Appendix VI

Cost Impact Discussion

1.0. General

So far during the discussions, various technical aspects have been covered with main focus on safety instrumented systems (SISs) and life cycle (LC). Naturally, it is clear that there is no substitute for LC studies and SIS to face hazardous conditions. However, nothing in the world is available free of cost. Therefore, LC and SIS efforts will also have some cost impact. In this part, short discussions will be put forward on cost impact on account of LC and SIS. In line with guidelines from HSE. UK cost benefit analysis will be very helpful in assessing whether risk reductions are reasonably practicable. One has to weigh cost and benefit prior to applying risk reduction measure. When cost divided by benefit is a very high value, that is, disproportionate factor is much greater than 1, it may be prohibitive. However, cost benefit analysis (CBA) has certain limitations like:
• A CBA cannot be used to argue against implementation of relevant good practice or any statutory duties.
• CBA of its own does not constitute as low as reasonably practicable (ALARP).
On the other hand, in order to run a unit, it has to be below ALARP. So certain risk reduction measures and SIS are inevitable. Naturally there will be cost impact on account of the same. So, the discussion that follows is based on the fact that LC and SIS are required and justified. Having accepted the above basis cost analysis, is still essential to know real impact as well as what could be the optimum one.

1.1. Safety Instrumented System Justification

From the management point of view, it is essential that the cost impact on account of buying, maintaining, and operating a safety system (i.e., SIS and associated LC). So, a CBA toward SIS will be helpful for all especially to convince the top management with reasoning. In view of IEC 61508 and 61511, LC is closely associated with SIS, so all the times two are not used separately, meaning that SIS will always have LC with it–hence associated cost for the same. SIS may be justified may be on account of a few reasons listed as follows:
• Essential for safety and no alternative methods exist
• Statutory requirement/international standard/code demands
• Lowest cost option for safety (for this CBA will be helpful)
• Prevention of environmental harm/violation of emission standard limits
• Protection against asset losses due to plant damage/lost production
• Protection against human safety
• Protection against loss of company image for not following good engineering practices, which have following major influencing factors:
Reputation
Share price
Customer base
Market share

1.2. Cost Impact on Safety Integrity Level

From the discussions in Chapters VII and VIII, it is clear that SIS and safety integrity level (SIL) are very closely coupled. Also, in order to achieve higher SIL, at times people may have to go for higher redundancies and/or fault tolerant design, which by itself will add cost to SIS. The higher the SIL value, the higher will be the cost of the system as is clear from Fig. APVI/1.2-1.
The figure clearly shows that costs escalate significantly with increasing SIL targets. This is because of the fact that on account of higher SIL, one may have to go for redundancies at various levels, as well as a few other factors such fault tolerance issues, safe programmable logic controller, and/or higher maintenance requirements (e.g., as per IEC 61511, SIS will be proof tested at an interval directed by the SIL) will increase the cost. With the same in mind, it is better to look at details into the system, by exploring life cycle costing process.
image
Figure APVI/1.2-1 Cost impact on safety integrity level.

2.0. Life Cycle Costing Process

Life cycle costing (LCC) shall be done in such a way that it is clear to the investor. Appropriately developed LCC, along with good engineering judgment, provides a rich set of information for making cost-effective, long-term decisions in a disciplined manner. A typical LC structure for SIS has been depicted in Fig. APVI/2.0-1. As is seen from this simplified figure, there are two clear parts in the cost structure.

2.1. Initial Capital Cost

The total SIS cost comprises two parts; one is initial capital cost, and the other is recurring cost. Fixed costs come from:
• Design engineering cost
• Supply charges for:
Sensors
Logic solver
Final element
• Training charges
• Other service related charges (including supplies), for:
Erection/installation
Testing
Commissioning (including start up)
These costs are mainly incurred at the initial stage of the project up to start up.

2.2. Recurring Charges

Recurring charges are incurred during main running conditions of the plant in terms of the operation and maintenance support charges. Again, there are two parts; one is fixed and the other is variable. These shall include mainly:
• Fixed charges:
Staff salary
Service charges (annual maintenance contract)
Training
• Maintenance charges in the form of
Spare inventory
Repair charges
Software maintenance
Testing (e.g., proof testing)
• Variable charges (incidental charges):
Hazardous events
Spurious trips
image
Figure APVI/2.0-1 Life cycle cost structure for safety instrumented systems (typical).

2.3. Variation of Costing Due to Different Reliability Model

In Clause 1.2, it has been shown that on account of variations in SIL requirements, there are variations in overall costing. Now from a reliability point of view, there will be variations in costing also, even if there is no change in SIL number. HSE.UK [3] gave a good example to show that in two cases for same SIL number (same overall PFDavg), the number of spurious trips are different due to different redundancies chosen at sensor and at LS levels. Since there is variation in numbers of spurious trips, hence there will be a change in overall costing. These are already discussed in Chapter VIII. For details, HSE.UK [3] may be referenced also. With this preliminary knowledge in mind, it is better to explore the LCC analysis part, which is also very interesting.

3.0. Life Cycle Cost Analysis

In a project/process/product, there is involvement of a number of disciplines. Each of these disciplines has different ways of looking at LCC. A person from project engineering would like to reduce the capital expenditure. A maintenance engineer is interested in ensuring that LCC be done in such a way that good maintenance strategy is adopted to reduce repair hours. A reliability engineer will like to have low failure rate, while production personnel will like to have higher operating hours. Accounting persons and share holders would like to see increase in net present values (NPVs) and stock value, respectively. Why all these are discussed? This is because LCC analysis is a joint effort requiring input from various disciplines. So good coordination is essential. With this in mind, let us look into the details of the analysis.
Net present value (NPV) is an important economic measure and concept used in projects to present actual situations taking into account discount factors, cash flow, and time. In LCC, analysis NPV is used. In order to make any major important decisions, project engineers/managers highly depend on life cycle costs. This calls for various considerations including how and when sustaining costs occur during the LC of the equipment or project. Adding expected equipment failure rates and renewals from a statistical viewpoint makes analysis about economics smarter and gets the rational decisions closer to real world conditions. In this connection, guidelines from international standard IEC 60300-3-3:2005 may be referenced. This standard is the basis for LCC analysis discussions. Engineers must supply facts (not opinions) for LCC calculations [2]. From Fig. APVI/2.0-1, some idea about the acquisition cost and sustaining cost (2–20 times acquisition cost) could be gathered. One needs to keep in mind that this evaluation process is not really a one-time process, but iterative to get the best alternative at that point of time. Before moving on to any other decision-making issue, it is better to have look how such process is carried out.

3.1. Life Cycle Cost Analysis Process

The process discussed here is in generic term as with different facilities there may be some variations. However, the procedure discussed here is more appropriate for process industries such as chemical and oil and gas plant LC cost analysis for SIS. Fig. APVI/3.1-1 gives a general overview of the process. As this is an iterative process, a feedback line from evaluation to problem definition has been shown. This is used to get best possible alternative by iteration.
There are the following seven main steps into which the whole system can be divided:
• Problems definition
• Cost elements definition
• System modeling
image
Figure APVI/3.1-1 Life cycle cost evaluation for safety instrumented systems. This figure has been developed based on idea from Y. Kawauchi, M. Rausand, Life Cycle Cost (LCC) Analysis in Oil and Chemical Process Industries, June 1999; http://frigg.ivt.ntnu.no/ross/reports/lcc.pdf.
• Data collection
• Cost profile development
• Evaluation
• Reporting
Each of these sub-steps has subdivisions also, which are shown by the ellipse associated with each step. Now, short discussions on each of them will be taken up in brief so that concepts about the same are well understood.

3.1.1. Problem Definition

There are three sub-steps in problem definitions. These are:
• Scope definition: As with any other issues, at the starting point it is necessary that the problem, scope, and boundary limit are well-defined so that there is no ambiguity in targeting the issue. Here the aspects, boundary limits of program phases, and the equipment and activities to be covered in modeling are defined to get a clear definition of the cost elements.
• Evaluation criteria: The criteria based on which evaluation is to be carried out are defined at the beginning as part of problem definition. The criteria normally cover the total cost, system performance, and effectiveness. The system performance characteristics (like availability, maintainability, and SIL of shutdown, etc.) and the effectiveness (like production capacity, product quality, etc.) shall be covered. In many cases regulation, codes, and standards specification play a great role at this point.
• Operational philosophy: This specifies modes and requirements of operation, maintenance strategies (predictive maintenance, proof test requirements, etc.). This is also important in case of problem definitions for LCC analysis. This is also somewhat dependent on an owner's prerogatives.

3.1.2. Cost Element Definition

There are two major issues here; one is cost breakdown structure (CBS), and the other is cost category. As mentioned earlier basic guidelines for the same is available from IEC 60300-3-3:2005 (with special reference to dependability).
• Cost breakdown structure (CBS): It is needless to tell that most vital work is to identify all cost items/cost elements that have considerable influence on the total LCC of the system. Fig. APVI/2.0-1 gives broad structure for costing. Naturally, the same is to be defined in a systematic manner. Also, development of CBS is required as per IEC 60300-3-3. CBS may be developed by defining items along three independent axes, which are, “life cycle phase,” “product/work breakdown structure,” and “cost categories.”
• Cost categories: It is difficult to define generalized cost elements that are applicable for every LCC analysis. This is because of the fact that LCC analysis may be applied to various types of systems each with different requirements. It is recommended that CBS and cost categories should be tailored for each application area for LCC analysis [1]. However, in line with IEC 60300-3-3, “acquisition costs” and “ownership costs” are fairly applicable in most cases. Based on this assumption, Fig. APVI/2.0-1 has been developed. At any time, it is possible to expand the cost categories on the highest level depending on the system to be analyzed. In many cases, like oil and gas or power generation, cost of deferred production (may be generally quantified based on the unavailability performance of the production system, and a unit cost of the product) are considered as it has a lot of impact on LCC. There are other cost categories to include costs like hazard cost, spurious trip cost, etc. as shown in Fig. APVI/2.0-1.

3.1.3. System Modeling

It is necessary to make a model taking in to consideration various factors such as:
• Availability
• Maintenance and inspection
• Logistics
• Risk involved
• Production regularity
• Environmental effect
• Human factor
These are already shown in Fig. APVI/3.1-1. These are influencing factors, so suitable care must be taken to model them to get realistic cost element.
• Availability: As already discussed in previous chapters, availability, maintainability, and human factor (depending on applicability) in SIS are very important and have tremendous impact on cost element.
• Maintenance and inspection: Two kinds of maintenance, viz. “corrective maintenance” and “preventive maintenance” have direct impact on cost element. The frequency of maintenance or inspection directly affect “availability” and the “operating cost” in terms of cost toward man-hours spent, spare part consumption, etc. Turnaround time, mean time to repair (MTTR), etc., are measures for maintenance costs for modeling.
• Logistics: Logistic support in the form of the following may be considered for modeling purposes:
Maintenance personnel
Training and training support
Supply support
Spare inventory
Support equipment
Computer resources
Packing, handling, storage, and transportation
Maintenance facilities
Technical data and information systems
• Risk: The potential risk related to a system is not only useful information, but it also needs to be considered in modeling. In this connection, IEC-60300-3-3 may be referenced (it recommends considering liability costs from risk analysis to be considered LCC analysis). It is also recommends to include warranty costs in the CBS of the LCC analysis [1].
• Production regularity is a term used to describe how a system is capable of meeting demand for deliveries or performance. It depends on system availability, production availability, and deliverability.
• Environmental: After the Convention on Climate Change, in the Kyoto Meeting in 1997, globally there is great concern over impact of production on the surrounding environment. Now almost all countries have their own laws or international laws and standards which need to be met by all concerned. Therefore, plant owners need to spend money toward pollution prevention, viz. SO2, CO2, and particulate emission from power plants; owners need to take necessary measures to limit such emissions within limits. Naturally, these will have impact on LCC.
• In the actual operation, the effect of human error cannot be overestimated. All such effects may have tremendous impact on hazardous situations, etc. So the contribution of the human error is not negligible in many cases, especially in cases where there are manual interventions frequently. There are broadly three categories of human error, viz.:
Omission error
Action error
Extraction error
Similarly, there many techniques to face and combat human error such as technique for human error rate prediction (THERP), human error assessment and reduction (HEART), etc. already discussed in previous chapters.

3.1.4. Data Collection

It is quite obvious that LCC analysis is done based on data, naturally the more accurate the data the more realistic the LCC will be. These data are input data necessary to carry out the analysis. There are two types of data; one is estimated data and the other is actual data. If latter one is available, then it can be directly applied to CBS. Otherwise one has to depend on estimated data based on expert judgments.
• Estimated data: When actual data is not available, the value of data may be estimated. The following types of methods are used for data estimations:
Stochastic: Specialized statistical method.
Parametric: Another statistical method used on historical data for estimation of cost factor and/or cost estimation, etc.
Analogous: Used to establish relationship between current data and previous data duly judged by experts.
• Actual data: There are many data sources available for various reliability, etc. data, but it is difficult to get actual cost data. So, one has to depend on operational and cost data from the database of the operating companies [1].

3.1.5. Cost Profile Development

One factor of LCC analysis is an affordability analysis with due considerations for long-term financial planning. Therefore, it is necessary to draw a cost profile over entire life. It is obviously noticed that the cost profile of each design case should be compared on a common basis or reference point when making financial judgments [1]. There are two aspects here:
• Model run: Cost profile is developed by running cost models developed in an LCC analysis with input data. Computer tools can be used to run a model or it could be manual calculations in a spreadsheet.
• Cost treatment: In this, various aspects like effect of inflation, interest rates, exchange rates and taxation, etc. are considered for financial judgment. Many times, the cost profiles are made on the basis of “constant prices.” This is adopted for those cases where it is difficult to accurately predict inflation and exchange rate, etc. In such cases, it is necessary to compare the alternatives on a common baseline.

3.1.6. Evaluation

It is the aim of this analysis is to find out the most desirable alternative configuration. In order to do the same, one needs to check if the baseline system meets the criteria defined in the problem definition. If not, the baseline system should be modified as an alternative system, and the LCC of the alternative system should be evaluated. There are several issues and checkpoints; major issues are listed as follows:
• Sensitivity analysis: The main task of sensitivity analysis is to find the impact of changes in input parameters on the result. This is done by making variations in the input parameters over a range to see if the impact on cost can help highlight the major factors affecting costs. There are several methods available for sensitivity analysis. Mainly “deterministic” and “stochastic” approach are used.
• Uncertainty analysis: Uncertainty analysis is done to consider possible ranges of the estimate and their effect on decisions. Three categories of uncertainties are:
Parameter
Modeling
Completeness
This will give confidence to decision-makers to make financial decisions.
• Cost driver: Identification of cost drivers in LCC analysis is one of the major issues. Cost drivers have major impact on the total LCC. Once a cost driver is identified, it is important to establish cause-and-effect relationships so that system design may be modified to effectively reduce the causes of cost drivers, which in effect will reduce the total LCC.

3.1.7. Reporting

Documentation of the entire process is extremely important and these reports could be used in the future as database.

3.2. Cost Analysis Timing

Theoretically, LCC analysis can be done any time in any phase of the project. In fact, it should be a continuous process. However, earlier identification of acquisition and ownership costs give the investor a better chance to balance reliability, performance, and maintenance.
Fig. APVI/3.2-1 provides curves of commitment and expenditure trends. There is another interesting curve, which is the dotted curve which shows how cost reduction chances vary with time during the LC of the plant/project/process/product. Little LCC opportunity exists with start after construction when it is not possible to significantly change LCC. An interesting feature to note here, is that about 95% of commitment has to be done prior to the end of acquisition cost period, but only 50% is spent at this time. Major expenditures will be done during the sustaining period. Another issue is uncertainty. An uncertainty curve within the range of LC period has been shown outside. Obviously, as the LC is toward the end of life, uncertainty asymptotically approaches to one, which is quite obvious. In view of the discussions, it is believed that tradeoff between uncertainty curve and commitment curve will decide the best timing for taking up LCC analysis.
image
Figure APVI/3.2-1 Life cycle cost decision at phases of project.

3.3. Life Cycle Costing Analysis Application

Application of LCC analysis shall include, but is not limited to the following:
• Assessing economic viability of projects/products.
• Evaluation and comparison of alternative design.
• Cost driver identification and improvement.
• Long term financial planning.
• Optimization of fund allocation for various activities and facilities.
• Evaluation and comparison of alternative strategies in different areas (e.g., maintenance).
• Evaluation and comparison of different approaches for renovation, etc.
• Assessment of product assurance criteria.

3.4. Codes and Standards

As indicated at the beginning, there will be variations of LCC analysis with industries and applications. There are various standards to cover the same; a few are noted as follows to conclude cost impact discussions.
• IEC 60300-3-1/23/9/11:2005 Life cycle costing (3-3).
• ISO 15663 Life cycle costing within the petroleum and natural gas industries.
• SAE ARP-4293: Life cycle cost - Techniques and applications.
• API RP 580/581: Risk based inspection.
Hope you enjoyed reading the book and look forward to your feedback in the form of review. Any feedback, comments (good or bad), or suggestions from you is very much valuable and is always welcome. – Author.

List of Abbreviations

AI/OAnalog input/output
CBACoast benefit analysis
CBSCost breakdown structure
DI/ODigital input/output
E/E/PEElectrical/electronics/programmable electronics
Table Continued

image

FEFinal element
FPGAField programmable gate array
HEARTHuman error assessment and reduction
HWHardware
IECInternational Electrotechnical Commission
I/OInput/output
I/P or O/PInput or output
ITInformation technology
LCLife cycle
LSLogic solver
MTTRMean time to repair
NPVNet present value
P&IDPiping and instrumentation diagram
SISSafety instrumented system
SWSoftware
THERPTechnique for human error rate prediction

image

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.12.120.202