Closing the loop – a digital approach to process safety management
07 October 2020
Process safety management (PSM) can be defined as the prevention of unplanned and uncontrolled loss of containment that can result in harm to people and the environment, and business losses.
Successful PSM is therefore a business imperative for any operating company in the hazardous process industries in order to protect personnel, the environment and its license to operate; mitigate the risk of major incidents; satisfy increasingly stringent regulations; reduce costly production downtime and reactive maintenance; and ensure stakeholder and public confidence is maintained.
In addition to avoiding a major process safety incident, managing process safety effectively, throughout the entire asset lifecycle – engineering, construction, operation, maintenance, modification, and closure – can also avoid significant financial and operational costs.
Preventable operator error is responsible for 40% of production losses; reducing this by even a single incident using effective PSM equates to a significant increase in revenues. Running an operation in a safe way also reduces process variability and gives operators more time to optimise performance.
Digitalisation provides the opportunity to improve how companies can manage process safety effectively and optimise the cost of safety. Digital solutions can extract relevant data from across the enterprise domain, enabling operations and management teams to continuously learn, adjust and improve PSM, and turn that data into actionable intelligence to ensure they have the right information at the right time, to make the right, informed decisions.
The role of regulation
Companies operating high hazard processes, ask yourselves the following three PSM questions:
Do we understand what can go wrong?
Do we know what systems we have to prevent this from happening?
Do we have sufficient information to assure us these systems are working effectively?
If you can answer ‘yes’ to all three questions, you likely have in place good risk identification and assessment processes, comprehensive risk controls, and reliable monitoring of those controls.
These are essential to satisfy PSM legislation enacted in the wake of real-world incidents. For example, the Seveso Directive – the main EU legislation dealing specifically with the control of onshore major accident hazards involving dangerous substances – came into being following an industrial accident in 1976 at a chemical manufacturing plant in northern Italy.
Furthermore, major incidents such as the Texas City Refinery explosion in 2005 have highlighted the need for addressing potential barriers at organisational, operational, and technical levels. This includes reviewing PSM leadership, placing a focus on process safety, as well as occupational safety. Managing change is key across all barriers, including ensuring operational procedures are up to date and being followed and critical maintenance is performed on time, and faults corrected. A top-down, integrated approach reviewing and tackling all barriers is essential for any company’s PSM policies and procedures.
In the recent past, the onus was on large chemical companies to implement their own process safety management systems and procedures. Now, organisations such as the Centre for Chemical Process Safety (CCPS) in the US and the UK Energy Institute (EI) have developed recognised PSM frameworks and guidance to ensure both large enterprises and SMEs are meeting their process safety obligations.
These PSM frameworks mean that companies can begin to benchmark themselves against the guidance and against other companies. This inevitably has implications for a company’s spend profile. Of course, there is a cost attached to process safety, but this must be weighed against the financial and reputational impact if something goes wrong. These frameworks allow firms to assess if their operations represent a substantial risk and if they should be investing further into PSM.
The four stages of PSM
At ABB, we carry out a wide variety of risk assessments including HAZID (hazard identification); HAZOP (hazard and operability); Process Hazard Reviews (PHR), functional safety including Layers Of Protection Analysis (LOPA) and Safety Integrity Level (SIL) verification; quantitative risk assessment; and human factors including alarm management and occupied building risk assessments.
Our approach to process safety in the hazardous process industries is focused on four key stages outlined in the Energy Institute PSM Framework: leadership; risk identification and assessment; risk management; and review and improvement. It has been developed over many years and builds upon operational heritage from Imperial Chemical Industries (ICI), who are accredited to be the founder of the HAZOP methodology.
In recent years, both OSHA and the HSE have started making sure that process safety leadership, the first stage, comes from the very top, and we are seeing PSM becoming a priority at board level, which is very encouraging.
Stage two, risk identification and assessment, relates to the first two key process safety questions. Our multi-stage Hazard Study approach starts by making sure that the facility is inherently safe in the first place, where possible, removing potential process safety issues. Next is HAZID where we take a top-down approach of looking for potential loss of containment and aim to identify improvements earlier in the design before HAZOP is performed. Typically, it is at the HAZOP stage when the design is fixed. If improvements have not been found before this stage, the end user can be left with costly safety system designs that are difficult to maintain and can have an impact on production.
Stage three, risk management, relates to the third of the three key process safety questions (as does stage four), ensuring that the various systems are working safely and correctly in order to keep the risk at an acceptable level. To do this, the information from the previous PSM stage is required so that the operation and maintenance teams can understand the impact if maintenance on the various safety systems is required to be deferred, if equipment is required to be taken out of service during normal operation and ensuring systems are in place during critical operations such as start-up/shut down of a facility.
The fourth and final stage is review and improvement. In addition to supporting greenfield studies (during stage two), we also support companies performing Process Hazard Reviews after a facility has been in operation for a period of time, typically five years. A number our of customers are finding that this approach of reviewing and refining their original basis of safety is much more efficient compared with redoing the entire HAZOP study.
Industry 4.0 and Process Safety Digital Twin
Digitalisation has introduced a new concept, Digital Twin, which is a digital copy of the ‘real-thing'. Simple examples of digital twins are a 3D model of a facility or a process model of a chemical process. The outcome from the risk identification and assessment stage can be considered the ‘Process Safety Digital Twin,’ that is a digital record of the basis of safety of a facility.
A HAZOP study, which forms part of the Basis of Safety, can last for several weeks or months, involving a multidisciplinary team, often including team members that are not part of the operating company. The knowledge that is collected is invaluable, but it is often stored in a non-digital format, where people that really need its knowledge – the operations and maintenance team – cannot use it or in some instances cannot find it. This effectively means it is a ‘long lost’ twin, not a digital twin.
For operating companies reading this article, how many of these questions can you truthfully answer yes to? Do you have a HAZOP (this should be yes!), do you know where your HAZOP report is stored? Have you looked at it in the last twelve months? Three months, one month, this week? Do you actively use the knowledge contained in it to conduct operations and maintenance?
This is where industry 4.0 comes into its own. By digitalising information contained in reports such as HAZOPs – either by transferring existing records for brownfield facilities or creating as part of the greenfield project – we create that ‘process safety digital twin.’ Operation and maintenance teams can then easily use this knowledge for decision support during activities to ensure risks are suitable managed (PSM stage three) – enable personnel to immediately pull up the relevant process safety hazardous event associated with specific pieces of equipment and critical safety systems.
Digitalisation, big data, and industry 4.0 has for several years involved the convergence of IT data (Information Technology), with OT data (Operational Technology). Before these initiatives, the question was ‘how do we get that data?’ Now it is ‘what do we do with it?’ This convergence of IT/OT systems is producing increased data, but only a small part of this data is safety related. The process safety digital twin (also referred to Engineering Technology data or ET) is the key to unlocking that data and turning them into useful, actionable information and can be used to keep facilities safe and avoid unwanted production outages.
Context is key. Is a facility potentially at risk of a major process safety incident or damage to a non-critical pump that has a standby? This is one element of the Watermelon effect (green on the outside, but as soon as you cut into it you see a lot of red). Another Watermelon effect is where digitalisation has not been adopted and companies rely solely on manually collated Process Safety Performance Indicators (PSPIs). How can we ensure the accuracy of this data? When teams are busy, are reporting periods missed? Has data been transposed incorrectly, or is there a practice to put in short- term effort to improve PSPIs, leading up to the reporting period?
By using digitalisation to automatically collect the IT/OT data in conjunction with a process safety digital twin (ET), we get the needles, not the whole haystack and avoid the Watermelon effect ensuring the risks on our facilities are managed (PSM stage three), and provide the platform for PSM stage four.
Closing the loop - optimising the cost of safety
What do we mean by closing the loop? Taking actual operational data and comparing it with the early engineering assumptions. In other words, PSM stage four – review and improvement. This stage is key, and as with any digital twin, the process safety digital twin must be validated to ensure accuracy, otherwise the Basis of Safety for a facility is wrong. If the Basis of Safety is too pessimistic, then that is costing a company in terms of maintenance and process interruptions. If the Basis of Safety is too optimistic, then a company is likely to operating at a too high a level of risk of a major process safety incident than is acceptable.
It is important to remember that the studies carried out in early stages of the functional safety lifecycle (which is part of the overall PSM), are predictions based on published document and historical knowledge of the review team. These assumptions are made well before the start of operations. When a company moves into the Operation & Maintenance phase, we need to gather actual operational data that allows us to close the loop, confirming the assumption or by adjusting the assessment and making improvements.
Since 2003, the operation and maintenance clauses within the Functional Safety standard IEC61511, have required companies to close the loop and validate the original design assumptions. Edition 2.0 has further emphasised the need to do this but has also included a ‘carrot,’ stating that maintenance can be reduced if evidence is gathered to demonstrate equipment is performing better than first predicted. As well as improving safety by the identifying bad actors, this now provides the opportunity to reduce routine proof testing maintenance and the costly associated process interruptions.
Digitalisation is again the solution to unlock that opportunity, automatically collecting IT/OT, including data from CMMS systems such as SAP and Maximo and being able to automatically compare and update the basis of safety, confirming the process safety digital twin (ET).
Karl Watson, ABB Energy Industries
Another example of closing the loop is the period review required by the PSM regulations. Typically, there is a requirement to perform this on a five-year cycle. As this approaches, most downstream operators begin to collect and collate information such as how many trips they have had, how many near misses reports and so on. This is not a very streamlined approach and often done in silos. But if it is in a digital format, we can start to measure the actual performance against what we thought in the process safety digital twin, throughout the five-year period, and not at the end (another potential Watermelon effect).
For example, if demand on a safety system is being originally estimated to occur every ten years, but in actual operation we have had two demands on that equipment in the last two years – the demand rate is more like once a year and therefore constitutes a much higher risk. Using the process safety digital twin, we can then assess why there are more demands on that system and improvements can be made. If the increased demand rate is due to controllers being left in manual mode due to poor control, we can tune the loop correctly, inform the operations team to keep it in automatic mode, which will reduce demands on the safety system, improving PSM, while also improving productivity by reducing plant trips.
Software to support the lifecycle of process safety management
Digitalisation provides the opportunity to manage process safety risks more effectively and help to identify improvements that will improve safety and also optimise the overall costs of safety. To take advantage of this opportunity, software solutions that support the safety lifecycle from engineering, to operation and maintenance, and also ‘closing the loop’, will play a pivotal part in the next generation of process safety management systems.
A lifecycle approach means that the actual performance can be immediately compared against the original design assumptions, in the context of risk. If new causes of demands are identified for a Major Accident Hazard, then it can facilitate the re-review of that hazard, updating the basis of safety and then making that update available to the O&M teams. For lower risk hazards, data is stored ready for the periodic, five-year review.
The benefits of closing the loop on process safety management are clear to see. By collecting real failure rate data and comparing it with initial predictions, we can demonstrate that assets and equipment may be functioning better than first predicted. This approach has paid dividends for a customer at its refinery in Europe where the number of tests that are needed to be performed have been reduced by as much as 1,400 a year. With these tests taking, on average, around four hours and typically involving at least two people, the company has been able to achieve $1 million in maintenance savings per annum, with the removal of any process interruptions to perform these tests.
About the author:
Karl Watson is a Global Process Safety Sales Solutions Architect for Energy Industries, ABB, and has over 30 years of experience working in process industries.Karl has been a TUV functional safety engineer for over 10 years. He has supported clients throughout various phases of the process and functional safety lifecycle. In his role as Global Process Safety Product Manager, Karl was responsible for creating ABB’s Safety Lifecycle Management software solution.
Karl’s current areas of responsibility include understanding customer process and functional safety needs at site, regional and enterprise levels, then identify the appropriate digital solutions to meet these needs, with the ability to also incorporate ABB’s safety and automation technologies, as well as safety consulting services, into these solutions.
Contact Details and Archive...