Our blogs Blogs home
Digitalization and software solutions

Digitalization and software solutions


Perfect storm for the next accident? Data analysis

This post is the second part of an article that we published in the beginning of the year – Perfect storm for the next accident?. Data analysis shows that the concerns raised at the beginning of the year are becoming reality.

The Safety Performance Indicators report from IOGP is one of the main sources for information related to safety indicators. The report using the data from 2015 was issued two months ago and presents some concerning results. Unfortunately, there is an increasingly evidence that risk management is being put aside.

The report is quite comprehensive (it has more than 140 pages containing a lot of detailed information related to Safety Indicators – it really worth reading it) but let’s try to concentrate on the most important findings.

Here they are:

The 2015 IOGP Safety Performance Indicators show that the Fatal Accident Rate for reporting companies has increased by 41% compared with 2014. The number of fatalities has increased from 45 in 2014 to 54 in 2015.

There was an increase of 41% in the number of fatalities and the following five causal factors appear consistently in the top ten for both fatal incidents and high potential events for each of the past five years.

  • Process (conditions): organisational: inadequate training/competence
  • People (acts): inattention/lack of awareness: improper decision making or lack of judgement
  • Process (conditions): organisational: inadequate work standards/ procedures
  • Process (conditions): organisational: inadequate supervision
  • Process (conditions): organisational: Inadequate hazard identification or risk assessment.

Five fatal incidents resulting in 18 fatalities were as a result of process safety events. With regard to the incident category, the largest proportion of the fatalities reported in 2015 were categorised as ‘explosions or burns’ (18 fatalities in 7 separate incidents). In 2014, only 20% of fatalities (remember that these were less compared to 2015) were in the category “explosions or burns”.

Fatal Accident Rate from API Safety Indicator Report

Fatal Accident Rate from API Safety Indicator Report

Unfortunately, the bad news does not stop there – because of current status of the oil and gas market, the report mentions a reduction of 15% in work hours. This pretty much means that the increase in fatalities actually occurred within a shorter period of time which makes the scenario even more concerning.

One of the biggest challenges our clients face when discussing the implementation of a Process Safety Management (PSM) system is “selling” the idea of safety systems and its benefits.

In a recent discussion with one of our clients, the Head of Process Safety of a LNG Plant, she mention how she “sold” the implementation of Safeti – she went to the board of directors and said “Risk analysis will save you money”.
She told me that when discussing with the board, she took a table showing all the Tier 1 and Tier 2 events that they had on the previous years.

[Quick interruption on the story here]

Tier 1, Tier 2, Tier 3 and Tier 4 is described in the American Petroleum Institute Guide to Reporting Process Safety Events – this is pretty much based on the Swiss Cheese Model (Reason, 1990) helps to compare and contrast:

  • Tier 1 events result in some level of harm (fire, LWC, release, etc.)
  • Tier 2 events result in a lesser level of harm.
  • Tier 3 and 4 indicators provide information about the strength (or lack thereof) of barriers and weaknesses in the equipment and hazard control systems.
The tier-based approach is based on the safety pyramid, where consequences increase in severity from the bottom to the top of the pyramid. Tier 3 and Tier 4 events are considered leading PSPIs, and Tier 1 and Tier 2 events are considered lagging PSPIs.

The tier-based approach is based on the safety pyramid, where consequences increase in severity from the bottom to the top of the pyramid. Tier 3 and Tier 4 events are considered leading PSPIs, and Tier 1 and Tier 2 events are considered lagging PSPIs. Tier classification if fundamental for Data analysis.

Reference material from webinar.

[Back to the story]

The event classified as Tier 1 brought their system down for 7 days. This pretty much means 7 days without production which obviously caused a direct impact to their revenue, missing cargo deliveries which impacted their image as reliable suppliers and constraining the entire supply chain of gas production for their company (they also produced and treated the gas prior exporting).

Risk management functions as a fundamental tool to prevent fatalities but if one goes about trying to explain the benefits of calculating iso-risk curves, FN-Curves, detailed fire and explosion analysis, this person will likely loose the audience pretty quickly. However, if one focus on money, this is a common language that everybody understands.

Furthermore, companies have a duty with their employees to use the latest technology to analyse the risks. This should get even more attention when compared to programmes to increase efficiency of a drilling bit or a pump. With new technologies enhancing existing models and calculations (e.g. the ability to properly account for model escalation), the decision-making process should be much easier and right decisions would avoid the trend that we are following.


0 Comments Add your comment

Reply with your comment

Your email address will not be published. Required fields are marked *