It is ironic that alarm systems, which are intended to inform and help the plant operator in the control of his process by correct and timely information, often have the opposite effect, at the precise time when the operator demands the most assistance from his control system. This four part article will take a look at some of the common factors that contribute to this undesired situation and how they can be eliminated or greatly reduced.
Many technical papers have been published on alarm management or the lack of it, specifically with respect to alarm showers or floods, which normally follow as a consequence of large plant upsets. This problem will be discussed, but with the focus on attempting to prevent or reduce the magnitude of such disturbances from occurring in the first place; ie getting to the root cause of the problem. Abnormal situation management (ASM) is a relatively new technique, that is attracting much attention due to the tremendous cost saving potential in reducing the frequency and mitigation of the consequences following major process upsets. These ASM monitoring systems are embedded within the normal process control and ESD systems, they attempt to automatically emulate the actions of a well trained, experienced and observant process operator. Even our best plant operators can and do make mistakes, which are more likely to occur under stressful situations, such as major plant upsets. Process instrumentation and control systems, it is claimed, do not suffer from such human traits. However, if process operators had more faith in the instrumentation and conversely instrumentation design engineers less faith, this would no doubt improve the overall situation. Instrumentation engineers also require to listen to the needs of process operators and to implement effective control, display and alarm management strategies to reduce these stress levels under such major process disturbances.
Standards and legislation
In the early 1990s, a consortium of companies in the USA formed a project group which was partly funded by the US government, with the task of looking into aspects of ASM and methods of eliminating or reducing the impact on process operations from major plant upsets. It had been estimated that such incidents cost the US economy in excess of $20 billion annually, which close to equalled their process industries annual profit margins - serious money. If companies are willing to invest in various ASM packages (mostly software), which could reduce their costs following such plant disturbances by say 25%, imagine their resultant profits and competitive advantage! In addition, a resulting 5-10% increase in annual production is also possible without increasing capital (process equipment) investment.
A number of excellent standards have been produced in recent years on guidelines in the design and application of alarm and instrument safety protection systems. Two notable ones being the Engineering Equipment and Materials Users Association - EEMUA 191 on alarm systems and IEC 61508 on functional safety of electrical/electronic/programmable electronic safety-related systems.
Owners of facilities, where it is deemed that the plant or process involves a hazardous installation, are required by the OHS Act No. 85 of 1993, under the Major Installation Regulations amendment of 1998, to carry out a risk assessment or HAZOP (hazard and operability study). This assessment is to be performed at intervals not exceeding three years. Amongst other requirements, this assessment must determine and record the probabilities of major incidents occurring and their likely consequences, together with the measures to remove or reduce the associated risk to acceptable levels. It is quite clear that reliable and effective alarm monitoring with automatic equipment protection where necessary, in accordance with the application required safety integrity level (SIL) as defined in IEC 61508, will satisfy this requirement. The act also addresses the responsibilities of local authorities in allowing the erection of new major hazard installations and the building of community housing (formal or otherwise) near such installations (remember Bhopal).
Why disturbances occur
Major process disturbances can be attributed to the following six predominant causes:
1. Process equipment failure, including associated services eg power, cooling water.
2. Human (operator/maintenance) error.
3. Instrumentation failure, including power and instrument air.
4. Process changes due to composition of feed material or controller set point, however, these have to be quite substantial to create large upsets.
5. Unstable (poorly tuned) and interactive control loops, including slow measurement response and poorly maintained control valves.
6. Natural elements such as extreme wind, rain, lightning, etc.
The most serious and sometimes catastrophic disturbances are due to process equipment failure, as little warning is normally given and often have serious consequences such as plant fires. Statistically, equipment related failures produce the highest proportion of process upsets, closely followed by operator errors, these contribute to more than 80% of major plant disturbances. These disturbances if not correctly controlled, can affect the safety of personnel within the plant and communities living in close proximity to the installation. Environmental implications and economic loss due to replacement of equipment and lost production can also follow serious plant upsets. Public opinion can influence a company image and profitability, specifically if that company produces a general consumer brand product.
Most disasters occur due to a 'combination' of related sequential events, early warning detection of any one related event with subsequent corrective action, will often break the chain reaction and prevent the likely disaster from occurring or at least mitigate the consequences. Nature gives us early warnings of many impending natural and man-made disasters, if only we would listen and take early action. Similarly, our own body with its life giving pumping system (heart), has a pressure safety valve or to be more precise a bursting disc located in our nose. If we choose to ignore too many nosebleeds (indicative of high blood pressure), we should not be surprised if one day we suffer from a heart attack.
Many plant disturbances can be reduced both in frequency and severity, with surprisingly little additional instrument hardware. What is required however, is a good working knowledge of the process mechanical equipment and the associated measurement and control systems. This together with a good software package and associated functional logic which interacts with the process control and safety shutdown systems, is all that is needed for successful implementation. It should however, be mentioned, that to implement an effective and successful ASM strategy involves some considerable time with the use of application specialists, much of which will be site based to gather data, test, diagnose and fine tune the configured system. Extensive input is also required from the facility owner, especially on existing installations, together with operator/maintenance staff and system awareness training. An additional benefit that inherently follows is an improvement in the organisations operating/maintenance practices and safety culture. The payback in higher operational safety and productivity can be enormous, as has been indicated in various studies, so much so, that at least one operating company has declined to publicly issue associated implementation reports, for fear of losing additional market share and their competitive advantage.
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version