This analysed 21 fatal occupational accidents in oil & gas relating to confined spaces using the HFACS (Human Factors Analysis & Classification System). They also conducted 33 semi-structured interviews to explore their experiences with the contributory factors identified with HFACS.
One motive of this study is that accident analysis tools & practitioners may not adequately evaluate the multiple & complex factors of accidents, including org factors, job design & engineering systems.
Results:
Of the accidents time period assessed, there was something like 36+ fatalities. H2S & N2 poisoning, and O2 deficiency inside the CS accounted for 70% of fatalities. Explosion was 12%, H2S poisoning outside of the CS at 11%, & electrocution at 5.5%. Similar stats for non-fatal accidents were evident except for H2S poisoning outside of the CS, which accounted for 36% of injuries.
For contributing influences – although this study started with the usual “unsafe acts” etc. (which, interestingly, are rarely justified by the investigator on how they came to this conclusion), the study associated the largest proportion of influencing factors at the organisational level – 44% of factors. This related to processes, training systems, ineffective risk management, & ineffective planning for inspection & maintenance. Lack of effective communication channels was also highlighted.
Further, ineffective organisational management of rescue equipment & planning, gas detectors and staff shortages for rescue and CS entries was identified as organisational influencing factors.
For resources, lack of rescue equipment, gas detectors & rescue teams & other staff shortages were identified. For PPE, lack of lifelines or low-quality PPE like loose-fitting respirators and high workload led to shortages in skilled workers & poor time allocation.
Issues around workers and contractors feeling unable to refuse unsafe conditions was a noted factor during interviews, and economic downturns affecting the “bargaining power” of contractors with principal contractors. Lack of monitoring or oversight on contractor performance during projects was seen as a factor leading to cost cutting with contractor resourcing. In one example with contractors where some personnel hold multiple supervisory positions at the same time.
Multiple issues with the permit to work system (PTW) were also identified during interviews. Usual issues like lack of procedures, lack of permit offices or recognisable permit issuers and lack of monitoring of the process. The latter, in my opinion, speaks to a common problem in safety where we implement systems but then put little effort into validating their impacts or effectiveness; nor have few ways or feedback loops to monitor their effectiveness.
Another area of concern was the frequent inaccessibility of isolating energy within CS, lack of safe access & egress means, improper designs, and similar couplings used between supplied air respirators & inert gas systems (potentially leading to asphyxiation).
One thing I like about this study is that it doesn’t try to attribute everything to worker behaviour, but one limitation as I see it is that we know about factors involved in these fatal accidents, but not the factors with successful work. Moreover, are the issues identified here also present most of the time with normal work?
Given that fatalities are, thankfully, rare, we’re only learning from a tiny (but critical) subset of work here.
SafetyInsights.org 