identified the fac-tors that influenced the course of events. There arefour causal factors for this event (see Table 1).Elimination of these causal factors would haveeither prevented the occurrence or reduced its sever-ity. Note the recommendations in Table 1 are writtenas if Mary’s house were an industrial facility.

Notice that causal factor two may be unexpect-ed. It wasn’t overheating of the oil or splattering ofthe oil that ignited the fire. If the wrong causal fac-tor is identified, the wrong corrective actions willbe developed.

The application of the technique identified thatthe electric burner element failed by shorting out.The short melted Mary’s aluminum pan, releasingthe oil onto the hot burner, starting the fire.

The analyst must be willing to probe the datafirst to determine what happened during the occur-rence, second to describe how it happened, andthird to understand why.

BIBLIOGRAPHY

Accident/Incident Investigation Manual, second edition,DOE/SSDC 76-45/27, Department of Energy.

Events and Causal Factors Charting, DOE/SSDC 76-45/14,Department of Energy, 1985.

Ferry, Ted S., Modern Accident Investigation and Analysis, sec-ond edition, John Wiley and Sons, 1988.

Guidelines for Investigating Chemical Process Incidents,American Institute of Chemical Engineers, Center forChemical Process Safety, 1992.

Occupational Safety and Health Administration AccidentInvestigation Course, Office of Training and Education, 1993.

Root Cause Analysis Handbook, WSRC-IM-91-3, Department ofEnergy, 1991 (and earlier versions).

Root Cause Analysis Handbook: A Guide to Effective

Investigation, ABSG Consulting Inc., 1999.User’s Guide for Reactor Incident Root Cause Coding Tree, revi-

sion five, DPST-87-209, E.I. duPont de Nemours, Savan-nah River Laboratory, 1986.

JAMES J. ROONEY is a senior risk and reliability engineer

with ABSG Consulting Inc.’s Risk Consulting Division in

Knoxville, TN. He earned a master’s degree in nuclear engi-

neering from the University of Tennessee. Rooney is a Fellow

of ASQ and an ASQ certified quality auditor, quality audi-

tor-hazard analysis and critical control points, quality engi-

neer, quality improvement associate, quality manager and

reliability engineer.

LEE N. VANDEN HEUVEL is a senior risk and reliability

engineer with ABSG Consulting Inc.’s Risk Consulting

Division in Knoxville, TN. He earned a master’s degree in

nuclear engineering from the University of Wisconsin.

Vanden Heuvel co-authored the Root Cause AnalysisHandbook: A Guide to Effective Incident Investiga-tion, co-developed the RootCause Leader software and was

a co-author of the Center for Chemical Process Safety’s

Guidelines for Investigating Chemical ProcessIncidents. He develops and teaches courses on the subject.


commentPlease

If you would like to comment on this article,

please post your remarks on the Quality Progress

Discussion Board at www.asq.org, or e-mail them

to [email protected].

Documents

Root Cause Analysis