faisal i. khan-3

I2SI: A comprehensive quantitative tool for inherent safety and cost evaluation

Faisal I. Khan and Paul R. Amyotte1

Faculty of Engineering & Applied Science,Memorial University of Newfoundland,

St. John’s, NL, A1B 3X5, Canada

1Department of Chemical Engineering, Faculty of Engineering, Dalhousie University, Halifax, NS, B3J 2X4, Canada

ABSTRACT

The traditional approach to process risk management involves providing layers of protection between the hazardous agent and the people, environment, or property. The layers of protection are intended to reduce risk by reducing either the likelihood of potential accidents or by reducing the magnitude of the impact. The risk can be reduced to very low levels by providing a sufficient number of layers of protection, and by making each layer highly reliable. Risk reduction strategies, whether aimed at reducing frequency or mitigating the consequences of potential accidents, can be separated into four categories: i) inherent, ii) passive (engineered), iii) active (engineered), and iv) procedural. Among these four classifications, the first one – inherent safety – is of current interest and is the subject of discussion in this paper.

Inherent safety is a proactive approach to process risk management. Considering the lifetime costs of a process and its operation, an inherent safety approach can lead to a cost-optimal option. In spite of being an attractive and cost-effective approach, the inherent safety methodology is not widely used. Many reasons have been attributed to this lack of widespread use; the non-availability of systematic tools for the application of inherent safety principles is perhaps the most important reason.

This paper presents details of an integrated inherent safety index (I2SI). The conceptual framework of this index was presented at the 37th Annual Loss Prevention Symposium of the AIChE (2003) and is to be published in an upcoming issue of Process Safety Progress. In addition to the framework, the current paper discusses additional features of the index such as the cost model and system design model, which were not presented or discussed earlier. I2SI is called an integrated index because the procedure considers the life cycle of the process with economic evaluation and hazard potential identification for each option. I2SI is comprised of sub-indices which account for hazard potential, inherent safety potential, and add-on control requirements. In addition to evaluating these respective characteristics, there are also indices that measure the economic potential of the option. To demonstrate the applicability and efficacy of I2SI, an application of the index to three acrylic acid production options is also discussed in the paper.

Keywords: inherent safety, hazard indices, inherently safer design, loss prevention, risk management

INTRODUCTION

In a typical approach to loss prevention, safety measures are engineered near the end of the design process, leaving add-on control measures to be the only option available. Control measures added late in design require continual staffing and maintenance throughout the life of the plant, greatly adding to the lifetime costs as well as necessitating repetitive training and documentation upkeep. The inherently safer approach to loss prevention tries to avoid or eliminate hazards, or reduce their magnitude, severity or likelihood of occurrence by careful attention to the fundamental design and layout. In this approach, less reliance is placed on ‘add-on’ engineered safety systems and features, and procedural controls which can and do fail. Unlike these ‘add-on’ approaches, which increase cost and are maintenance intensive, applications of inherent safety can lead to enhanced safety and lower capital and operating costs (Edwards & Lawrence, 1993; Hendershot, 2000; CCPS, 2001). The inherent safety approach uses basic design measures to achieve hazard elimination, prevention and reduction. Classically, an inherently safe plant or activity cannot (under any circumstances) cause harm to people or the environment (Mansfield, 1996). The salient features of an inherently safe plant are:

i it uses harmless materials,

ii it contains small inventories of hazardous materials which are insufficient to cause significant harm even if released, and

iii the hazardous materials are held in forms or under conditions that render them effectively harmless (diluted, at ambient temperature and pressure, etc.).

In practice, such an ideal plant or process rarely exists as it is often the very reactive/special nature of materials that makes them useful in industry. It is therefore more helpful to think in terms of inherently safer plants or processes in lieu of inherently safe plants or processes. Inherently safer processes or plants pose less inherent risk as compared to conventional process/plant.

Professor Trevor Kletz (Kletz, 1985) was the first to formalize the principles of inherent safety. There are five commonly used inherent safety principles or guidewords (elimination, minimization, substitution, moderation, and simplification) which are considered to be the most general and widely applicable. Remaining principles developed by various workers are essentially sub-categories of the main principles (for example, avoiding knock-on effects may be viewed as a special case of limitation of effects, which in turn may be thought of as moderation; similarly making incorrect assembly impossible may also be viewed as a form of simplification). It should be noted that a somewhat different set of primary guidewords is sometimes used by health and safety practitioners. In the present work we have chosen to employ minimization, substitution and simplification, while separating moderation into attenuation and limitation of effects (limiting of). The main thrust behind such guideword-based approaches is to help designers strike a balance between hazard avoidance, prevention, control and mitigation. This encourages the use of basic design features to achieve such a balance, rather than over-reliance on ‘add-on’ active and passive systems (e.g. emergency shutdown systems and firewalls) that may fail. The

2

framework of a guideword-based inherent safety approach for loss prevention is discussed by Amyotte et al. (2003).

It has been proven that, considering the lifetime costs of a process and its operation, an inherently safer approach is a cost-optimal option. Lifetime costs include the fixed cost of the facility, as well as the costs for operations, maintenance, and safety measures. Conventional systems may be cheaper in terms of fixed and operational costs; however, considering maintenance and safety measure costs, these systems may turn out to be costlier than those based on the principles of inherent safety (which may well have higher fixed costs). There are numerous examples in the process industries for such situations (e.g. Kletz, 1998; Gupta et al., 2003). Inherent safety can be incorporated at any stage of design and operation; however, its application at the earliest possible stages of process design yields the best results (Khan & Amyotte, 2002, 2003a). Although it is an attractive and cost-effective approach to risk management, inherent safety has not been used as widely as other techniques, such as HAZOP and quantitative risk assessment. There are many reasons for this; key among them are a lack of awareness and the non-availability of a systematic methodology and tools.

There have been several efforts made by different agencies to develop inherent safety evaluation tools (Khan & Amyotte, 2003a) Examples include the INSET tool kit sponsored by the then European Community, the overall inherent safety index prototype proposed by Edwards and coworkers at Loughborough University, UK, the inherent safety index proposed by Heikkila and co workers at VVT, Finland, the fuzzy-based inherent safety index proposed by Gentile and coworkers at Texas A&M University, USA, an index and expert system for inherent safety evaluation of process flowsheets by Palaniappan and coworkers at the National University of Singapore, Singapore, and a hierarchical approach for chemical process inherent safety evaluation by Shah and coworkers at the Swiss Federal Institute of Technology, Switzerland. In addition to these tools there are several programs or methods that have been adopted within different corporations to evaluate the potential for inherent safety considerations in design and operation. Examples include the Rohm and Haas major accident prevention program, Exxon Chemical’s inherent safety review process at various points in the process life cycle, Union Carbide’s index based system for inherent safety reviews, and the software tool developed by Sandoz to assist chemists and engineers in identifying hazards and inherently safer process options. Details of available tools and techniques for inherently safer design and evaluation are discussed in CCPS (2001), Gentile et al. (2003), and Khan & Amyotte (2003a,b). Recently, Amyotte et al. (2003) have presented a study on the application of inherent safety in dust explosion risk management. The study emphasized with illustrative practical examples the use of inherent safety principles at the outset of process design or equipment selection to prevent or mitigate dust explosions.

There remains, however, the need for a systematic and easy to use tool that may answer most of the safety design questions. Khan et al. (2003) have presented an analysis of available tools for risk assessment and their effectiveness in inherent safety evaluation. In another effort, Khan & Amyotte (2003a) have discussed in detail the current status of inherent safety practice in North America in general and Canada in particular. Gupta & Edwards (2002) have presented an overall picture of inherent safety and future directions needed for advancement of its use. They have identified the non-availability of effective

3

tools for inherent safety evaluation as one of the major limiting factors restricting the application of inherent safety (see also Khan & Amyotte 2002, 2003a,b). The current authors have attempted to bridge this gap through a new approach to inherent safety evaluation. This is a structured guideword based approach similar to the well-known and practiced HAZOP study procedure. The current paper presents an integrated inherent safety index (I2SI). This index is intended ultimately to be applicable throughout the life cycle of process design. The main features of I2SI are that it:

i is based on a guideword approach similar to the aforementioned, well-accepted and practiced HAZOP procedure,

ii requires details that are readily available,

iii can be worked out quickly, thus providing a swift means of hazard and inherent safety potential identification,

iv provides net scores which enable easy interpretation of results,

v enables comparison of the inherent safety potential posed by available alternatives, thus helping in design decision-making, and

vi does not require a high level of expertise to use.

INTEGRATED INHERENT SAFETY INDEX (I2SI)

Conceptual Framework

The conceptual framework of the I2SI is shown in Figure 1. This framework of I2SI was first presented at the 37th Annual Loss Prevention Symposium of the AIChE (Khan & Amyotte, 2003b) and is to be published in an upcoming issue of Process Safety Progress. To avoid repetition of information, we are not presenting a detailed description of the framework here; however, to facilitate the current discussion, a brief illustration of the essential components of the I2SI framework is presented. The current paper aims to discuss additional features of the index such as the cost model and system design model, which were not presented or discussed earlier. Additionally, application of the complete I2SI system (inherent safety as well as cost indices) is presented via a case study.

The I2SI is comprised of two main sub-indices: a hazard index (HI) and an inherent safety potential index (ISPI). The hazard index is a measure of the damage potential of the process after taking into account the process and hazard control measures.

The inherent safety potential index, on the other hand, accounts for the applicability of the inherent safety principles (or guidewords) to the process. The HI is calculated for the base process (any one process option or process setting will be considered as the base operation or setting), and remains the same for all other possible options. The HI and ISPI for each option are combined to yield a value of the integrated inherent safety index (I2SI) as shown in Equation (1).

(1)

4

Both the ISPI and HI range from 1 to 200; the range has been fixed considering the minimum and maximum likely values of the impacting parameters. This range gives enough flexibility to quantify the index. As evident, an I2SI value greater than unity denotes a positive response of the inherent safety guideword application (i.e. an inherently safer option). The higher the value of the I2SI, the more pronounced the inherent safety impact.

In order to evaluate alternate process options (entire process) for the same product, one needs to have a value of I2SI for the complete system. This can be calculated using Equation (2).

(2)

where i represents the process unit, and N is the total number of process units.

Safety Indexing Procedure

The conceptual framework and sequence of steps involved in the computations of the HI and ISPI are discussed is detail in Khan and Amyotte (2003b). A brief description of each is presented in the following sections.

Hazard Index (HI)

The hazard index (HI) is comprised of two sub-indices: a damage index (DI) and a process and hazard control index (PHCI). The damage index is a function of four important parameters, namely: fire and explosion, acute toxicity, chronic toxicity, and environmental damage. The DI is computed for each of these parameters using the curves in Figures 2a-c and Figures 3a-c, which effectively convert damage radii to damage indices by scaling up to 100.

Figures 2a-c were developed for the scenarios of fire and explosion, and toxic release and dispersion for acute as well as chronic cases, for a range of chemicals using the SWeHI approach (Khan et al., 2001). Within the SWeHI methodology, the damage radii term, DR, incorporates DR1 for fire and explosion hazards and DR2 for toxic hazards. The quantification of DR1 (fire and explosion damage radii) and DR2 (toxic damage radii) is done through a series of steps, which are detailed elsewhere (Khan et al., 2001). A brief outline of the steps is presented here.

Quantification of damage radii for fire and explosion hazards

Important steps involved in quantifying DR1 are as follows:

i) Classification of the various units in the plant into three main categories.

ii) Evaluation of energy factors.

iii) Assignment of penalties.

iv) Estimation of damage radii (DR1), which in the SWeHI methodology is termed as B1.

According to the first step, the various process units involved in the chemical process industries have been classified into three different groups: i) storage units, ii) units

5

involving physical operations such as heat transfer, mass transfer, phase change, pumping and compression, and iii) units involving chemical reactions. Subsequent steps are unique to each unit and are carried out as per the SWeHI methodology (Khan et al., 2001).

Quantification of damage radii for acute and chronic toxicity hazards

The parameter DR2 quantifies the toxic load (acute and/or chronic) over an area in terms of the radius of the area (in meters) that is lethally affected by a toxic load having a 50% probability of causing fatality. This parameter is similar to the toxic damage index B2 of the SWeHI system and is derived using transport phenomena and empirical models based on the quantity of chemicals involved in the unit, the physical state of the chemicals, the toxicity of the chemicals, the operating conditions, and the site characteristics (Fowcett & Wood, 1993; Tyler et al., 1996).

The estimation of DR2 is done with a core factor and several penalty factors. The core factor (G in SWeHI terminology) is dependent on the release conditions such as the state of the chemicals involved and the plant operating conditions. Penalties are assigned based on the chemical characteristics, operating conditions and surrounding (ambient) characteristics. Finally, the penalties and the core factor are combined to give DR2 (see Khan et al., 2001 or Khan & Amyotte, 2003b for G factor and penalty calculations).

Quantification of damage index for environmental damage

The DI calculation for environmental damage (Figures 3a-c) is adapted from the previous work of Khan et al. (2004a). Here, the environmental damage index is characterized by considering the impairment impact for air, water and soil. Monographs were developed for each environment (air, water, and soil) as shown in Figures 3a-c. It may be seen in Figures 3a-c that for each environment, there are three lines classified as A, B, and C. This classification was done to account for the characteristics of the released chemicals according to the NFPA (National Fire Protection Association) rankings for toxic/corrosive and reactive chemicals (NFPA, 1992). Chemicals having an NFPA ranking less than 2 were classified as A, 2 and 3 as B, and 4 as C. These three DI’s – one for each environment – are combined through Equation (3) to give the DI for environmental damage.

(3)

The damage indices computed for each of these effects (fire and explosion (fe), acute (ac) toxicity, chronic (ch) toxicity and environmental (en) impairment) are combined as per Equation (4) to give the final DI.

(4)

Process and Hazard Control Index (PHCI)

The other sub-index, PHCI, is calculated for various add-on process and hazard control measures that are required or are present in the system. The framework of the PHCI calculation is given in Khan and Amyotte (2003b). Briefly, this index is quantified

6

subjectively based on a mutually agreed scale among process safety experts. The index ranges from 1 to 10 for any control arrangement and is quantified based on the necessity of this control arrangement in maintaining safe operation. This necessity is divided into nine groups as listed in Table 1. For any given control system, based on necessity (from Table 1), an index (PHCI) may be derived from Figure 4. This process is repeated for all possible control systems. The PHCI’s of these different control systems are finally combined through Equation (5) to give the final PHCI.

(5)

where p stands for pressure, t for temperature, f for flow, l for level, c for concentration, iv for inert venting, b for blastwall, fr for fire resistance wall, s for sprinkler system, and d for forced dilution.

Finally, the DI is divided by the PHCI to calculate a value of hazard index (HI) as shown in Equation (6).

(6)

Inherent Safety Potential Index (ISPI)

Similar to the Hazard Index (HI), the inherent safety potential index (ISPI) is comprised of two sub-indices: an inherent safety index (ISI) and a second process and hazard control index (PHCI). To quantify the ISPI, the first step is to calculate an inherent safety index (ISI). Again, the framework for this procedure has been previously discussed by Khan and Amyotte (2003b).

The ISI computation follows the same procedure as a HAZOP study in which guidewords (in the present case, inherent safety principles such as minimization, substitution, etc.) are applied to the process system. Based on the extent of applicability and the ability to reduce the hazard, an index value is computed for each guideword. Figures 5a-d show index values for various ranges of applicability for four of the guidewords. To decide on the abscissa values (extent of applicability and ability to reduce the hazard), guidelines are developed as reported in Table 2. For the guideword simplification – where the current authors experienced difficulty in quantifying this subjective parameter – the index value can be decided using the guidelines presented in Table 3 (which admittedly are subjective themselves).

Figures 5a, b and d present the characterization of the ISI for minimization, substitution and limiting of, respectively. In the case of attenuation, there are three main operating parameters identified that control the attenuation process: i) temperature, ii) pressure, and iii) toxicity/corrosiveness of the chemicals. Monographs were developed and are shown in Figure 5c for these operating parameters; the index value may be estimated based on the extent of applicability of the guideword to these operating conditions. It is worth mentioning that the extent of attenuation guideword applicability in the case of toxicity may be measured in terms of reduction in LC50 values (LC50intial/LC50changed). The final inherent safety index for attenuation may be estimated by combining the three indices through Equation (7).

7

(7)

Subsequently, the ISI values calculated for the different guidewords (as defined in Figure 5) are combined through Equation (8) to give a final overall ISI.

(8)

After analyzing the applicability of inherent safety principles (guidewords), there may still be a requirement to install add-on process and hazard control measures. These are accounted for using the same procedure as utilized in the hazard index (HI) calculation procedure and discussed in an earlier section. This index is quantified subjectively based on a mutually agreed scale among process safety experts. The index ranges from 1 to 10 for any control arrangement and is quantified based on the necessity of this control arrangement in maintaining safe operation. This necessity is divided into nine groups as listed in Table 1. For any given control system, based on necessity (from Table 1), an index (PHCI) may be derived from Figure 4. This process is repeated for all possible control systems. The PHCI’s of these ten different control systems are finally combined through Equation (5) to give the overall PHCI.

Finally, the inherent safety potential index (ISPI) is computed in a manner similar to the hazard index (HI), by dividing the ISI with the PHCI as shown in Equation (9).

(9)

Cost Indexing Procedure

In order to account for the financial aspects of inherent safety, a cost indexing system was developed. The conceptual framework of the cost indexing procedure is shown in the right-hand side of Figure 1. This indexing system is comprised of two sub-indices: a conventional safety cost index (CSCI) and an inherent safety cost index (ISCI), the details of which are presented in the following sections.

Conventional Safety Cost Index (CSCI)

The conventional safety cost index (CSCI) is computed as shown in Equation (10).

(10)

The numerator, CConvSafety, is the sum of the costs of process control measures and add-on (end-of-pipe) safety measures (i.e. CConvSafety = CControl + CAdd-on). Calculation of these two components of CConvSafety is described in the sections subsequent to the following explanation of the denominator in Equation (10), CLoss.

Cost of Losses

For any given incident scenario, the loss index is comprised of four components: production loss (PL), asset loss (AL), human health loss (HHL), and environmental cleanup cost (ECC), which are governed as shown in Figure 6. These four components are summed to give the total cost of loss via Equation (11).

8

(11)

A brief illustration of each loss type is given below.

Production Loss – For a given scenario, the production loss is calculated based on production hours lost multiplied by the cost of the each production hour:

CPL = Likely downtime (hours) * Production value ($/hour).

Asset Loss – Incidents (scenarios) involving fire, explosion, or other similar events may cause loss of physical assets, such as damage to property, loss of equipment, etc. Asset loss may be simply calculated as:

CAL = Asset density ($/area) * Damage area.

It is important to note that the damage radii employed in the current work represent a 50% probability of damage.

Human Health Loss – For a given scenario, human health loss is calculated in terms of the number of fatalities/injuries and the costs associated with fatality and/or injury:

CHHL = Damage area * Population density (people/area) * Cost of fatality/injury ($).

As mentioned above, the damage radii represent a 50% probability of fatality. The authors acknowledge that there can be high degrees of subjectiveness and discomfort associated with assigning a dollar value to fatality and/or injury. While the value of a human life is immeasurable, it is possible to employ indicators such as insurance costs, rehabilitation costs, worker compensation rates, etc. Further, as the current indexing procedure finds its most widespread application in relative terms, such considerations should not introduce significant uncertainty into the results.

Environmental Cleanup Cost – The environmental cleanup cost for soil, water and air environments are calculated based on the mass or volume contaminated. The mass of contaminated soil is calculated by adopting a general soil density of 2650 kg/m3 and a depth of contamination of 0.5 m. The volume of contaminated water is calculated by considering the contaminated area multiplied by a 1-m depth of contamination; for contaminated air, the area multiplier is a height of 10 m. Thus, the individual cleanup costs of:

CSoil = Mass of contaminated soil * Cleanup cost($/mass) * NH,

CWater = Volume of contaminated water * Cleanup cost ($/volume) * NH, and

CAir = Volume of contaminated air * Dilution or cleanup cost ($/volume) * NH

are summed to yield the total environmental cleanup cost via Equation (12).

(12)

The term NH represents the NFPA rank of the chemical as related to health hazards (NFPA, 1992). To aid the user, we have developed a generalized table for cleanup costs (Table 4), based on the detailed review of remediation costs of contaminated sites conducted by Khan et al. (2004b).

Process control measure costs

9

The cost of process control measures may be calculated using Equation (13).

(13)

where Ci represents the cost of a given process control measure implemented N times, and n is the total number of control systems implemented. The cost of individual control measures may be taken from Table 5, which was developed through a detailed survey of available control devices from various suppliers. To better represent the survey data, cost is subdivided into three different categories according to the severity of operating conditions.

Class A: Process system/component operating in normal capacity/normal severity, and requiring a conventional control system; for example, control measures for steam pipes, liquid chemicals, etc.

Class B: Process system operating under high capacity/hazardous chemical/severe operating conditions, and requiring an advanced control system; for example, control measures for pressurized gases, flammable liquids, high gas/liquid flowrates, steam, etc.

Class C: Process system operating under very high capacity/ highly hazardous chemical/extremely severe operating conditions, and requiring an advanced control system; for example, control measures for liquefied gases, flammable gases, high gas/liquid flowrates, steam, handling fine dusts, etc.

Add-on safety measure costs

In a manner similar to the process control measure costs, the cost of add-on safety measures may be estimated by Equation (14).

(14)

where Cj represents the cost of a given add-on safety measure implemented N times, and n is the total number of add-on safety systems implemented. The cost of individual add-on measures may be taken from Table 6, which was developed using the same procedure as for Table 5.

Inherent Safety Cost Index (ISCI)

The inherent safety cost index (ISCI) is computed by Equation (15).

(15)

The denominator in Equation (15), CLoss, is the same as in Equation (10) for the conventional safety cost index. The numerator, CInhSafety, is the sum of the costs of inherent safety implementation, process control measures and add-on (end-of-pipe) safety measures (i.e. CInhSafety = CInherent + CControl + CAdd-on). The costs of process control and add-on safety measures are calculated using the formulations discussed in the preceding sections (recognizing, of course, that these costs may change with the implementation of

10

inherent safety measures). A method to quantify the costs for inherent safety implementation is described in the following section.

Inherent safety implementation costs

Inherent safety implementation costs are estimated based on the extent of application of the inherent safety guidewords and the costs associated with their application, as shown in Figure 7. The total cost of inherent safety implementation is represented by Equation (16).

(16)

where CM, CS, CA, CSi, and CL represent the costs of minimization, substitution, attenuation, simplification, and limitation of effects, respectively. EM, ES, EA, ESi, and EL

are the extent of applicability of the respective inherent safety principles (see Table 2).

APPLICATION OF I2SI TO ACRYLIC ACID CASE STUDY

To demonstrate the efficacy of the proposed indexing system we are revisiting the acrylic acid production case study earlier conducted by Palaniappan et al. (2002). Palaniappan et al. (2002) have used this case study to demonstrate the applicability and validation of their newly proposed inherent safety assessment tool iSafe.

There are two main process routes for acrylic acid manufacture (one- and two-step routes). The two-step process involves oxidation of propylene to produce acrolein as the first main step. Acrolein is subsequently oxidized to yield acrylic acid and water; there is also the possibility of side reactions producing carbon dioxide and water instead of acrolein or acrylic acid. These are catalytic reactions and are carried out at high temperature (250 – 320 oC) in the vapor phase. In spite of high process yields, past studies including Palaniappan et al. (2002) have confirmed that this process route is not as inherently safe as the one-step process.

The one-step process involves catalytic oxidation of propylene in the vapor phase at 190 oC and 3 atm pressure. There are two possible side reactions producing carbon dioxide and acetic acid with water. To optimize the production yield and operating conditions of the one-step process, many combinations of process units are possible. Three main options were studied by Palaniappan et al. (2002), which are revisited here. Flowsheets of the three options are presented in Figures 8a-c.

A detailed inherent safety investigation of these three process options for one-step acrylic acid production was conducted using I2SI. In order to maintain homogeneity and to facilitate comparison with the results of Palaniappan et al. (2002), the same process and operational data were used in our study. Additional information required, such as process control, add-on safety and cost data, is generic in nature and was adopted from Peters et al. (2002).

RESULTS AND DISCUSSION OF CASE STUDY

Option 1

11

Option 1 is considered as the base case; the process flowsheet for this option is given in Figure 8a. Intermediate and final results from the I2SI computations for the different units of this option are presented in Table 7. It can be seen that among all the units, the only one having an I2SI value greater than unity is the compressor (light shading in Table 7) – thus signifying that the compressor unit is inherently safer. This is further evident in the low value of the hazard index (HI) and the relatively higher value of the inherent safety potential index (ISPI). The reactor has the lowest value of I2SI, mainly due its high hazard index and comparatively low inherent safety potential index. The high hazard index is due to propylene and oxygen handling at high temperature and pressure. Further, there is a high probability of side reactions (which are not inherently safer features). The other units – distillation columns (particularly column III), absorber and mixer – have similar values of I2SI. It may be noted that the feed mixer is handling highly flammable propylene in significant quantities, leading to a high hazard index value for this unit.

Because option 1 is the base case, no additional inherent safety features (guidewords) are applicable to this option. This means that the conventional safety and inherent safety cost indices are the same (Table 7). These indices (CSCI and ISCI) are greater than unity for most of the process units, signifying that the costs of the safety measures on these units are higher than the expected losses. For the reactor and absorber units, the expected loss is higher (just, in the case of the absorber) than the safety measure costs. Considering a cost index value of unity as a balance condition where safety costs equal the expected loss, most of the process units in option 1 are performing in a suboptimal manner from a financial perspective.

Option 2

Option 2 is a revised version of option 1, as shown in Figure 8b. The main revisions incorporated in option 2 are: a quench tower to reduce the temperature, change of solvent to lower the severity of the operating conditions (temperature, pressure and quantity in use), an extraction column, and a solvent mixer to optimize the use of solvent and efficiency of acid extraction. These modifications bring about a positive impact on inherent safety through the substitution, attenuation, and limitation of effects principles. At the same time, they create a negative impact on inherent safety by increasing the complexity of the process.

The results of the I2SI computations for option 2 are presented in Table 8. Compared to option 1, most of the units in option 2 have high values of I2SI (greater than unity, as denoted by light shading), highlighting the enhancement of the inherent safety features of the process. Analyzing the individual process units, it may be observed that the hazard index values for most of the units have not changed from option 1 (where applicable); it is the enhancement of the inherent safety potential index that has contributed significantly to the increase in I2SI values. However, it is important to note that there are still two units with I2SI values less than unity. These units are the feed mixer and reactor; the modifications incorporated in option 2 have no favourable or adverse effects on these units. The maximum I2SI value is observed for distillation column I, followed by the extractor, distillation column II, and the quench tower.

The results of the conventional safety and inherent safety cost indices for this option are also presented in Table 8. The two cost indices are the same for the compressor, feed

12

mixer and reactor because no inherent safety measures are implemented on these units. By incorporating inherent safety revisions, the safety cost index for distillation column I has been reduced by about one-half (ISCI of 0.50 compared with CSCI of 1.09). This, combined with a high I2SI value of 4.64 for column I, clearly demonstrates that the revised option 2 has enhanced the inherent safety of the process and also achieved a significant reduction in cost. A similar trend of lower values of ISCI than CSCI is shown in Table 8 for several other units (light shading). The solvent mixer, however, shows a different trend in cost indices. Although the solvent mixer demonstrates a degree of inherently safer operation (I2SI > 1), there is an attendant increase in cost (ISCI of 2.20 versus CSCI of 1.63) as shown by the dark shading in Table 8.

Option 3

Option 3 is again a revised version of option 1, as shown in Figure 8c. The main revisions incorporated in option 3 (over option 1) are: a quench tower, change of solvent, an extraction column, a solvent mixer, and the use of solvent recycle. As with option 2, these modifications bring about a positive impact on inherent safety through the substitution, attenuation, and limitation of effects principles. At the same time, addition of the new units and the recycle stream creates a negative impact on inherent safety by increasing the complexity of the process (re. the principle of simplification).

The results of the I2SI computations for option 3 are presented in Table 9. I2SI and the other indices for the air compressor, feed mixer, and reactor are the same as for options 1 and 2, mainly because most of the process modifications are done further downstream of the reactor section. Compared to option 1, most of the units in option 3 have high values of I2SI (greater than unity, as denoted by light shading), again highlighting the enhancement of the inherent safety features of the process. Comparing these data with option 2, it is observed that the solvent mixer and the three distillation columns in option 3 have slightly higher values of I2SI. As was the case for option 2, process units such as the reactor and feed mixer in option 3 have I2SI values less than unity, clearly demonstrating the scope of inherent safety implementation. Similar to option 2, the maximum I2SI value in option 3 is observed for distillation column I, followed by the extractor, distillation column II, and the quench tower.

The results of the conventional safety and inherent safety cost indices for this option are also presented in Table 8. As with option 2, the two cost indices are the same for the compressor, feed mixer and reactor because no inherent safety measures are implemented on these units. Again similar to option 2, inherent safety modifications bring the inherent safety cost index for distillation column I (0.49) down to about one-half the value of the conventional safety cost index for this unit (1.09). The opposite trend is observed for the splitter (ISCI of 1.48 versus CSCI of 1.30) and the solvent mixer (ISCI of 2.18 versus CSCI of 1.63). The cost index trends for option 3 are similar to option 2, and the cost index values for most of the units are of the same order as observed for option 2. This demonstrates that option 3 is financially as good as option 2 (or even slightly better).

13

SUMMARY AND CONCLUSIONS

There have been several efforts in the last decade to develop inherent safety evaluation tools (Khan & Amyotte, 2003a). Examples include the INSET tool kit sponsored by the then European Community. One of the first quantitative tools, an overall inherent safety index prototype, was proposed by Edwards and coworkers at Loughborough University, UK and subsequently revised by several researchers. This prototype aimed to evaluate process routes for inherent safety. In another attempt, Heikkila and co workers at VVT, Finland, proposed an inherent safety index for process equipment and process route selection. Subsequently, this index was revised by Gentile and coworkers at Texas A&M University and a fuzzy-based inherent safety index was proposed. Palaniappan and coworkers at the National University of Singapore have proposed an index and developed an expert system for inherent safety evaluation of process flowsheets. In another recent effort, Shah and coworkers at the Swiss Federal Institute of Technology have proposed a hierarchical approach for chemical process inherent safety evaluation. In addition to these tools, there are several programs or methods that have been adopted within different corporations to evaluate the potential for inherent safety considerations in design and operation.

Recently, Khan and Amyotte (2003b) have proposed the conceptual framework of a new quantitative index – an integrated inherent safety index (I2SI). In the current paper, the authors have discussed the complete, unique design and application of the I2SI indexing system. I2SI is built upon a framework similar to the HAZOP study procedure, in which the principles of inherent safety are applied as guidewords to assess and evaluate the inherent safety potential of each processing node. This inherent safety potential is later compared with the conventional safety arrangement applicable to the same system. Such comparison clearly highlights the relative advantages of inherently safer options, and can help in decision making. The cost index module of the I2SI system helps to assess the relative cost of inherent safety against conventional safety measures. I2SI is applicable to all stages of process design and development.

To evaluate the effectiveness of the I2SI system, it was used to study three different options of acrylic acid production. Palaniappan et al. (2002) used the same case study to evaluate their newly proposed expert system (iSafe) for the design of inherently safer processes. Considering option 1 to be the base case, the other two options (2 and 3) were found to be inherently safer than option 1, with option 3 showing the most significant inherent safety benefits. These results are similar to the findings of Palaniappan et al. (2002).

The I2SI procedure developed in the current work permits further analysis of the three options by examining both the performance of individual process units (or nodes) and the economic aspects of the options. Studying the process in detail, it was observed that in options 2 and 3, maximum inherent safety potential is realized in distillation column I followed by the acid extraction column. Still, two units (the feed mixer and reactor) in all three options have I2SI values less than unity, highlighting that these units have high hazard indices and less inherent safety potential. They thus require more conventional safety measures. In this manner, the I2SI methodology is able to provide insight into a

14

process and identify the areas or nodes where attention should be focused to prevent, minimize and control hazards.

The cost index module of I2SI has been demonstrated to provide interesting results. Considering option 1 as the base, it was observed that options 2 and 3 perform better from a cost perspective as well as being inherently safer than option 1 (with no significant difference between options 2 and 3 in terms of cost performance). Analyzing option 3 in detail, it was observed that the inherent safety features of the splitter and the solvent mixer proved to be costlier than the conventional safety measures, whereas significant inherent safety savings were observed for the distillation, extraction and absorption columns. Overall, option 3 comes out to be cost-optimal as compared to option 1 (the base case). This provides clear evidence that inherently safer options can be cost-optimal.

The authors agree that there are certain aspects of the I2SI indexing procedure that require subjective judgment. We also agree that this indexing procedure is data intensive, although most of the data required by I2SI are available off-the-shelf and should not require much extra effort to obtain. To deal with such data intensiveness and uncertainties in the data, the use of fuzzy mathematics (as also suggested by other researchers) is planned to refine some of the mathematical treatment in I2SI. There is also additional work planned within our research group on life cycle aspects of the index methodology.

ACKNOWLEDGEMENT

The authors gratefully acknowledge the financial support of the Natural Sciences and Engineering Research Council (NSERC) of Canada.

REFERENCES

Amyotte, P.R., Khan, F.I., & Dastidar, A.G. (2003). Reduce dust explosions the inherently safer way. Chemical Engineering Progress, 98, 36-43.

CCPS (2001). Making EHS an integral part of process design. New York: Center for Chemical Process Safety, American Institute of Chemical Engineers.

Edwards, D.W., & Lawrence, D. (1993). Assessing the inherent safety of chemical process routes: is there a relation between plant costs and inherent safety? Process Safety and Environmental Protection, 71, 252-258.

Fowcett, H.H.,& and Wood, W.S. (1993). Safety and accident prevention in chemical operation (3rd ed.). New York: Wiley.

Gentile, M., Rogers, W., & Mannan, M.S. (2001). Development of an inherent safety index based on fuzzy logic. AIChE Journal, 49(4), 959-968.

Gupta, J.P., & Edwards, D.W. (2002). Inherently safer design – present and future. Process Safety and Environmental Protection, 80, 115-125.

Gupta, J.P., Hendershot, D.C., & Mannan, M.S. (2003). The real cost of process safety – a clear case for inherent safety. Process Safety and Environmental Protection, 81, 406-413.

15

Hendershot, D.C. (2000). Process minimization: making plants safer. Chemical Engineering Progress, 96, 35-40.

Khan, F.I., & Amyotte, P.R. (2002). Inherent safety in offshore oil and gas activities: a review of the present status and future directions. Journal of Loss Prevention in the Process Industries, 15, 279-289.

Khan, F.I., & Amyotte, P.R. (2003a). How to make inherent safety practice a reality. Canadian Journal of Chemical Engineering, 81(1), 2-16.

Khan, F.I., & Amyotte, P.R. (2003b). Integrated inherent safety index (I2SI): a tool for inherent safety evaluation. In Proceedings of 37th Annual Loss Prevention Symposium, New Orleans, LA, March 31-April 2, 2003 (pp. 263-291). New York: American Institute of Chemical Engineers.

Khan, F.I., Husain, T., & Abbasi, S.A. (2001). Safety weighted hazard index (SWeHI): a new user-friendly tool for swift yet comprehensive hazard identification and safety evaluation in chemical process industries. Process Safety and Environmental Protection, 79, 65-80.

Khan, F.I., Sadiq, R., & Amyotte, P.R. (2003). Evaluation of available indices for inherently safer design options. Process Safety Progress, 22, 83-97.

Khan, F.I., Sadiq, R., & Veitch, B. (2004a). Life cycle index (LInX): a new indexing procedure for process and product design and decision-making. Journal of Cleaner Production, 12, 59-76.

Khan, F.I., Husain, T., & Hejazi, R.F. (2004b). An overview and analysis of site remediation technologies. Journal of Environmental Management, 71, 95-122.

Kletz, T.A. (1985). Inherently safer plants. Plant/Operations Progress, 4, 164-167.

Kletz, T.A. (1998). Process plants: a handbook for inherently safer design. Bristol, PA: Taylor & Francis.

Mansfield, D.P. (1996), The development of an integrated toolkit for inherent SHE. In International Conference and Workshop on Process Safety Management and Inherently Safer Processes, Orlando, FL, October 8-11, 1996 (pp. 103-117). New York: American Institute of Chemical Engineers.

NFPA (1992). Hazardous materials response handbook. Quincy, MA: National Fire Protection Association.

Palaniappan, C., Srinivasan, R., & Tan, R. (2002). Expert system for the design of inherently safer processes. 2. flowsheet development stage. Industrial & Engineering Chemistry Research, 41, 6711-6722.

Peters, M.S., Timmerhaus, K.D., & West, R.E. (2002). Plant design and economics for chemical engineers (5th ed.). New York: McGraw-Hill.

Tyler, B.J., Thomas, A.R., Doran, P., & Greig, T.R. (1996). A toxicity hazard index. Chemical Health & Safety, 3, 19-25.

16

17

Inherent safety cost index

Inherent safety

potential index

Hazard index

Select process unit

Identify. chemicals in use. operating conditions. inventories. design options/alternatives

Estimate damage radii & then damage index (DI)

Estimate process and hazard control index (PHCI)

Evaluate potential of applicability of inherent safety principles to the unit

Estimate inherent safety index (ISI)

Estimate process and hazard control index (PHCI) after implementing inherent safety principles

Estimate integrated inherent safety index (I2SI)

Estimate the losses, CLoss

Estimate conventional safety cost, CConvSafety:

Control measures, Add-on safety

Estimate conventional safety cost index (CSCI)CSCI = CConvSafety/ CLoss

Estimate inherent safety cost, CInhSafety: Inherent safety, Control measures, Add-on safety

Estimate inherent safety cost index (ISCI)

ISCI = CInhSafety/CLoss

Have all units been evaluated?

Stop

No

YesFigure 1 Conceptual framework of the I2SI.

Figure 2a Damage index graph for fire and explosion. Figure 2b Damage index graph for acute toxicity.

Figure 2c Damage index graph for chronic toxicity.

18

5

15

25

35

45

55

65

75

85

95

105

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 300

Damage radius, m

Dam

age

inde

x

5

15

25

35

45

55

65

75

85

95

105

20 50 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400

Damage radius, m

Dam

age

inde

x

1

11

21

31

41

51

61

71

81

91

101

300 400 500 1000 1500 2000 3000 4000 5000 5500 6000 7000 10000

Damage radius, m

Dam

age

inde

x

0.001

0.01

0.1

1

10

100

0.0001 0.001 0.01 0.1 1 10

Dam

age

inde

x

Class A

Class B

Class C

Damage radius, km

Figure 3a Damage index graph for air pollution. Figure 3b Damage index graph for water pollution.

Figure 3c Damage index graph for soil pollution.

20

0.001

0.01

0.1

1

10

100

0.0001 0.001 0.01 0.1 1 10 100Damage radius, km

Dam

age

inde

x

Class A

Class B

Class C

0.001

0.01

0.1

1

10

100

0.001 0.01 0.1 1 10 100 1000Damage radius, km

Dam

age

inde

x

Class A

Class B

Class C

Figure 4 Monograph for process and hazard control index (PHCI).

21

1

2

3

4

5

6

7

8

9

10

11

0 1 2 3 4 5 6 7 8 9 10

Extent of requirement of a particular process or hazard control system

Proc

ess

and

Haz

ard

Con

trol

Inde

x (P

HC

I)

Figure 5a Inherent safety index for Minimization guideword. Figure 5b Inherent safety index for Substitution guideword.

Figure 5c Inherent safety index for Attenuation guideword. Figure 5d Inherent safety index for Limiting Of guideword.

1

11

21

31

41

51

61

71

81

91

101

1 2 3 4 5 6 7 8 9 10

Extent of guideword applicability

Inhe

rent

Saf

ety

Inde

x

1

11

21

31

41

51

61

71

81

91

101

1 2 3 4 5 6 7 8 9 10


Inhe

rent

Saf

ety

Inde

x

1

11

21

31

41

51

61

71

81

91

101

1 2 3 4 5 6 7 8 9 10


Inhe

rent

Saf

ety

Inde

x

1

11

21

31

41

51

61

71

81

91

101

1 2 3 4 5 6 7 8 9 10


Inhe

rent

Saf

ety

Inde

x

Temperature

Pressure

Toxicity/corrosiveness

22

Consider one process unit and identify the process involved & operating conditions

Identify chemicals and collect their physicochemical characteristics

Identify types of hazards Fire and explosion Toxic/corrosive release & dispersion Combination of both

Fire and explosion hazard

Calculate production loss (CPL)

Toxicity hazardChronic & Acute

Calculate damage radius for 50% probability of causing fatality, injury, and physical damage

Estimate likely downtime in case of each hazard and select the maximum

Environmental contamination hazard

Total loss for unit (CLoss) = CPL + CAL + CHHL + CECC

Repeat the process for all units

Calculate asset loss (CAL)

Calculate human health loss (CHHL)o Fatality/injury

Calculate environmental cleanup cost (CECC)o Soil (CSoil)o Water (CWater)

o Air (CAir)CECC = CSoil+CWater+CAir

Figure 6 Procedure to calculate losses.

23

Figure 7 Framework for inherent safety cost computation.

Consider one process unit and identify the process involved & operating conditions

Is guideword Minimization applicable?

Evaluate the extent of applicability of the guideword, EM

Assess the cost of Minimization, CM

Yes

No

Is guideword Substitution applicable?

Evaluate the extent of applicability of the

guideword, ES

Assess the cost of Substitution, CS

Yes

No

Is guideword Attenuation applicable?

Evaluate the extent of applicability of the guideword, EA

Assess the cost of Attenuation, CA

Yes

No

Is guideword Simplification applicable?

Evaluate the extent of applicability of the guideword, ESi

Assess the cost of Simplification, CSi

Yes

No

Is guideword Limiting Of applicable?

Evaluate the extent of applicability of the guideword, EL

Assess the cost of Limiting Of, CL

Yes

No

Repeat the process for all units

CInhSafety = CInherent + CControl + CAdd-on

CInherent = CM/EM + CS/ES + CA/EA + CSi/ESi + CL/EL

24

Figure 8a Process flowsheet of option 1.

Figure 8b Process flowsheet of option 2.

Figure 8c Process flowsheet of option 3.

25

Table 1 Guidelines to decide extent of requirement of control arrangementsDescription Extent of Requirement

Essential 10Very Important 9Important 8Not greatly important but required 7Required 6Requirement is moderate 5Good if available 4Requirement does not affect process 3Not required 1-2

Table 2 Guidelines to decide the extent of applicability of inherent safety guidewords.Description Extent Indicator

Completely applied and hazard eliminated 10Completely applied and most significant hazard reduced 9Completely applied and hazard reduced 8Completely applied and hazard moderately reduced 7Significantly applied and hazard eliminated 6Significantly applied and hazard reduced 5Applicable and hazard may be eliminated 4Applicable and hazard may be reduced 3May be applicable and hazard may be eliminated 2May be applicable and hazard may be reduced 1

Table 3 Guidelines to decide ISI value for guideword simplification.Description Inherent Safety Index

Process simplified to large extent and hazard eliminated 100Process simplified to large extent and most significant hazard reduced

90

Process simplified to large extent and hazard reduced 80Process simplified to large extent and hazard reduced moderately

70

Process simplified and hazard eliminated 60Process simplified and hazard reduced 50Process simplified moderately and hazard reduced 40Process simplified moderately and hazard reduced moderately

30

No significant process simplification and hazard reduced moderately

20

No significant process simplification and no substantial hazard reduction

10

26

Table 4 Range of cleanup costs for different chemical types.

PollutantCost

Soil Media (000$/ton)

Water Media (000$/m3)

Air Media (000$/m3)

Heavy metals 0.10-0.30 2.0-4.0 3.0-5.0Metals 0.07-0.15 1.5-3.0 1.0-2.0Organic solvents 0.05-0.08 1.0-2.0 0.5-1.5Inorganic solvents 0.01-0.05 0.5-1.5 0.5-1.0

Table 5 Classification of process control measure costs.Control system Cost (000$)

Class A Class B Class CPressure control 2-4 4-9 9-15Temperature control 1-3 3-6 6-12Flow control 3-6 6-11 11-18Level control 2-5 5-9 9-12pH control 1-3 3-6 6-12Additional control system (density control, concentration control, etc.)

2-5 5-11 11-19

Table 6 Classification of add-on safety measure costs.Control system Cost (000$) of one unit

Class A Class B Class CAlarms 0.5-1.5 2-4 4-11Detectors 2-3.5 4-8 9-20Firefighting equipment 6-10 10-20 21-30Blastwall 5-9 10-16 16-25Sprinkling system 3-5 5-15 15-25Inert gas blanketing system 4-10 10-17 18-30Fire resistance wall 4-8 9-15 15-30Other safety measures 3-7 8-14 14-32

27

Table 7 Integrated inherent safety index and cost indices for option 1 (base case).Main

Process Steps/Units

Damage Index,

DI

Process &

Hazard Control Index,PHCI1

Inherent Safety Index,

ISI

Process &


Hazard Index,HI =

DI/PHCI1

Inherent Safety

Potential Index,ISPI =

ISI/PHCI2

Integrated Inherent

Safety Index,I2SI =

ISPI/HI

Compressor 7.2 43 10 43 0.17 0.23 1.39Feed mixer 29.4 56 10 56 0.53 0.18 0.34Reactor 47.4 92 10 92 0.52 0.11 0.21Absorber 30.4 68 10 68 0.45 0.15 0.33Distillation column I

22.1 63 10 63 0.35 0.16 0.45

Distillation column II

20.6 58 10 58 0.36 0.17 0.49

Distillation column III

25.5 58 10 58 0.44 0.17 0.39

Cost IndicesMain

Process Steps/Units

Expected Loss ($),

CLoss

Conventional Safety Cost ($),

CConvSafety = CControl + CAdd-on

Inherent SafetyCost ($),


Inherent Safety Cost

Index,ISCI =

CInhSafety/CLoss

Conventional Safety Cost

Index,CSCI =

CConvSafety/CLoss

Compressor 1.91E+04 23000 0 + 11500 + 11500 = 23000

1.20 1.20

Feed mixer 4.31E+04 44500 0 + 17500 + 27000 = 44500

1.03 1.03

Reactor 8.45E+04 63000 0 + 17500 + 45500 = 63000

0.75 0.75

Absorber 5.23E+04 51500 0 + 15500 + 36000 = 51500

0.98 0.98

Distillation column I

5.50E+04 60000 0 + 17500 + 42500 = 60000

1.09 1.09


3.19E+04 56500 0 + 17500 + 39000 = 56500

1.77 1.77


2.62E+04 33000 0 + 12000 + 21000 = 33000

1.26 1.26

28

Table 8 Integrated inherent safety index and cost indices for option 2.Main

Process Steps/Units

Damage Index,

DI

Process &



ISI

Process &


Hazard Index,HI =

DI/PHCI1

Inherent Safety


ISI/PHCI2

Integrated Inherent

Safety Index,I2SI =

ISPI/HI

Compressor 7.2 43 10 43 0.17 0.23 1.39Feed mixer 29.4 56 10 56 0.53 0.18 0.34Reactor 47.4 92 10 92 0.52 0.11 0.21Quench tower

24.5 56 55.8 51 0.44 1.09 2.50

Absorber 30.4 68 55.8 57 0.45 0.98 2.19Acid extractor

30.4 68 73.2 47 0.45 1.56 3.48


22.1 63 73.3 45 0.35 1.63 4.64

Solvent mixer

19.6 42 27.7 42 0.47 0.66 1.41


20.6 58 55.7 58 0.36 0.96 2.70


26.5 51 47.7 45 0.52 1.06 2.04

Cost IndicesMain

Process Steps/Units

Expected Loss ($),

CLoss






Index,ISCI =

CInhSafety/CLoss


Index,CSCI =

CConvSafety/CLoss

Compressor 1.91E+04 23000 0 + 11500 + 11500 = 23000

1.20 1.20

Feed mixer 4.31E+04 44500 0 + 17500 + 27000 = 44500

1.03 1.03

Reactor 8.45E+04 63000 0 + 17500+ 45500 = 63000

0.75 0.75

Quench tower

3.51E+04 40500 6250 + 12000 + 13500 = 31750

0.90 1.15

Absorber 5.23E+04 51500 6250 + 9500 + 23000 = 38750

0.74 0.98

Acid extractor

4.69E+04 38500 5416 + 9000 + 17000 = 31416

0.67 0.82


5.50E+04 60000 7611 + 8500 + 11500 = 27611

0.50 1.09

Solvent mixer

1.44E+04 23500 7666 + 7500 + 16500 = 31666

2.20 1.63

Distillation 3.19E+04 56500 10083 + 8500 + 1.01 1.77

29

column II 13500 = 32083Distillation column III

2.62E+04 33000 7428 + 6500 + 12500 = 26428

1.01 1.26

30

Table 9 Integrated inherent safety index and cost indices for option 3.Main

Process Steps/Units

Damage Index,

DI

Process &



ISI

Process &


Hazard Index,HI =

DI/PHCI1

Inherent Safety


ISI/PHCI2

Integrated Inherent

Safety Index,I2SI =

ISPI/HI

Compressor 7.2 43 10 43 0.17 0.23 1.39Feed mixer 29.4 56 10 56 0.53 0.18 0.34Reactor 47.4 92 10 92 0.52 0.11 0.21Quench tower

24.5 56 55.8 51 0.44 1.09 2.50

Absorber 30.4 68 55.8 57 0.45 0.98 2.19Splitter 21.1 51 22.1 51 0.41 0.43 1.05Acid extractor

30.4 68 65.7 47 0.45 1.40 3.13


22.1 63 74.4 45 0.35 1.65 4.71

Solvent mixer

19.7 42 28.4 42 0.47 0.68 1.44


20.6 58 59.2 58 0.36 1.02 2.87


26.5 51 49.2 45 0.52 1.09 2.10

Cost IndicesMain

Process Steps/Units

Expected Loss ($),

CLoss






Index,ISCI =

CInhSafety/CLoss


Index,CSCI =

CConvSafety/CLoss

Compressor 1.91E+04 23000 0 + 11500 + 11500 = 23000

1.20 1.20

Feed mixer 4.31E+04 44500 0 + 17500 + 27000 = 44500

1.03 1.03

Reactor 8.45E+04 63000 0 + 17500+ 45500 = 63000

0.75 0.75

Quench tower

3.51E+04 40500 6250 + 12000+ 13500 = 31750

0.90 1.15

Absorber 5.23E+04 51500 6250 + 9500 + 23000 = 38750

0.74 0.98

Splitter 1.85E+04 24000 6416 + 7500 + 13500 = 27416

1.48 1.30

Acid extractor

4.69E+04 38500 6500 + 8000 + 13500 = 28000

0.60 0.82


5.50E+04 60000 6711 + 8500 + 11500 = 26711

0.49 1.09

31

Solvent mixer

1.44E+04 23500 7416 + 7500 + 16500 = 31416

2.18 1.63


3.19E+04 56500 7833 + 8500 + 13500 = 29833

0.94 1.77


2.62E+04 33000 7028 + 6500 + 11500 = 25028

0.96 1.26

32

faisal i. khan-3

Documents