article_426_matrikon

Five justifications for alarm management

July 17, 2005
Does the financial ROI combined with the overall operational benefits justify the expense of implementing an alarm management solution? This article finds that the most effective way is to perform an alarm assessment.
By Jeff GouldTHE MOST OFTEN asked question by anyone considering an alarm management project is probably, “How do I justify alarm management?” There are plenty of articles written on how an alarm system should perform but how can a company justify and validate the necessity to do alarm management?Justification is not just financial ROI; rather, it is the combined benefits and needs associated with solving a problem. The question that needs to be answered is this: does the financial ROI combined with the overall operational benefits justify the expense of implementing an alarm management solution?The most effective way to justify alarm management is to perform an “alarm assessment.” This is the first step in the six phases of any successful alarm management project:
  1. Benchmark & Assessment
  2. Alarm management philosophy
  3. Alarm analysis / rationalization.
  4. Implementation and execution
  5. Continuous improvement
  6. Maintenance

This article will examine the first phase, Benchmark & Assessment, as it relates to justifying alarm management projects. For more information about the other Alarm Management Project phases, please refer to Senior Applications Engineer, Michael Marvin’s whitepaper, “Five Steps: Integrating Alarm Management to Boost Production.”

The Alarm Management Assessment
Alarm management assessment is comprised of three sub-sets:

  1. Technical Benchmark & Assessment
  2. Business Justification
  3. Project Plan
The Technical Benchmark & Assessment
Technical Assessment addresses measurable and philosophical information associated with the current alarm system. Quantifiable metrics are based on static alarm configurations (static alarms are the alarms that are configured on an analog or digital input, or control block on a control system) and real-time, or dynamic alarms (alarms that ring in and appear on the operator’s console). Assessing the static alarms reveals whether too many alarms are configured, thus generating too many real-time alarms for the operators to handle. Similarly, by analyzing real-time alarms, operators can determine whether there are too many alarms (irrelevant or nuisance alarms) in normal load conditions and during upset situations.For the sake of simplicity, the target alarm counts for each metric are summed up in the following two figures.
NOTE: A common misperception is that the target average must be 144 alarms per day. If the plant is operating normally 24 hours a day and 7 days a week then 144 is the correct objective. If, however, upsets occur as is the case with the majority of facilities, then a more typical average number will likely be around 200 alarms per day.A common pitfall when benchmarking and assessing an alarm system is the tendency to not collect event information. The top 100 operator actions that are archived into an alarm and event historian, as well as the incidents that are logged into the incident database, will provide significant information to apply to the creation of an accurate business case for alarm management.The last component of Benchmark and Assessment is a review of the alarm philosophy, assuming one exists, to determine whether it sufficiently outlines management of change procedures (who is changing parameters, when it happens and why it is being done) and whether philosophy objectives are met. Finally, interviews and reviews should be completed to assess the operators’ perception and experience with the current alarm system.Only after this information has been collected, can the next subset, Business Justification, be completed effectively and easily.The Business Justification
Alarms are designed to help the operators mitigate safety, environmental and financial loses. Incident investigations repeatedly show a very real financial loss associated with any safety and environmental incident, therefore, any properly configured alarm that is missed by an operator has a financial loss associated with it. All too often, it is the dramatic events that are scrutinized and brought to attention, yet it is the small, lower priority alarms that create a significant compounding financial effect. To accurately define the priority of alarms, all alarms must have an Impact Category assigned to them to determine both the level and area of concern.
Once the consequence of not responding to an alarm is understood, the time that an operator has to respond to the alarm must be considered. The alarm priority is set taking both of these elements into effect.
Since every alarm should alleviate loses, improper alarm configurations end up continually costing the plant profitability and opportunity. Based on this understanding of what an alarm is and the significance of each alarm, a successful business plan can be created and a justification can now be made regarding the size and scope of an alarm management solution. An effective guideline to follow is that each successful justification must address one or all of the following three key concerns:
  1. Environment, health, and safety issues
  2. Insurance costs
  3. Plant and operational reliability
Justification #1: Regulatory RequirementsMost alarm systems are not properly configured and thus do not meet regulatory requirements. Regulatory bodies understand and recognize that a sound alarm system is a critical component of a safe and effectively operated facility.The EEMUA #191 (Alarm Systems: A guide to Design, Management and Procurement) is the guideline that was commissioned by the Health and Safety Executive (HSE) in the United Kingdom. The guideline’s purpose is to provide the HSE with a reliable tool to measure the effectiveness of any plant’s alarm management program. Numerous UK based plants have been mandated by the HSE to implement alarm management programs. Failure to meet these requirements often means the potential loss of the site’s operating license in addition to hefty fines.The significance of the EEMUA guideline extends beyond the UK. Other major regulatory and governing bodies have adopted the EEMUA guidelines as the basis to audit and measure plant safety practices. These bodies include the ISA (The Instrumentation, Systems, and Automation Society), Responsible Care, and the European group NAMUR (Normen Arbeitsgen Mess Und Regeltechnik), to name a few.The following table illustrates how facilities that do not have an alarm management program compare to EEMUA real-time alarm targets.
Justification #2: ISO Adherence
Some sites have identified alarm management as part of their ISO practices. If specified as part of the ISO procedures, the site must have the proper procedures and tools in place to ensure their alarm system is properly managed. Without the right tools it is usually difficult and time consuming to achieve these objectives - often, it is impossible.Justification #3: Operational Financial Benefits
Evaluating incident databases and recording whether a missed alarm resulted in a particular incident, can determine the financial impact associated with alarm management. For most plants this will often be around 15 related incidents per year based on close analysis of the data. When developing an alarm management plan as part of a six sigma project, it is important to document both the financial situation before and after an alarm management program so that progress can be tracked. Based on estimated incident costs, an average savings of $250K is often achieved by reducing incidents and improving plant reliability.The next task in evaluating the financial benefits is to examine alarm and event data, assuming this data exists. By analyzing alarm floods, significant and costly events that were not immediately evident, can be quickly identified. Alarm floods need to be analyzed to determine if there is a cost, what the cost is, and whether proper alarm configuration would prevent similar events from occurring again. Moreover, additional benefits will usually arise with the identification of previously unnoticed control or equipment related problems.Operator actions have direct costs associated with them. Analyzing the top 50 operator actions will provide concrete problems and costs that often go unnoticed. Controllers that are often switched from auto to manual, or set points that are frequently changed, are indicative of control or equipment problems. With an experienced person on site, the associated cost can be quickly determined.Finally, a very real benefit, although less tangible, is that an improved alarm system will significantly reduce the number of operator alarms. Reduced alarm volume means the operator has significantly more time, as much as 50%, to monitor and optimize the process. Process optimization can result in significant throughput and quality benefits.Justification #4: Insurance Findings, Premium Savings, and Requirements
A growing number of insurance companies are demanding that sites and organizations implement alarm management programs because of the correlation between alarm management and equipment damage. The reason for this change in policy is that each year within the petrochemical industry alone, North American insurance companies pay out $22 billion in claims related to equipment damage. Insurers that do require alarm management programs, typically offer a reduction in insurance premiums. For self-insured plants, financial savings can be even more significant and are identifiable once an assessment has been completed regarding the cost of damaged equipment caused by poor alarm management.Justification #5: Operational Reliability
Alarm management plays a significant role in any corporate operational excellence program. Since alarms settings are based on equipment protection limits or process limits, exceeding these limits results in excess wear on equipment and loss of production or lesser quality product. These alarm indications can provide insight into how equipment is being used and how it is affected by both manual and automatic operations. With this insight, adjustments can now be made to prolong the operating life and reliability of the process. Most maintenance departments will attest that the life of an asset is strongly tied to its operational use.The Chicken and the Egg Conundrum
How can the justification to acquire the appropriate alarm management tools be made if the very same tools are required to develop the business case justification? Matrikon offers an alarm assessment service to address this paradox. Matrikon’s alarm management tools serve a dual purpose: to perform a concise and efficient assessment of the plant’s current alarm management system and to provide the client with a firsthand evaluation of Matrikon technology.The data and analysis gathered from an assessment will provide a clear understanding of the problems and weaknesses in the alarm system. This becomes the foundation of a sound business justification and project plan.Project Plan OutlineThe project plan is a logical outline of the proposed work and deliverables. The details from phase one, benchmarking & assessment, will help determine whether an entire plant’s alarm system needs to be configured or if it can be limited to key areas.All project plans should now have the remaining 5 phases:
  1. Benchmark & Assessment
  2. Alarm management philosophy
  3. Alarm analysis / rationalization
  4. Implementation and execution
  5. Continuous improvement
  6. Maintenance

Again, for more information about the other Alarm Management Project phases, please refer to Senior Applications Engineer, Michael Marvin’s White Paper, “Five Steps: Integrating Alarm Management to Boost Production”.

Conclusion
Alarm assessment is a well-defined process that provides the technical situation, business justification and plan of action for a particular site. The main ideas addressed are:

  • Every alarm configured has a financial impact associated with it.
  • Every alarm that is missed, both small and large, costs money.
  • Is the alarm system properly configured?
  • Are all levels of operation reaping the benefits from the alarm system?

Assessments are standard practice for experienced integrators and can typically be performed within 1-2 weeks after which a very high quality and accurate assessment is delivered.

  About the Author

Jeff Gould is the ProcessGuard Product Manager responsible for overseeing the development, marketing, and sales of Matrikon’s Alarm Management product line. Jeff joined Matrikon in 1996 and helped to develop Matrikon’s driver business, and for the last few years has been responsible for growing and shaping Matrikon’s alarm management solutions. Jeff graduated as a computerengineering technologist from NAIT, and his background includes several years of software development management, and sales and marketing of software solutions for the process control industry.