7th FRAM Workshop11-13th September – Munich

HENRI F. VON BUREN

ACCIDENT ANALYSIS INVOLVING MOBILE CRANES:THE FRAM ALTERNATIVE

What is the plan?

Using a case study of a mobile crane overturning accident for the validation of the functional resonance analysis method - FRAM for use in industrial accidents analysis and investigations.

The trigger

The scope:mobile crane overturning accidents

Mobile Cranes

Mobile cranes were involved in:

• 80 of 95 (84%) of overhead power line incidents

• 37 of 59 (63%) of crane collapses

• 35 of 59 (60%) of struck by boom/jib incidents

Types of Cranes Involved:Types of Cranes Involved:

At least 71% of all crane-related incidents involved mobile At least 71% of all crane-related incidents involved mobile cranescranes

Crane-Related Deaths in Construction by Year, 1992-2006

Source: U.S. Bureau of Labor Statistics Census of Fatal Occupational Injuries Research File

92 93 94 95 96 97 98 99 2000 01 02 03 04 05 06

92 93 94 95 96 97 98 99 2000 2001 2002 2003 2004 2005 20060

10

20

30

40

50

60

Year of incident

Causes of Crane-Related Deaths in Construction, 1992-2006

90 deaths

157 deaths

132 deaths

Other causes***

Caught in/between

Struck by cranes or crane parts

Falls**

Struck by crane booms/jibs*

Crane collapses

Struck by crane loads

Overhead power line electrocutions

0% 5% 10% 15% 20% 25% 30%

% of deaths

157 deaths

132 deaths

89 deaths

78 deaths

56 deaths

47 deaths

30 deaths

43 deaths

Total deaths: 632

* Included 64 struck by falling booms/jibs** Included 21 falls from cranes, 9 falls from crane baskets, 8 from crane loads.***Other causes included 9 highway incidents.Source: U.S. Bureau of Labor Statistics Census of Fatal Occupational Injuries Research File

Main Causes of Worker Deaths, by Frequency

• Electrocutions – from overhead power lines

• Struck by crane load

• Crane collapse

• Struck by falling boom/jib

Crane CollapsesWhy Workers Died:Why Workers Died:

1992 - 2006

Number of Collapses: 81

Number of Deaths: 89

Source: U.S. Bureau of Labor Statistics Census of Fatal Occupational Injuries Research File

51%Other/ unknown

causes

12%Overloaded

9%Crane load/ boom shifted

14%Crane cables/

rigging/ stabilizers

broke

15%Uneven/

unstable or icy surface

Dominant scenarios - Netherlands(Swuste, 2006)

• Which accident scenarios occur during mobile crane activities?

• Which scenarios are dominant?

Results, literature (Studies)

Häkkinen ea, 1978, 1993

Butler, 1978 Suruda ea., 1997; Sheperd ea., 2000

1. Load instability 1. Mechanical problem

1. Overhead wiring

2. Rigging the load 2. Crane instability 2. Load instability

3. Hoisting persons 3. Jib instability 3. Jib instability

4. Assembling and dismantlement

4. Structural problems

4. Hoisting persons

5. Crane instability 5. Crane instability

6. Accessibility 6. Assembling and dismantlement

7. Jib instability 7. Accessibility

8. Person in crane reach

Results, expert meetings (NL)

Ranking: Activity, Central event

crane instability, 1 Positioning, outriggers2 Assembling3 Misleading signals4 overloading5 Wrong plans, equipment6 Crane failure

Load instability 7 Exceeding limits8 Overloading9 Movement with hanging load10 Simultaneously hoisting11 Swinging load

Jib instability 12 Overloading13 Load hits boom14 Breaking cables

Sweep area not respected 15 Excavator, remote control

ACCIDENT MODELS

“We often propose solutions to problems

that we do not understand and then

are surprised when the solutions fail to have the

anticipated effect”.

(Leveson, 2012)

An unique model?

Different models - different consequences

Hollnagel (2010)

Accident analysis: event tree

Tombamento do guindaste

Operador não percebeu a patola

recolhida

DesSolo lavado por chuvas

nDescumprimento do

procedimento tOperador não

realizou a Inspeção

diária/LVSE o

Instabilidade do guindaste nÂngulo

pequeno com a

Falta de Treinamento

sobre conteúdo do manual

Instabilidade do guindaste

Capacidade de carga do solo excedida

Manual em Inglês e Chinês (não está disponível em

português)

Resistência Estrutural Insuficiente

Solo lavado por chuvas

Patolamento incorreto Reaterro

inadequado

Característica particular da operação (Distância do CG se

aproxima do ponto de tombamento)

Orientação do manual de operação

do guindaste não seguida

Dispositivo de seg. LMI não alarmou

Flexão da Lança Telescópica

Não tinha conhecimento do

procedimento

Ângulo pequeno com a horizontal

Capacidade de carga da operação

execida

Patola esquerda recolhida

APR não contemplava a

atividade

Operador não fez a inspeção diária

Descumprimento do procedimento

Existe procedimento, porém check list é

impreciso

Falta de treinamento no procedimento

Operador não fez inspeção visual

Abertura total da lança

telescópica

Rotação lateral da lança em 90º

Altura que o material deveria

ser elevado

Posição em que a peça deveria ser

instalada

Pode ser by-passado ou funciona sob certas

condições

Falta de comunicação na

passagem de turno

Ausência de sistemática de Gestão de mudanças

Falta de apoio para manobra

Não havia rigger no local

Não havia TST no local

Não havia supervisão no local

Foi buscar cinto de segurança

Não havia TST TRAN turno noite

Não havia Supervisor

TRAN à noite

Pouca experiência na função

Everyday operations: lifting of loads with the use of mobile cranes

Step 1: description of the functions needed for a lifting of loads with a crane

Analyse de profession : opérateur / opératrice d’appareils de levage - 2001, 2006 and 2009 + INTERVIEWS with the operators

Origin of the essential functions of this system

Functions

T

I O

C

P R

Posicionando

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Releasing

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Locking

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Position-ing

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Planning

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Engineering

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Discharging

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Disconnect

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Leveling

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Checkingequip.

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Preparingthe lift

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

CreatingPHA

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

RiggingPlan

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Activating

Tempo Controle

Precondição

Input Output

RecursoT

I O

C

P R

LMIsteps

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Turningon

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

ApplyingSMS

proced.

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Basicops.

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Signs

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Supervision

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Stabilization

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Movingload

Tempo Controle

Precondição

Input Output

Recurso

FRAM model

Step 2: Characterizing Potential and Actual Variabilities – Simple solution)

Characterizing the Performance Variability – Elaborate solution

Step 3: the Aggregation of Variability.Upstream and Downstream Couplings

Model Instantiation

Instantiation of the model: the overturning of the mobile crane

Panel and cables to be liftedLocation of

the crane

The status of the outriggers

LMI interface device(load moment indicator)

INDICATOR OF EXTENDED OUTRIGGERLAMP INDICATING THE HALF OPENING

OF THE OUTRIGGER

First instantiation

T

I O

C

P R

Verificandoequip.

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Preparandoiçamento

Tempo Controle

Precondição

Input Output

Recurso T

I O

C

P R

ElaborandoAPR

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

PlanoRigging

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Disponibi-lizando

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

OperaçõesLMI

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Dando apartida

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Aplicandoproced.

SMS

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Abrirretrairlança

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Levantarabaixarlança

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Girar

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Sinalização

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Supervisão

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Patolar

Tempo Controle

Precondição

Input Output

Recurso

Input

T

I O

C

P R

Movimentarcarga

Tempo Controle

Precondição

Input Output

Recurso

Signs and supervision

Interconnections between the functions

Second instantiation

T

I O

C

P R

OperaçõesLMI

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Dando apartida

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Aplicandoproced.

SMS

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Operaçõeselementares

Tempo Controle

Precondição

Input Output

Recurso

Output

T

I O

C

P R

Patolar

Tempo Controle

Precondição

Input Output

Recurso

Input

T

I O

C

P R

Movimentarcarga

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Sinalizando

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Supervisão

Tempo Controle

Precondição

Input Output

Recurso

T

I O

C

P R

Comunicação

Tempo Controle

Precondição

Input Output

Recurso

Step 4: Sensors to monitor the stability of the crane

Device to monitor the actual state of opening of the outriggers

Critical points

If the display showing the outriggers extension was crucial , the signal/noise relation could not be so lower that it avoid its detection. In fact, the operator could have ignored the stabilizing variable to be concentrated in others (cognitive tunnel vision).

The crane´s on-board automation system creates an unforgiven environment, which does not tolerate a mistake made by the operator. Not sute that introducing more tech will help.

The economic/financial situation of the subcontractor (owner of the equipments) wasn´t approached on the research. New data recently obtained indicates problems to adapt to the new Construction boom reality; lack of operators to new big cranes, lack of training to newcomwers/small machines operators and unreliable SMS (documents in chinese?).

Conclusion: FRAM, can we use it?

The experts seems to be satisfied with the answers obtained with the traditional methods (RESISTANCE).

FRAM requires the existence of a group of specialists with experience in the subject (cranes and liftings).

Mostly, it is important to have a good background on Human Factors, on Complex Sociotechnical Systems and of course on the technologies to evaluate the Potential and Actual Variability of the function involved and about Risk evaluation in general.

Nevertheless, for complex sociotechnical systems it constitutes a very useful Method. Don´t use it if you want to analyze minor accidents that are common in the Construction sites.

Future developmentsThe use of FRAM for Risk Analysis in the Electromechanical Construction and Assembly sector.

We can thinking ahead also using Fuzzy Logic to the quantification of variability descriptions, since the variability is described verbally (as a quality).

“It is entirely possible to introduce fuzzy reather than crisp descriptors, and this may be a step in the direction toward quantification”.

(Erik Hollnagel, 2012, p.94)

“When the only tool you have is a hammer, everything begins to look like a nail”.

(Lotfi Zadeh)