Process safety targets: planning to fail

Corporations are familiar with the concept of setting objectives which are SMART (Specific, Measurable, Achievable, Realistic, and Timely) and go forth and set goals across the companies’ areas of operations. For process safety, these targets are referred to as process safety management (PSM) elements and consider the safety of operations, maintenance performance, asset integrity, change management, major hydrocarbon releases and other measures (14 in total within the OSHA process safety management model). These process safety metrics are then developed into key performance indicators (KPIs) which are intended to provide important information on the effectiveness of the safety systems, and an early warning of impending catastrophic failure. One of the key requirements for the successful implementation of process safety performance measures is site specific data. Typically, the data is already available within the organisation but not being utilised to receive the full benefit.

A view of perception and interpretation

The intent of process safety KPIs is to form an important prerequisite for companies’ leadership to ensure that they know what is happening and how effectively the major hazards are being controlled. To successfully deliver this message, the KPIs need to be relevant, correct and understood by those who manage them.

An example of perception and interpretation of a wider performance target is the rumour that former US President, Bill Clinton, was said to have read one book per day for a year. Upon hearing this information, most people assume he was reading classic novels, and believe this to be a formidable feat of omnivorous reading. But it could be suggested that most parents are probably exceeding this target by reading stories to their children on a daily basis. Perhaps we will never know exactly what Mr Clinton was reading, hence demonstrating there is an automatic human position to accept these statements without challenge.

The same effect can be applied to performance metrics in which the results go unquestioned. It is apparent that senior managers are making the KPI results more visible within the organisations and viewing them regularly, but not the information that lies behind the on target ‘green’ metric on a process safety report. It appears to be the case that for the operator, the most important aspect is for the target to be green without necessarily understanding the measures or the data that lies beneath, and some operators are suffering from what could be called ‘target blindness’. This is where senior management are so used to seeing the same suite of metrics month on month with no change. In these cases, it is important to implement continuous improvement to refresh or challenge the performance data in order to ensure that the process safety measures are not losing their importance or meaning.

For external stakeholders who are interested in the process safety performance of a company, an audit of the process safety metrics is an obvious place to start. External audits have the benefit of bringing a fresh set of eyes to review the performance metrics and identify deficiencies, whilst bringing learnings and best practice from other operators. However, during this process it is surprising how often gaps within the safety performance can be found by external auditors.

An example of this was seen for a large complex energy installation within the Middle East for the tracking of the management of change (MoC) process performance. The MoC performance metric was clear: 90% of MoC documentation to be completed within the required timeframe. The target had been reported as being on track for several years and was shown as being ‘green’ within the monthly process safety report metrics. However, closer inspection of the data that sat behind the metrics score revealed that the site was reporting a target that was 90% of the 90% target. This meant 81% overall performance, which was actually well below target and should have been displayed as being ‘red’ (out of target). Therefore, there was a risk that a higher level of nonconformity meant that the company was running at a higher than anticipated risk level for project documentation, which could result in a process safety incident. Inaccurate process documents following a modification has been cited many times within the industry as a contributing factor to process safety incidents.

Learnings from industry losses

Shortcomings within the management of process safety performance has been a key finding in some of the industry’s most significant incidents in which senior management in the energy industry had failed to adequately monitor the status of key process safety performance indicators, leading to poor decision making.

Some examples of this include:

• On 23 October 1989, the Phillips 66 petrochemical facility in Pasadena, US, exploded, killing 23 workers at a major chemical plant, injuring a further 314 workers and resulting in one of the largest property losses in the petrochemical industry. The explosion resulted from a release of approximately 40 t of flammable process gases during a routine maintenance operation, which were released through an open valve almost instantaneously. The initial blast registered 3.5 on the Richter scale, and the ensuing fire took 10 hours to bring under control. The magnitude of the explosion and the significant loss of life that occurred was tragic proof that a catastrophic loss of containment could lead to severe consequences and helped modernise process safety management programmes in the US and globally.
• In 2000, three incidents at the Grangemouth, UK, refinery occurred, including a large process unit fire caused by a significant leak of hydrocarbons during start up procedures. One of the key findings from the investigation was that high hazard energy installations require a specific focus on process safety management and cannot rely on occupational safety indicators to provide an indication of process safety performance.
• On 23 March 2005, the Texas City Refinery, US, exploded when a release of hydrocarbon vapour was ignited, killing 15 workers, injuring 180 others and severely damaging the refinery. During the investigation it was found that at the time of the incident, the process safety KPIs were either under development or not sufficient enough to drive the site’s performance for process safety.

Each of these incidents call into question whether there is a thorough understanding of the importance of process safety management information to provide leadership in major hazard risk control in some companies. Unfortunately, significant incidents resulting in fatalities, injuries, environmental damage, and property loss have continued over the last 30 years since the explosion in Pasadena. The more recent losses at the offshore rig fire, explosion and oil spill in the Gulf of Mexico in 2010, and the toxic chemical release in LaPorte, US, in 2014 are reminders that global improvements to the management of process safety elements and their associated metrics are yet to achieve the goal of preventing significant process safety incidents.

Best practice

Business objectives change over time, i.e. make X amount of money or be more efficient. As such, leadership teams look to apply a ‘ratchet effect’ to optimise the company from a financial point of view. However, safety metrics have a different goal and sit outside of other company targets. The goal for process safety will always be for zero incidents, hence why applying a ratchet to improve this performance is not as applicable as for other company targets. Upon adopting process safety KPIs, a company may set a road map towards year-on-year continuous reductions in incidents and a tightening of the PSM metrics to fulfil the ‘achievable’ element of a SMART objective. But the most powerful question to ask for continuous improvement is: is the system working?

In terms of best practice, guidance for documented process safety systems is provided by OSHA and ISO Standards. It is worth mentioning that these standards provide a structure, and hence the implementation of that structure has to be suitable for the challenges and hazards within the organisation. Traditional indicators such as lost time injuries are important for companies to track, but are not true PSM elements and do not provide a good indication of process safety performance.

Written by Dr Jason Shirley, ECP Energy & Chemical Professionals, UAE.

Read the article online at: https://www.hydrocarbonengineering.com/special-reports/18062021/process-safety-targets-planning-to-fail/