Engineering Failure Analysis 54 (2015) 1–12

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier .com/locate /engfai lanal

An investigation into the cause of loss of containment from thesupply of mini-bulk lubricants

http://dx.doi.org/10.1016/j.engfailanal.2015.04.0051350-6307/� 2015 Elsevier Ltd. All rights reserved.

E-mail address: turlough.guerin@hotmail.com

Turlough Guerinc/o The Climate Alliance Ltd, 5 Retreat Crescent, Sunbury, Victoria 3429, Australia

a r t i c l e i n f o a b s t r a c t

Article history:Received 16 January 2015Received in revised form 6 April 2015Accepted 7 April 2015Available online 13 April 2015

Keywords:Root causeSpillProcessChange managementForklift

An Intermediate Bulk Container (IBC) was punctured during its handling, releasing oil ontosoil at an environmentally-sensitive region of Australia. The telehandler did not pierce theplastic of the IBC directly (as was expected) but rather one of the tynes had caught on theunderside of the metal base plate, despite numerous controls being in place at time of spill,revealing a previously unreported mechanism for a fluid spill from handling of petroleumhydrocarbons. The diverse investigation team used a root cause analysis (RCA) techniqueto identify the underlying cause: the inspection process was inadequate with contributingfactors of not using a spotter and design of IBC did not anticipate conditions. Engineeringcontrols were put in place as part of the change management process to help prevent spillsfrom occurring from piercing from telehandler tynes on the current project site.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Leaks and spills of petroleum hydrocarbon are a major concern in the upstream oil industry, from both a construction [1]and operational perspective [2–4]. The petroleum industry is greatly concerned about safety and is one which has a strongsafety culture with product spills being no exception. Construction in remote locations requires a flexible and yet securelogistics system for delivery of fuel, oil and chemicals. Intermediate Bulk Containers (IBC) units are used to hold various typesof liquids including oils, acids and concrete accelerants (Table 1). IBCs are ideally suited for such applications because of theirflexibility for handling and scalability as the construction work front changes.

Despite the industry’s best endeavors, loss of containment may occur and we need to understand the root causes, con-sequences and implications of such events. One of the tools commonly used to investigate these losses of containment, rootcause analysis (or RCA), will lead investigators to take both short-term immediate corrective actions, as well as finding theunderlying root causes (latent failures) hidden in the way work is done that will help avert similar incidents or spills in thefuture [5].

The project’s hazard identification process (or HAZID) did not identify handling of IBCs as posing a risk to the project, butrather accepted the risk to deploy these as an improved approach to other options. Given there is an underlying requirementin all professionally-managed construction projects to ensure the design stage of the project identifies and considers thepotential risks [6–10] including those from transport and storage of chemicals, the current resource construction projectdecided to use IBCs as an enhanced and preferred method to procuring 205 L (or 44 gal) drums strapped to pallets, or to pur-chase vessels larger than IBCs, such as ISO containers or other large transportable tanks.

Table 1Examples of chemicals commonly transported to and stored atresource construction and mining sites.

Engine lubricants Greases Brake fluidGear lubricants Detergents SealantsCoolants Solvents AcetoneSodium hydroxide Hydrochloric acid Bleach

2 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

2. Purpose

The current study used RCA methodology, in a practical context, to investigate the cause(s) and contributing factors thatled to an oil spill from a commonly used bulk handling container or Intermediate Bulk Container (IBC) on a large constructionproject in a remote and environmentally-sensitive area.

3. Method

3.1. Description of IBCs

An Intermediate Bulk Container (IBC) or IBC Tote or Pallet Tank, was the type of vessel from which loss of containmentoccurred in the spill event in the current study. It is a single-use container designed for the transport and storage of bulkliquid and granulate substances (e.g. oil, chemicals, food ingredients, solvents, pharmaceuticals, etc.).

3.2. Description of operation and location

The contractor company that was operating the IBC handling equipment when the spill occurred was the primary earth-works contractor engaged to supply services to the oil and gas company that had land tenure on the island on which theresources construction project was being built. The contractor organization was operating up to 370 plant items includingforklifts and tele-handlers used for handling IBCs. The site was an LNG construction site located offshore Western Australiaon a remote island classified as a Class A Nature Reserve. The site where the spill occurred was located approximately 2 kmfrom the ocean in the center of the construction works where the LNG plant was being built. The topography where the spilloccurred was flat and well trafficked by personnel on the project.

3.3. Management systems descriptions

The contractor operated under a management regime that was comprised of an integrated health, safety, quality andenvironmental management system. Each component of the system was certified to the relevant ISO standard includingISO 9001 and 14001. To ensure alignment of the contractor’s system with that of the oil and gas operator, this integratedsystem was audited externally by the operator every 6 months.

3.4. Root cause analysis methodology

The team conducted the investigation for this incident in accordance with a fishbone, cause and effect (or Ishikawa) rootcause analysis (RCA) process [11]. The steps of this process were: 1. Formation of the investigation team; 2. Collection ofincident data including machine handler (operator’s) background; 3. Development of the sequence of events; 4.Undertake a protective systems analysis; 5. Perform RCA analysis using the cause and effect (fishbone) model (method)as a team exercise linking verification data to each decision made by the team during analysis; and 6. Development of appro-priate corrective actions. Table 2 lists the proforma options for possible root causes that were used in the RCA process. AnRCA investigation team was assembled with a range of complementary skills. These were the Company ConstructionDirector, The Company Area Construction Manager, 2x Company Environmental Coordinators, Contractor EnvironmentalEngineer, Company Environmental Superintendent, The Operator, The Health and Safety Representative (for theOperator), The Contractor HSE Manager, The Contractor Project Director, The Contractor’s National Construction GeneralManager, and the Company Incident Investigation Manager.

3.5. Collection of incident data

Data collection included one-on-one interviews, review of project and procedural documentation, employee trainingrecords, photographs from the incidents, and procurement manifests and related documents. It also included the goodsmanifests, discussion with suppliers of the IBC, and licenses.

Table 2Proforma root causes of fluid spills used in investigations on LNG construction site.

Primary root cause Root cause descriptiona

Procedures and safe work practices Accepted to deviate from work routineLack of job oversightMistake or mental slipNone exists or availableNot complete or accurateNot enforced, audited, or inspectedNot trained on procedureOther priorities conflictedRisk of not following not understoodWillful deviation

Design Design standards inadequate or not usedDid not anticipate the conditionsDid not consider human factorsInadequate reviewInherent safety design not incorporated

Inspection and quality control No inspectionQuality control needs improvementHold point not performedInspection not requiredNo hold pointForeign material exclusion during work needs improvementInspection instructions needs improvementInspection technique needs improvement

Training and competency No trainingUnderstanding needs improvementDecided not to trainMissed required trainingNo learning objectiveTask not analysedContinuing training needs improvementInstruction needs improvementLearning objective needs improvementLesson plan needs improvementPractice/repetition needs improvementTesting needs improvement

Misunderstood verbal communication Long messageNoisy environmentRepeat back not usedStandard terminology needs improvementStandard terminology not used

Supervision PreparationSelection of workerSupervision during workFall protection needs improvementLock out/tag out needs improvementNo preparationPre-job briefing needs improvementScheduling needs improvementWalk-through needs improvementWork package/permit needs improvementFatiguedNot qualifiedSubstance abuseTeam selection needs improvementsUpsetInadequate job hazard/safety analysis

Risk management Inadequate process hazard analysisIndividual snap decision (quick decision made without assessing the risk)

Preventative maintenance/repeat failure Equipment parts defectivePreventative/predictive maintenance/not preventative maintenance for equipment

No communication or not timely Preventative/predictive maintenance/preventative maintenance for equipment needs improvementCommunication system needs improvementLate communication

Turnover needs improvement No standard turnover process

(continued on next page)

T. Guerin / Engineering Failure Analysis 54 (2015) 1–12 3

Table 2 (continued)

Primary root cause Root cause descriptiona

Turnover process needs improvementTurnover process not usedTurnover less than adequate

a Notes: Additional primary root causes (without any column entry for Root Cause Description): contractor safety, communications, human factors,management of change, incident and near miss investigation, emergency response, natural phenomenon, auditing, leadership accountability, and pre startup safety review.

4 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

3.6. Identification of protective systems

Protective systems are defined as software, hardware or management systems which reduce the potential for having theincident or the consequences of the incident. These include job safety procedures and hazard identification documentation.The most commonly used procedures and documentation on Australian construction sites are safe work procedures (SWPs),job hazard analyses (JHAs), and Step Back 5 � 5 s, i.e., a quick pre-job analysis.

3.7. Communication processes

After project spill events, that are considered large by the project’s standards i.e. >1000 L, similar to the one described inthe current study, the contractor and the operator companies prepared project-wide communications that are distributed toall project personnel.

4. Results and discussion

4.1. Background and site description

IBCs (1000 L) have been used on the project since project commencement to transport various types of bulk fluids, as wellas for temporary storage on the project site. The chain of custody with IBCs commences from point of manufacturer to thesupplier(s), continuing to contractor and company base supply chains (in accordance with quarantine requirements) prior toarriving on the project. The construction site was located on a small island located approximately 60 km off the north-westcoast of Western Australia, Australia. Because of its high conservation value, the island was declared a public reserve for floraand fauna and has been classified as a ‘Class A’ Nature Reserve for the past 100 years. Using process chemicals, such as pet-roleum hydrocarbons, on the project in such a sensitive environment is a high risk activity in relation to potential environ-mental harm in the event of an uncontrolled release and a consequence it was critical that a concerted effort be made by theproject to minimize chemical spills.

4.2. Employee’s background

The investigation revealed that the telehandler operator (involved in the spill event), had over 10 years’ experience inforklift operation, held the relevant high-risk license, including successful completion of the verification of competency(VOC) and on site challenge test training. The operator commenced working for the contractor on the project inNovember 2010, almost 2 years prior to the spill. The operator conducted various tasks in the course of employment, andcarried out this specific task of handling IBCs for the workshop on previous occasions.

4.3. Effectiveness of the protective systems

The investigation team analysed the protective systems relevant to this incident and the resulting evaluation is summar-ized in Table 3. This analysis showed that there were no inspection requirements in any of the project documentation in rela-tion to IBCs or related items in transport or for storage, particularly inspections that ensured the identification of vesselintegrity. Of the 25 separate protective systems, processes and controls in place at the time of the spill, it is noteworthy thatonly 5 were categorized as being higher up the safety hierarchy than ‘‘Administrative’’.

4.4. Sequence of events and outcomes from the incident

A timeline was constructed to summarize activities before, during and after the event. These details are presented inTable 4. The operator commenced the lifting activity by placing the tynes partially into the pockets of the IBC and then exitedthe machine to check the alignment/position of the fork tynes before extending the telescopic boom. The operator thenreturned to the telehandler and commenced extending the telescopic boom into the IBC pocket (i.e. pallet base/belly) toengage the lift. The operator commenced extending the boom and then heard a ‘‘popping’’ noise and observed the IBC

Table 3Evaluation of the protective systems and controls relevant to the spill incident.

Protective system Type of control based onhierarchy

In place?(Y/N)

Effective?(a)

(Y/N)Comments

IBC – selection and assessmentof type, design and structuralintegrity (IBC risks notidentified in HAZID)(b)

Elimination/substitution N N Other IBC Units (from another areas, which are currently orhad been in use prior to incident on the Project) wereidentified as damaged during the investigation

Self-bunded sea container Isolation/engineering N N These containers are not usually used for transport. Thesecontainers are typically used for chemical storage on theproject

Spotter Isolation N N JHA referenced use of spotter if required (when handling IBCswith tyned equipment). Operator assessed spotter was notrequiredNOTE: Use of spotter may have influenced the outcome

Work method statement (WMS) Administrative N N Not developed for this particular taskPlant acceptance HSE checklist Administrative N N No document was available for the investigationSWP Administrative Y N SWP for Forklift Operations is not specific on when a spotter

is required. SWP did not identify potential hazard ofobstructions on underside of IBC

JHA Administrative Y N Operator signed onto JHAJHA did not specify when a spotter is requiredJHA did not identify potential hazard of obstructions onunderside of IBC

Communication of similarincidents

Administrative Y N Operator not aware of previous related incidents associatedwith IBC holding an acid (Hydrochloric Acid spill of similarmagnitude, Incident on project site on 19/3/2011)

IBC inspection – prior to lift Administrative Y N General area around IBC inspected, inspection did not includeunderside of IBC and metal plate

Sea container stacking/filling Isolation Y Y Correct use of strapping‘‘Step Back 5 � 5’’ (i.e. pre-work

risk analysis conducted by allemployees on project site)

Administrative Y Y Four step back 5 � 5 s were completed by the operatorthroughout the day of the incident

Telehandler pre-start check Administrative Y Y Pre-start check was conducted, no issues identified withtelehandler

Inspection of area (seacontainer) prior to lift

Administrative Y Y Inspection took place, did not include underside of IBC,according to the JHA there was no requirement to do so

Supervision Administrative Y Y Supervisor on call and involved in step back 5 � 5 s;Supervisor not required to be present for each lift

Communication Administrative Y Y Communication between operator and workshopsuperintendent in planning the move of the IBCs and theincident response

Training – JHA, hazardidentification

Administrative Y Y Personnel developing JHAs receive feedback from HSEAdvisors

Training – spill response Administrative Y Y Occurs on a 6 monthly basis for all site personnelRandom drug and alcohol testing Administrative Y Y Process in place and effectiveControl of spill Administrative/

protective equipmentY Y Swift and effective control

Containment Administrative/protective equipment

Y Y Earthen bund quickly constructed

Clean up Administrative/protective equipment

Y Y Clean up required inspection after soil excavation

(a) The investigation team made an assessment as to whether the control was effective.(b) HAZID is a hazard identification process involving a cross section of stakeholders identifying potential hazards prior to project initiation.

T. Guerin / Engineering Failure Analysis 54 (2015) 1–12 5

collapse immediately. Of critical importance in the incident investigation was the finding that the fork tynes had not piercedthe plastic bladder of the IBC directly.

4.5. Outputs from the root cause analysis

From the close inspection of the IBC base, deformed metal had caught on the IBC bladder to cause the spillage (Fig. 1). Theinvestigation also revealed that the tynes of the telehandler had sufficient clearance to enter the base of the impacted IBCunder normal conditions where no such metal deformation is expected (Figs. 2–4). The primary root cause of the spillwas determined to be ‘‘Inspection/Quality Control – Inspection and Acceptance Process is Not in Place or Adequate’’(Fig. 5) and verification for this decision (Table 5). There were also two contributing factors to the cause of the incident:(1) The design of IBC cage did not anticipate conditions on project. The IBC is designed to be moved a limited number oftimes, and (2) The use of a spotter may have prevented the outcome as the spotter may have been able to identify the defor-mity in the pallet base plate. The rigor of the inspection required to have picked up such a deformity in the base plate of the

Table 4Timeline of events related to the spill on day of event.

Time Description of events

15:00 Operator completes step back 5 � 5 at workshop, for the unloading of the sea container15:00–15:15 Operator contacted workshop office to ascertain where pallets were to be positioned once removed15:15 Operator moved two pallets of spare parts from the sea container in front of IBC, and placed in nominated area workshop15:15–15:25 Operator released straps securing IBCs within sea container

Operator positioned tines of telehandler into pockets of IBC (approx. 20–100 mm in pocket)Operator exited telehandler to check positioning of tinesOperator re-entered telehandler to commence extending boom and tynes into IBC pocket

15:25 Operator heard a ‘‘popping’’ sound, saw IBC collapse quickly and witnessed oil spilling onto the ground15:25–15:30 Operator exited telehandler and immediately notified workshop superintendent of spill

Workshop superintendent and operator immediately returned to location with spill kits to commence control and containmentIBC repositioned by operator using the telehandler as instructed from workshop superintendent to eliminate any further leakingof oil from IBC

15:30 Workshop superintendent notified contractor’s Environmental Engineer of spill15:30–15:40 Loader available in the area commences construction of an earthen bund to contain spill

Fig. 1. (Top left) Timber and tynes of the telehandler used to hold punctured IBC in tilted position, (top right) underside of IBC showing impact of damagedmetal (belly) plate and location in which the tynes are inserted, (bottom left) underside of IBC, (bottom right) undamaged IBC behind damaged IBC (in seacontainer) for comparison.

6 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

pallet would have required the use of a flashlight and the inspector would have had to have leaned down and looked into thepallet slot (and known what to have looked for).

4.6. Comparing the RCA method with the 5 Why analysis

Why the RCA method (using the fishbone, cause and effect, or Ishikawa diagram) is referred to in the literature (for largespills) is because the underlying causes of an incident such as a spill can be determined with a high degree of confidence.

A shortened version of the RCA is called the ‘‘5 Why’’ method. This method involves asking why a particular eventoccurred, in the series of events that led to an incident, up to 5 (or even more) times [12,13]. In simple terms, once askingthe ‘‘Why’’ question yields no further reasons for the cause(s) of an incident, then the root cause is determined andcommunicated. Whereas, on many commercial construction and resource projects, the shorter and less involved processof the ‘‘5 Why’’ method, is commonly used instead. The advantage of the 5 Why method is it can be done using limitedresources (usually by an individual or small team) but it relies on professional judgment (or practical experience) to ensure

Fig. 2. (Top left) Punctured IBC in location at time of spill, (Top right) tynes of telehandler involved in the spill, (Bottom left and right) dimensions of tynesof telehandler indicating that there was sufficient clearance for insertion of tynes into IBC base without any damage expected. Distal end on bottom left andload end on bottom right.

belly plate shown 5 mm below top bar of IBC pallet base

1200 mm

95 mm

20 mm

20 mm

5 mm

395 mm 135 mm140mm

100 mm

Fig. 3. A side view of the base of the IBC i.e. the pallet showing dimensions relevant to the telehandler interaction.

IBC placement (involved in Spill)

Front of Sea container 1200 mm

800

mm

Container wall

Distance of IBC from container (inner wall) approx. 50 mm

1100 mm(width of workshop

pallet in front of impacted

IBC)

4700mm

2300

mm

Fig. 4. Location of IBC relative to rest of the sea container contents and the telehandler in the delivered container.

T. Guerin / Engineering Failure Analysis 54 (2015) 1–12 7

Contributing Factor: Design of IBC cage did not anticipate project

conditions

956L Oil spill to ground

Oil in IBC Verification: hot

shot container had just arrived; spilt oil

Hole in IBC Verification:

Photographs, spill report, witness statement, site

inspection

Force to allow oil out of IBC

Verification: Investigation, photographs

Spill contained Verification: site

inspection, photographs

Proper Condition:

Common for oil to be stored in

IBC in sea container

Plastic of IBC punctured by metal

plate Verification:

inspection of IBC, team consensus

Proper Condition:

IBC was full of oil

Temporary earthen bund constructed to

contain spilled oil Verification: site

inspection, witness statements

IBC position tilted by operator with telehandler to

prevent further oil spilling

Verification: site inspection, witness

statements

Force applied to metal plate Verification:

inspection of IBC, investigation

Forklift tynes contacted metal plate on

underside of IBC Verification: inspection of IBC, investigation,

photograph

Direction from workshop superintendent Verification:

Witness statements

Immediate notification of spill by Operator to workshop superintendent Verification: witness statements

Proper Condition: Swift effective spill response by superintendent and

operator

Fault with IBC causing metal plate on underside to distort and hang lower than usual (possible integral design

limitation) Verification: Investigation, other IBCs

inspected onsite were damaged, witness statement of operator

Repeated handling of IBC in supply chain

causes damage Verification: Other

IBCs witnesses with damage

(photographs)

Fault not noticed prior to lift

Verification: Witness

statements, investigation

Integrity and condition of IBC base/pockets

not formally inspected prior to transport to

project site. Verification:

Photographs from Contractor Logistics

Underside of IBC not inspected by Operator

prior to lift Verification: Witness

statement

Operator exited machine to check

alignment/positioning of fork tynes

Verification: Self spotted, operator witness statement

Spotter not used Verification:

Witness statement

Force/resistance not felt by operator. Machine telescopic boom is operated

by hydraulics Verification: Witness statement, SME

agreement

Proper Condition: Hydraulic

telehandler would not allow operator

to perceive resistance

Root Cause: Inspection/Quality

Control – Inspection and acceptance

process is not in place or not adequate

No requirement in JHA or procedures to check, not common

practice Verification: JHA,

SWP

Faults/snags on underside of IBC not

recognized as a hazard

Verification: JHA, SWP

Proper Condition: Operator checked

positioning alignment of tynes

prior to lift

Proper Condition: Design permits fork tynes to be

entered into pockets for lifting

IBC

No Mandatory requirement to use

spotter Verification: JHA, SWP & Company H&S Procedure Forklift Trucks & Powered Pallet

Proper Condition: Procedures

were followed

Contributing Factor: Use of Spotter may

have prevented outcome

Fig. 5. The results of the RCA Why Tree for the oil spill incident.

8 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

Table 5Data collected to verify the cause of the oil spill incident.

Data description Comments

Authority to operate/inspect/maintain for operator for the telehandler current Discussed by investigation teamOperator’s license to perform high risk work was current Discussed by investigation teamStep back 5 � 5 (several that day) and VOC for telehandler Discussed by investigation teamOEM’s health and safety procedure ‘‘Manual Forklift Trucks and Powered Pallet

Movers’’Note

Pre-start on telehandler machine Discussed by investigation teamContractor’s SWP for Forklift Operations NoteJHA for telehandler Signed onto by operator on day of spillInspection of pierced IBC Workshop superintendent stated that new IBCs are requested from

the supplierWitness statement of telehandler operator Formal statement obtained.Supervisor, workshop superintendent Formal statement obtained.Additional photographs taken of incident area and other used IBC Units Note – refer to figuresMultimodal Dangerous Goods Form, completed for the sea container from

previous day, supply baseConfirms that ‘‘the goods have been packed/loaded into seacontainer in accordance with the applicable provisions’’

Photographs taken of IBC immediately after it was loaded into the sea containerat supply base prior to shipment to quarantine checking

Shows that IBC was handled at supply base as part of quarantinerequirements

Table 6A qualitative comparison of the Why tree (RCA) and ‘‘5-Why’’ process for determining root causes.

RCA 5-Why

Detailed, time consuming High-level, rapidRequires extensive training (several days) Short course (1–2 h) requiredThorough, extensive collaboration in decision making, involving a broad

range of experts from diverse backgroundsMinimum of consultation/collaboration to decide on the causes,designed to be conducted by individuals

Requires significant level of resources (requires a team of at least 5–6) Can be done using limited resources (can be done by an individual orsmall team)

Provides a robust analysis of the root causes and contributing factors Relies on professional judgment to ensure correct line of questioningResults can be relied upon to change business process or systems Limited reliance can put on the results of a single analysisIs applied in commercial contexts where a significant injury, death or major

environmental incident has occurredIs applied in commercial contexts as a root cause analysis where anincident investigation is required

Table 7An estimate of financial costs of the spilla.

Cost element Description Value ($AUD)

Loss of product 1000 L of high quality lubricant 5000Time commitment of environmental

engineer60 h of time committed to remediation, investigation and close out 9600

Soil clean-up Operation of a wheel loader (I�2 h) and supervision/disposal of 10 t contaminatedsoil

11,000

Root cause analysis Preparation, delivery and review 3000Interviews This involved meeting with all stakeholders involved 640Engagement of suppliers Discussions regarding IBC and its history 240Reporting Preparation of entire investigation report and presentation to operator 6500Administration Collection of all required documentation, communication of findings 2000

TOTAL 37,980

a Notes: Conservative estimates based on pay rates of approximately $AUD 160 per hour for an engineer.

T. Guerin / Engineering Failure Analysis 54 (2015) 1–12 9

the line of questioning is appropriate for the event under investigation. A qualitative comparison of these two types of rootcause assessments is given in Table 6. In the majority of cases this depth of analysis is fit-for-purpose, enabling the businessto learn quickly and keep going, business as usual, plus locking in the required prevention strategies. When the 5 Whymethod is used to determine underlying causes, this means that full analysis and understanding of the root causes and con-tributing factors are not always thoroughly determined and therefore their prevention may not be reduced (as a result of the5 Why analysis) when business (process, system) changes are made as a result of the findings. As a consequence, the actualroot causes are not necessarily properly identified and therefore communicated as widely as would be desirable to helpreduce fluid spills.

Fig. 6. (Top left) spilt oil migrated out of un-bunded sea container (Top right) floor of container showing absence of internal catchment (Bottom left) rear ofcontainer and (Bottom right) side of container showing earthen bund erected to limit flow of oil.

10 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

4.7. Costs

An estimate was compiled for the breakdown of the costs of the current spill, and the subsequent investigation (Table 7).The clean-up of the contaminated soil and commitment of time by the contractor’s Environmental Engineer were the singlelargest cost items from the spill incident (Fig. 6).

4.8. Development of corrective actions

The hierarchy of controls was applied to the corrective actions and the key actions are listed below:

1. Design, construction and use of a protection plate to prevent fork tynes puncturing IBC Units as they are handled, particu-larly on major projects where there are numerous points of handling of these vessels (Figs. 7 and 8).

2. Undertake chain of custody inspections of IBCs containing lubricants and other chemicals, to identify and reject any dam-aged IBCs with metal base frame deformities from supplier.

While these recommendations are specific to the project site of the current study, the implications for other sites and pro-jects are broader given the popularity of the use of the 1000 L IBC for supplying and distribution of lubricants, chemicals andother hazardous materials across resource construction projects.

5. Conclusions

In handling IBCs, it is necessary to check and/or reject IBC units for damage during chain of custody paying particularattention to the underside/metal base integrity. A cursory examination of the costs arising from the spill suggest that suchevents are likely to have a material impact on the costs for the project. Insights and innovation in this project were from theapplication of rigor and diverse thinking and viewpoints to the investigation even though it was a relatively minor spill inrelation to those reported in the industry. By conducting the RCA, the unusual root cause was identified (i.e. the interaction ofa forked tyne with deformed pallet base plate), which would have most likely been overlooked unless a high level of rigorwas applied to the investigation. An opportunity for further research would be to communicate the findings from the currentstudy with vendors of the IBCs assessed in the study to determine the merit of integrating design changes to alleviate spillsfrom such equipment in the future.

Fig. 7. (Top left) Engineering control i.e. IBC cage, put in place as a result of the spill investigation, (Top right) engineered base plate manufactured on site tominimize impacts of tynes on IBCs (Middle left) other IBCs identified on the project site with damage (Middle right) after the investigation showing minordamage to base (Bottom left) other IBCs with extensive damage (Bottom right) damage to the side of another IBC.

Fig. 8. Design of plywood protection board for handling IBCs. Notes: (1) Height of the fork lift tyne slots not to exceed the height of the IBC frame/skid slots.(2) Cut outs for hand holds, (3) tyne spacing thickness 400–500 mm.

T. Guerin / Engineering Failure Analysis 54 (2015) 1–12 11

12 T. Guerin / Engineering Failure Analysis 54 (2015) 1–12

References

[1] Guerin TF. Root causes of fluid spills from earthmoving plant and equipment: implications for reducing environmental and safety impacts. Eng FailAnal 2014;45:128–41.

[2] Ismail Z, Karim R. Some technical aspects of spills in the transportation of petroleum materials by tankers. Saf Sci 2013;51:202–8.[3] Sánchez-Arias LE, Remolina DA, Alvarez-León R. Evaluation of a recovery technique for mangrove soils affected by oil spills, using as indicator plantules

of Rhizophora mangle L. (Rhizophoraceae). Pan-Am J Aquat Sci 2013;8:79–88.[4] Stevenson K. Care and concern for people we’ll never meet – process safety in the design of the Wheatstone platform, Perth, WA; 2012. p. 942–7.[5] Otutu F, Agba SA. Experience with tripod beta methodology applied to serious environmental incident investigation. In: Proceedings – 13th middle east

oil and gas show and conference, technology solution for the oil and gas business: new middle east focus, MEOS 2003, June 9, 2003 – June 12, 2003.Society of Petroleum Engineers (SPE), Bahrain, Bahrain; 2003. p. 1023–30.

[6] Behm M, Gambates J, Toole MT. Construction safety and health through design. In: Hill DC, editor. Construction safety management andengineering. Des Plaines, Illinois: American Society of Safety Engineers; 2014. p. 103–26.

[7] Fortunato III BR, Hallowell MR, Behm M, Dewlaney K. Identification of safety risks for high-performance sustainable construction projects. JConstruction Eng Manage 2011;138:499–508.

[8] Behm M, Culvenor J. Safe design in construction: perceptions of engineers in Western Australia. J Health Saf Res Practice 2011;3:9–23.[9] Gambatese JA, Behm M, Rajendran S. Design’s role in construction accident causality and prevention: perspectives from an expert panel. Saf Sci

2008;46:675–91.[10] Behm M, Veltri A, Kleinsorge IK. The cost of safety. Professional Saf 2004;49:22–9.[11] Ishikawa K. Introduction to quality control. Taylor and Francis; 1990.[12] Hendershot D. Tell me Why. J Hazard Mater 2007;115:39–49.[13] Pojasek RB. Asking ‘‘Why’’ five times. Environ Qual Manage, vol. 10 (Autumn); 2000. p. 79–84.