Three-Stage ISD Matrix (TIM) Tool to Review the Impact of Inherently Safer Design Implementation

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Contents lists available at ScienceDirect

Process Safety and Environmental Protection journal homepage: www.elsevier.com/locate/psep

Three-Stage ISD Matrix (TIM) Tool to Review the Impact of Inherently Safer Design Implementation Mardhati Zainal Abidin a , Risza Rusli a,∗ , Azmi Mohd Shariff a , Faisal Irshad Khan b a

Chemical Engineering Department, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610, Perak Darul Ridzuan, Malaysia b Safety and Risk Engineering Group, Faculty of Engineering and Applied Science, Memorial University, St. John’s, NL A1B 3X5, Canada

a r t i c l e

i n f o

a b s t r a c t

Article history:

Inherently safer and friendlier plant design offers a simpler, cheaper, safer solution that

Received 15 May 2015

consumes less energy, requires less maintenance, and produces less waste and pollution.

Received in revised form

It is a solution that the chemical industry needs to continually adopt in the years ahead.

2 October 2015

Nevertheless, obtaining an inherently safer process/technology with respect to all potential

Accepted 6 October 2015

hazards is quite unfeasible and may lead to conflicts in the alternative process selection.

Available online 23 October 2015

To resolve safety conflicts, thorough understandings of all the hazards associated with the

Keywords:

ing inherently safer design alternatives using a combination of three-stage ISD matrix tool

Three-stage ISD matrix

and guide word approach. The proposed methodology was applied to the ammonia supply

process options are vital. This paper presents a systematic screening procedure for review-

Trade-off

system with the objective to understand the trade-off of inherent safety toward the overall

ISD conflict

process. The results show that the proposed tool is capable of helping users understand the

Qualitative ISD review

impact of modification toward the safety and implementation cost.

Systematic screening procedure

© 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Inherent safety

1.

Introduction

In 1970, Kletz introduced a theory of inherently safer design (ISD), which intends to eliminate or reduce hazards. The theory is based on four principles: eliminate/substitute, minimize, moderate, and simplify. Historically, the “missed ISD opportunities” in chemical process industries (CPI) have led to major accidents such as the Bhopal and Flixborough explosion, where the consequences were devastating. The Flixborough explosion in 1974 was among the most notable cases that have spurred the needs for ISD in CPI. The accidental release of flammable and combustible liquids from a reactor piping system led to the occurrence of vapor cloud explosion and fire (Mannan, 2004; Sanders, 2003). In this case, high-inventory processes can be avoided by increasing the

∗

reaction and conversion rate via a mixing process as well as by properly sizing the piping system (Kletz, 2001). Another dreadful disaster in CPI is the Bhopal accident, which happened due to the release of a large amount of methyl isocyanate (MIC) from a storage tank. This accident has been repeatedly cited as an accident that could have been prevented through the implementation of ISD (Edwards, 2005; Etowa et al., 2002; Khan and Abbasi, 1999). There were few reports that showed the Bhopal facilities were heavily dependent on engineered and procedural systems that were not maintained properly, thus the system were inefficient to prevent accidents from happening (Gupta, 2002; Wiley et al., 2006). Learning from these past accidents, and in light of the incident at Bayer Corp Science facilities involving MIC in 2008, the Chemical Safety and Hazard Investigation Board (CBS) requested a committee of

Corresponding author. Tel.: +605-3687567; fax:+605-3656176. E-mail address: [email protected] (R. Rusli). http://dx.doi.org/10.1016/j.psep.2015.10.006 0957-5820/© 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

31

Fig. 1 – The business case for safety by design (Hale et al., 2007).

independent experts to re-evaluate the carbamate pesticides production process using MIC that led to the implementation of ISD via a minimization concept in 2010 (Bayer Crop Science, 2012). Nowadays, companies such as Bayer, Dow, and Exxon Chemicals recognize the importance of ISD and have incorporated these strategies into their safety management programs (Srinivasan and Natarajan, 2012).

1.1.

Overview on ISD conflict

The implementation of ISD should be done in a hierarchical manner where the first-order inherent safety involves the step to avoid or eliminate hazard, and when the first order of the inherent safety is not applicable, the second-order inherent safety will be considered. The implementation of the secondorder inherent safety consists of two steps: severity reduction and likelihood reduction (Moore et al., 2008). Although the concept of ISD seem to be simple, many factors need to be considered during ISD selection such as the trade-off issues from ISD implementations, e.g., performance, environment, the conflict between inherent safety principles, the conflict between hazardous properties, and business and economic factors (Khan and Amyotte, 2003). It should also be noted that making a facility inherently safer does not automatically reduce the risk. If such measure involves reducing the chemical or physical hazards of an operation, this usually translates into a lower severity of consequences if a loss event occurs. Since the risk is a function of both severity of consequences and likelihood, any changes that increase the likelihood of a loss event more than it reduces its potential severity event would actually increase the overall risk (Center for Chemical Process Safety, 2011). The conflict between ISD principles has been discussed extensively in literature. For example, certain minimization technologies for ISD can also steer to other potential problems such as operating with higher energy inputs, higher temperatures, the requirement of more complex process/control system, and instability issues (Etchells, 2005). An example

of minimization principle given by the Center for Chemical Process Safety (2010) shows that although a continuous reactor is a safer choice compared with batch reactor by reducing the impact of accidents, it relies heavily on controller instrumentation, thus it should be considered inherently less safer. Another work by Luyben and Hendershot (2004) provides an example of a situation in which minimizing the reactor size will lead to a more aggressive response and cause instability problems for the controller. A large deviation in process variables can push the process into unsafe regions of operation and affect the product quality, thus the requirement for controllers and safety measures of the process will be higher than the original one. Other examples of an ISD conflict including selection between volatile and toxic solvent, and selection of ammonia-based and chlorofluorocarbon refrigerant (Hendershot, 2011, 2006, 1995). Based on the above discussion, it is concluded that selecting the best ISD option is a challenging task. A broad perspective is required and an overly narrowed focus on one or two high visibility concerns may result in selecting a technology that does not represent the best overall inherent safety balance. More than that, economic and business factors should be considered in selecting an ISD alternative. While the objective of ISD is to strive toward a safer and cheaper design, there is a crucial balance between tangible and intangible costs and benefits that shape the company’s decisions (Fig. 1) (Hale et al., 2007). Sometimes, striving toward a safe design requires additional costs. Whereas inherently safer and friendlier plants are often cheaper than hostile ones, considering the former lifetime costs of a process and its operation, this is not true for all cases (Table 1). In the case of reactor intensification that was discussed, conflicts will arise due to ISD implementation via minimization with a simplification concept, and this will give a direct impact toward the cost. While implementing the reactor intensification concept will reduce capital and accident costs, the cost for added safety equipment will be higher in order to ensure a smooth process. It is important to highlight that costs-added safety equipment do not solely come from

32

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 1 – Cost saving for inherently safer principles implementation (Kletz and Amyotte, 2010). Feature

Effect on cost saving

Intensification

Large

Substitution

Moderate

Attenuation

Moderate

Limitation of effects Simplification Avoiding knock-on effects Layout

Moderate

Open construction Week roof tank

Moderate

Large

Negative

Nil

Incorrect assembly impossible Status clear

Nil

Nil

Tolerance

Modest

Ease of control

Moderate

Software

Nil

Reason

Smaller equipment and less need for added safety equipment Less need for added safety equipment Less need for added safety equipment Less need for added safety equipment Less equipment

More land needed; some increase in cost Building not needed Safer design no more expensive than bad Good design usually no more expensive than bad Good design usually no more expensive than bad Fixed pipe cheaper than hoses or bellows Less control equipment needed; less maintenance Good design usually no more expensive than bad

the vast amount invested for installation, but the costs are incurred from maintenance, design, and application of a complex system (Bernechea and Viger, 2013). Thus, a wise decision must be made to ensure the selection of ISD is beneficial from both the safety and economic point of view.

1.2.

Overview on hazard review for ISD

The first step in resolving inherent safety conflicts is to ensure a thorough understanding of all the hazards associated with the process options. All hazards must be identified, and their potential consequences must be understood. Therefore, several studies have highlighted the importance of incorporating ISD concepts at the hazard review stage. In normal practices, the hazard review tools can be used to understand and to identify the hazard. Table 2 summarizes the features of the tools that can be used for process hazard review, and the applicability of the tools for different stages of the process lifecycle is shown in Table 3. While some of the tools are suitable to be used for inherent hazard review, the remaining techniques for hazard review have been modified and used for inherently safer review (Table 2). A summary of available review techniques is presented in Table 2, in which, among the limitations include the fact that the techniques require experienced practitioners and they are not inherently structured. The lacking of a formal, structured, systematic questioning procedure similar to HAZOP for hazard review at the earlier stage has also been highlighted by Kletz (1999). One of the tools that was developed for ISD evaluation based on a guide word approach similar to the HAZOP procedure is the integrated inherent safety index (I2SI) (Khan and

Amyotte, 2005, 2004). The guide word used in the I2SI methodology is based on the extent of applicability and the ability of five ISD principles: minimization, substitution, attenuation, simplification, and limiting of effect to reduce the hazard. There are twenty guide words used in I2SI, e.g., applicable and hazard may be eliminated, significantly applied and hazard eliminated, no significant process simplification and no substantial hazard reduction, etc. I2SI is a tool that considers the life cycle of the process with economic evaluation and hazard potential identification for each ISD option. While this tool offers a quick solution that can help in decision-making, there are certain aspects that require the expertise and judgment from a practitioner. Different approaches can be proposed for ISD implementation based on four main principles, thus it can lead to a broad suggestion under each category that are bound to the subjectivity issue. While the subtle judgment that hazard analysts and process experts make while performing hazard evaluations can often be the driving force behind the results, unrecorded judgments will hinder the knowledge transfer process for future users. To resolve the above limitations, a systematic screening procedure to review the impact of ISD modification in an explicit and structured manner is proposed. This tool combines the well-known interaction matrix and guide word techniques in order to identify the trade-off of ISD modification toward the risk (both severity and probability) and the cost. The main features of this tool are:

1) Three sets of interaction matrices were developed based on an extended concept of inherent safety heuristic (ISH) suggested by Rusli and Shariff (2010). The elimination procedure similar to Delvosalle et al. (2006) was adopted in this tool with the objective to systematically guide the user identification of ISD variable based on four main ISD principles. 2) The objective of this tool is to help the user understand the impact of ISD trade-off in a qualitative manner. This tool can be used as a compliment to more detailed hazard evaluations such as risk-based methodology. 3) In this tool, the user must record the impact of ISD modification in terms of risk and economy. These features will ensure that the analysts highlight their assumptions when documenting their work so future users can identify the places where additional research is necessary to obtain better hazard information or data. As organizations gain experience in using these approaches, they will appreciate the assumptions made during a study that are as important as any of the results.

2.

Methodology

The process flowchart of the systematic screening procedures for reviewing inherently safer design alternatives using a combination of three-stage ISD matrix tool and guide word approach is shown in Fig. 2. The difference between this tool compared with HAZOP and I2SI are shown in Table 4. The procedure comprises five main steps: hazard identification for base case, suggestion of alternative designs to reduce the hazard, comparison of alternative design toward the base case using the TIM Tool, list of guide words, and review stage. The objective of hazard identification in the first step is to obtain the information about the hazard for the base case, which involves two key tasks: (1) identification of specific

Table 2 – Hazard review tools (Center for Chemical Process Safety, 2011; Crowl and Louvar, 2002; Gould et al., 2000). Description

Advantages

Concept hazard analysis (CHA)

A technique that adopts the literature review of previous incidents to identify the areas of the process of specific concern during the concept and early design stages. A technique to identify the main hazards present (i.e., hazards associated with the chemicals present) during the concept and establish all criteria that the plant must adhere to fulfill specific legislations. Technique to evaluate hazards early in the life of a process by formulating a list of hazards, the potential causes, effects, and possible corrective and/or preventive measures. Brainstorming techniques to formulate a series of questions (such as what, when, how, and where) precursor to HAZOP to identify the strengths and weaknesses of a system A written list of items or procedural steps to verify the status of a system

Good basis for a more detailed study

Concentrates only on major hazards

Aids in the production of a more inherently safe process

Good basis for further studies

Concentrates only on major hazards

Aids in the production of a more inherently safe process

• Systematically identifies the accident scenarios. • Easy to perform • Good basis for further studies Can be used as a start for more detailed techniques

• Will only identify and examine the major hazards • Requires experienced practitioners

Aids in the production of a more inherently safe process

• Needs experienced practitioners • Hazards can be missed

Aids in the production of a more inherently safe process

Pre-HAZOP

Technique to identify potential hazards in the design and development phase

An inherent safety checklist has been provided by the Center for Chemical Process Safety (2011, 2010), which prompts a user to consider alternative means for reducing the hazard level inherent in the process by answering a set of questions that was developed based on five ISD principles. Inherent hazard analysis was built using the same concept but with lesser number of questions. Aids in the production of a more inherently safe process

HAZOP

Technique that was developed to identify and evaluate safety hazards in a process plant and to identify operability problems.

• Relatively quick • Low level of detail required • Good basis for further studies • Systematic and comprehensive technique • Examines the consequences of the failure

• No checklist can anticipate every potentially hazardous situation. • Checklists are limited by author’s experience. • The most thorough checklists also tend to be the longest, which make them very tedious to complete. • Not easy to apply to novel processes. • Unable to identify all hazards • Concentrates on the major hazards only

Inherent safety HAZOP-inherent safety strategies of substitute, minimize, moderate, and simplify can be used as guide word. The integrated inherent safety index (12SI) tool (Khan and Amyotte, 2005, 2004) that is applicable for the life cycle of process was developed using this concept.

What if

Brainstorming approach in which a group of experienced people familiar with the subject process ask questions or voice concerns about possible undesired events

• Easy to apply • More likely to uncover unique or unexpected hazards in the process

• Time consuming and expensive • Additional guide words are required for unusual hazards • Requires experienced practitioners • Focuses on one-event causes of deviation only • Requires a detailed source of information • Not inherently structured • Lack of skill/experience will cause the important hazards to be overlooked • Time consuming for complex processes

What if/checklist

Combines the creative, brainstorming features of the what-if analysis method with the systematic features of the checklist analysis method

Capitalizes on the strengths and compensates for the individual shortcomings of the separate approach

Concept safety review

Preliminary hazard analysis (PreHA)

Critical examination of system safety (CEX) Checklist

• Can be customized for a particular process/company • Easy to apply • A simplistic assessment that can be performed by inexperienced practitioners

Disadvantages

Relies on the experience of the checklist author and the experience of the review team

Application/modification for ISD review

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Review technique

What if inherent safety reviews is a brainstorm process deviation scenario that identifies inherent safety improvements for reducing/eliminating the potential for the scenario to develop. The ISD strategies can be used to determine the safety improvement. Combination of brainstorm process deviation scenarios and inherent safety checklist to identify chance for safety improvement

33

34

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 3 – Applicability of hazard review techniques (Center for Chemical Process Safety, 2011; Crowl and Louvar, 2002; Gould et al., 2000). Design stage/techniques Concept Process Design Commissioning Operation Modification Decommissioning

CHA

Concept safety review

PreHA

䊉 䊉 䊉 䊉 䊉 䊉 䊉  䊉              Rarely used or inappropriate

CEX

  䊉 䊉 䊉 䊉 䊉

Fig. 2 – Process flowchart for ISD screening procedure.

undesirable consequences and (2) identification of material, system, process, and plant characteristics that can produce those consequences. The second step involves the proposal of alternative design to reduce the hazard. Proposals of alternative designs were done based on the standard industrial practice and available patent/literature.

Checklist

What if

䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 䊉 Commonly used

What if-checklist

Pre-HAZOP

HAZOP

䊉 䊉 䊉 䊉 䊉 䊉 䊉

䊉 䊉 䊉    

  䊉 䊉 䊉 䊉 䊉

The third step is comparison of alternative designs toward the base case using the TIM Tool. The objective of this step is to identify the list of ISD variables for process alternatives that deviate from the base case. A detailed framework of screening process using the TIM Tool is shown in Fig. 3. The three-stage matrix used for the review process consists of matrix ISD heuristic–ISD guide word, matrix ISD guide word–ISD indicator, and matrix ISD indicator–ISD variable (Tables 5a–5c). These matrices were developed based on a workflow of an extended concept of ISH as suggested by Rusli and Shariff (2010) and based on a classification given by the Center for Chemical Process Safety (2010) and Kletz and Amyotte (2010). In order to use this tool effectively, experience and deep understanding of inherent safety concepts are essential. The screening process using the TIM tool involves five general steps starting with matrix ISD heuristic–ISD guide word (Table 5a): Step1: Screening the parameters under ISD heuristic (A1–A3 in Table 5a) one by one. Irrelevant parameters in row (A1–A3) were eliminated in this step. Step 2: After the elimination of irrelevant parameters from the ISD heuristic category, empty columns under the ISD guide word (B1–B4) were eliminated. Step 3: The remaining parameters under the ISD guide word (B1–B4) after the elimination process in Step 2 were screened one by one. Irrelevant parameters in column (B1–B4) were eliminated. Step 4: The remaining parameters under the ISD guide word (B1–B4) after the elimination process in Step 3 will be transferred to the next matrix (matrix ISD guide word–ISD indicator) (Table 5b). Step 5: Step 2–4 will be repeated in sequence for Stage 2 (matrix ISD guide word–ISD indicator) as shown in Table 5b and for Stage 3 (matrix ISD indicator–ISD variable) as shown in Table 5c. The final output for this step is the list of the ISD variable involved and the value. Three guide words—higher, lower, and change phase—will be assigned for each ISD variable based on their deviation from the base case.

Fig. 3 – Flow of screening process using TIM Tool.

35

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 4 – Comparison of traditional HAZOP, I2SI with the TIM Tool. HAZOP

I2SI

Stage

Suitable for detailed design stage

Purpose of guide word

To identify the deviation of process parameters for one selected process Several guide words—typically no/not/none, more, less, part of, reverse, other than, as well as are combined with parameters (flow, pressure, temperature, reaction, level, composition).

Guide word

Applicable throughout the life cycle of process design To identify the extent of applicability and the ability of five ISD principles to reduce the hazard Twenty guide words (e.g., completely applied and hazard eliminated, completely applied and most significant hazard reduced, completely applied and hazard reduced, etc.)

The final step involved the reviewing process of alternative designs. Based on the guide word of ISD variable that had been listed in the previous step, the assessor must identify the impact of ISD modification toward certain criteria. For examples, in this paper, the impact of ISD modification toward safety level and economic is required. Therefore, there were two key questions that needed to be answered by the assessor. While reviewing the safety impact, the assessor must answer “How does the deviation of the ISD variable affect the severity and accident probability?”. While reviewing the economic impact, the assessor must answer “What is the cost affected by the changes using this principles?”.

3.

TIM Tool Suitable for early/preliminary design stage To identify the deviation of the ISD variable for process alternatives from the base case Three guide words (higher, lower, and change phase) combined with the ISD variable

(over 99% NH3 ). Table 6 summarizes the results of the first step, which is to identify hazards via two criteria: identification of undesirable consequences, and identification of material properties, system, process, and plant characteristics. The data used in this step can be found in Study (2006). In order to meet process requirements as well as to reduce the risk potential, two alternatives of the ammonia supply system were suggested (Table 7). Alternative 1 using aqueous ammonia with 23 wt% NH3 in liquid form and alternative 2 using anhydrous ammonia in vapor form. An example of the three-stage ISD matrix usage is shown in Fig. 4, particularly to identify the ISD variable for the first alternative with respect to the base case. The steps for the systematic screening procedure (Section 2) to identify the ISD variable that compares alternative 1 to the base case is shown in Fig. 4 (a–i) and explained as below: Step 1: The screening process for variable A1–A3 in Fig. 4a was done one by one. Referring to the process alternative information (Table 7), the use of ammonia is inevitable and cannot be replaced by the safer material. Therefore, the approach to reduce the hazard via ISD heuristic A1 (hazard elimination) is implausible, thus it was excluded.

Case study

The ability of the three-stage ISD matrix to review design alternatives will be demonstrated using a case study that had been discussed by Study (2006) regarding the selection of the ammonia supply system to the selective catalytic reactor (SCR). A base case for the ammonia supply system involves the transportation of liquefied anhydrous ammonia via a pipeline. Anhydrous ammonia is pure ammonia

Table 5a – Matrix ISD heuristic–ISD guide word. ISD heuristic

ISD guide word

Hazard elimination Consequence reduction Likelihood reduction

A1 A2 A3

B1

B2

B3

B4

Substitute X

Minimize

Moderate

Simplify

X

X X

Table 5b – Matrix ISD guide word–ISD indicator. ISD guide word

B1 B2 B3 B4

Eliminate/ substitute Minimize Moderate Simplify

ISD indicator C1

C2

C3

C4

C5

C6

C7

Hazardous substance X

Process Route

Inventory

Energy

Process condition

Complexity

Loss of containment

X

X

X

X X

X

36

X X X X X X X

Catalyst

X X

X

Dilution Pressure Temperature

X X X C2 C3 C4 C5

Hazardous substance Process route Inventory Energy Process condition Complexity Loss of containment

New safer raw material

New safer solvent

New safer process chemistry

X X

Volume

X X

Process phase

X X

New equipment

D8

C1

C6 C7

Process layout Resistant material Min no of unit/ utilities Strength of equipment

D14 D13

Table 6 – Hazard identification for the base case. Hazards identification

D7 D2 D1

ISD indicator

Table 5c – Matrix ISD indicator–ISD variable.

D3

D4

D5

D6

ISD variable

D9

D10

D11

D12

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Material Undesirable consequences Material properties, system, process, and plant characteristics

Description

Anhydrous liquid ammonia High toxicity level exposure of vapors or liquid has the potential to cause serious injury or fatality Material properties

Color: colorless Standard: gas Relative density, gas: 0.6 (air = 1) Relative density, liquid: 0.7 (water = 1) Vapor pressure: 124 psi at 20 ◦ C (68 ◦ F) Boiling point: 33 ◦ C Solubility in water: completely soluble Percent volatility: 100% Lower explosive limit (LFL): 15% Upper explosive limit (UFL): 30% Immediately dangerous to life and health (IDLH): 300 ppm Acute exposure guideline levels (AEGL) for 10-min exposure durations AEGL-1: 30 ppm AEGL-2: 270 ppm AEGL-3: 2700 ppm System/Process Liquefied ammonia in pipeline

Step 2: The elimination of A1 from the ISD heuristic category, causes the related variable under the ISD guide word to become irrelevant (appeared as empty column) in Fig. 4b. Therefore, the empty columns under the ISD guide word (B1) were eliminated. Step 3: The remaining parameters under the ISD guide word (B2–B4) were screened one by one to identify implausible ISD approaches. In this case, there was no elimination because as can be seen from the alternative design information (Table 7), minimize (B2), moderate (B3), and simplify (B4) are plausible (Fig. 4c). Step 4: The remaining parameter from Fig. 4c under the ISD guide word (B2–B4) were then transferred to the next matrix (matrix ISD guide word–ISD indicator) (Table 5b) as shown in Fig. 4d Step 5: Steps 2–4 were repeated in sequence for (matrix ISD guide word–ISD indicator) (Table 5b) for Stage 2 (Fig. 4d–f) and matrix ISD indicator–ISD variable (Table 5c) for Stage 3 (Fig. 4g–i). The lists of ISD variable generated by Stage 3 are shown in Table 8. Three guide words (higher, lower, and change phase) were assigned to the list of ISD variables based on the deviation of parameter from the base case (Table 8). These information will be used as input for the reviewing stage. The final outcome of this tool (the review stage) for alternative 1 and alternative 2 are shown in Table 9 and Table 10, respectively. As shown in Tables 9 and 10, the outcome of the review process is the answer for (1) “How the deviation of the ISD Variable will affect the severity and accident probability?” and (2) “What is the cost affected by the changes using this principles?”. The estimation of base-case cost is shown in Table 11 and the cost of ISD implementation is shown in Tables 9 and 10. In

37

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 7 – Ammonia supply system options (Study, 2006). Options Base case Alternative 1 Alternative 2

Piping length (m) Anhydrous ammonia liquid Aqueous ammonia (23 wt% NH3 ) Anhydrous ammonia vapor

NH3 mass (kg)

Pump

Utilities 45.36 kg/h steam Vaporizer skid 2267.96 kg/h steam Vaporizer skid Low-pressure steam

182.88

235.87

–

609.60

272.16

1

609.60

4.54

–

this paper, the approach used to estimate the value of ISD implementation cost was adopted from Khan and Amyotte (2005, 2004) supported with the cost data available in Peters et al. (2004). As can be seen in Tables 9 and 10, the evaluation of process alternative 1 and alternative 2 using the three-stage ISD

matrix has resulted in two main outputs: (1) review on the impact of modification on safety that involves severity and probability and (2) review on the impact of modification on capital cost (equipment) and operating cost (utilities and raw materials). Compared to the base case, the inherently safer features in alternative 1 are classified as a moderation action,

Fig. 4 – TIM tool implementation for alternative 1.

38

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 8 – List of guide word for ISD variable deviation. Options Base case

ISD variable Anhydrous ammonia liquid

Alternative 1

Aqueous ammonia (23 wt% NH3 )

Alternative 2

Anhydrous ammonia vapor

Value

Keyword

Volume (inventory)

235.87 kg

–

Volume (energy) Process phase Dilution No of unit/utilities Process layout Volume (inventory) Volume (energy) Dilution No of unit/utilities Process layout Volume (inventory)

45.36 kg/h Liquid 99% NH3 – 182.88 m 272.16 kg 2267.96 kg/h 23% NH3 1 pump 609.60 m 4.54 kg

Higher Higher Higher Higher Higher Lower

Volume (energy) Process phase Dilution No of unit/utilities Process layout

– Vapor Steam dilution No vaporizer skid 609.60 m

Lower Change phase Higher Lower Higher

which involves dilution of higher concentrations of ammonia (anhydrous to aqueous ammonia with 23 wt% NH3 ). However, in this alternative, the amount of ammonia to be handled is larger than the amount for the base case. Moderation is the reverse of intensification, for if the conditions were made less extreme, a larger inventory may be needed. Comparing to the base case, the cost for a piping system for alternative 1 is higher than the cost for the base case, an increase from $3.46E + 03 to $1.15E + 04. In alternative 1, the system will become more complex with the positive displacement pump that is required for the aqueous ammonia supply. For the base case and alternative 1, a vaporizer skid was used to convert liquefied ammonia into vapor form prior to injection into SCR. The dilution of ammonia causes a higher requirement of energy for vaporization. Therefore, among these three options, alternative 1 required a larger amount of steam (2267.96 kg/h) compared with the amount required for the base case (45.36 kg/h), with annual cost noted as $1.26E + 05 and $2.51E + 03, respectively. These results are in agreement with the data given by Chemiton Enterprise, who concluded that aqueous ammonia require higher evaporation energy than anhydrous ammonia (Table 12). In the second alternative, anhydrous ammonia was supplied to the SCR in vapor form. This alternative offers the reverse of inherently safe features compared with alternative 1, where a smaller amount of ammonia will be handled in vapor form, thus reducing the impact of its release to the surrounding. However, the cost of the piping system for alternative 2 is similar to the cost for alternative 1, which is $1.15E + 04 as a 0.05 m pipe was used for all three options due to structural integrity concerns. As the ammonia was supplied in vapor phase, the requirement for a vaporization process prior to injection in the SCR was eliminated, thus automatically offering a simpler process with lower energy requirements. Another inherently safe feature in alternative 2 via moderation is the use of lower pressure steam as a diluent to ensure even dispersal of ammonia before injection into the SCR. While alternative 2 requires redundant instrumentation due to the fact that the mass flow measurement for vapor is more difficult to control compared to liquid flow, it offers cost reduction via the elimination of ancillary equipment such as

pump and vaporizer skid as well as their operating cost. The comparison of cost for the three process options is plotted in Fig. 5. As can be observed in Fig. 5, alternative 2 gives the lowest cost ($ 1.36E + 04), followed by the base case ($5.16E + 04), and finally due to the trade-off, alternative 1 poses the highest cost ($ 1.61E + 05). Whereas in this work, the proposed tool can help the assessor understand the impact of ISD modification on the capital and operating costs, therefore, the decision must be made carefully. Although, alternative 2 offers several inherently safer features with the lowest ISD implementation cost, at certain threshold quantity anhydrous ammonia is classified as a hazardous material, and is subject to strict regulations and risk management procedures for transport, storage, and handling. These requirements result in additional costs and complications in obtaining permission and may generate local community concerns on transporting hazardous materials. Based on the above discussion, it can be concluded that the trade-off between ISD principles can give a significant impact toward the ISD implementation cost. It is worthy to note that while the tools can help the assessor evaluate the impact of ISD modifications, experience and good engineering judgments are vital for the decision-making process.

Fig. 5 – Cost comparison of three ammonia supply options.

Table 9 – Review of alternative 1 for ammonia supply system. TIM tool

Deviation of ISD Variable

Impact toward safety

Impact toward economics

Higher material volume

Higher inventory, larger hazard zone of material release

Capital cost–equipment cost

Stage 1: ISD guide word minimization Stage 2: ISD indicator inventory Stage 3: ISD variable volume Input: ISD heuristic consequence reduction

Higher energy volume

–

Operating cost–utility cost for evaporation Operating cost–utility cost for pump operation

Stage 1: ISD guide word minimization Stage 2: ISD indicator energy Stage 3: ISD variable volume Input: ISD heuristic consequence reduction

Higher dilution

Lower pressure difference between the storage system and the outside environment, reducing the rate of release in case of a leak

Costly

ISD implementation cost 1.15E + 04

1.26E + 05 3.24E + 02

Operating cost–aqueous ammonia cost

Low cost

8.65E + 03

Capital cost–equipment cost Capital cost–equipment cost

Moderately costly

1.89E + 03

Stage 1: ISD guide word moderation Stage 2: ISD indicator process condition Stage 3: ISD variable dilution

Input: ISD heuristic likelihood reduction

Higher no of unit

Stage 1: ISD guide word simplification

Higher no of unit

Stage 2: ISD indicator complexity Stage 3: ISD variable no of unit/utilities Input: ISD heuristic likelihood reduction Stage 1: ISD guide word simplification Stage 2: ISD indicator complexity Stage 3: ISD variable process layout Total ISD implementation cost

Higher complexity of process layout

Lower initial atmospheric concentration of the hazardous material, smaller hazard zone downwind of the spill Additional pump, increase the failure probability Vaporizer skid for ammonia evaporation process, increase failure probability

Additional piping length, increase failure probability

1.31E + 04

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Input: ISD heuristic consequence reduction

ISD implementation cost

Capital cost–equipment cost – Costly

1.61E + 05

39

40

Table 10 – Review of alternative 2 for ammonia supply system. Deviation of ISD variable

Input: ISD heuristic consequence reduction

Lower material volume

Lower inventory, smaller hazard zone of material release

Capital cost–equipment cost

Stage 1: ISD guide word minimization Stage 2: ISD indicator inventory Stage 3: ISD variable volume Input: ISD heuristic consequence reduction

Lower energy volume

–

No utility cost for evaporation

Stage 1: ISD guide word minimization Stage 2: ISD indicator energy Stage 3: ISD variable volume Input: ISD heuristic consequence reduction

Change phase

Gas: larger hazard zone of material release

Operating cost–anhydrous ammonia cost

Stage 1: ISD guide word moderation Stage 2: ISD indicator process condition Stage 3: ISD variable process phase Input: ISD heuristic consequence reduction

Higher dilution

Lower initial atmospheric concentration of the hazardous material, smaller hazard zone downwind of the spill

Operating cost–utility cost for dilution by steam

Stage 1: ISD guide word moderation Stage 2: ISD indicator process condition Stage 3: ISD variable dilution Input: ISD heuristic likelihood reduction

Lower no of unit

Removal vaporizor skid, lower failure probability

No capital cost–equipment cost

Higher complexity of process layout

Additional piping length, increase failure probability

–

Stage 1: ISD guide word simplification Stage 2: ISD indicator complexity Stage 3: ISD variable no of unit/utilities Input: ISD heuristic likelihood reduction Stage 1: ISD guide word simplification Stage 2: ISD indicator complexity Stage 3: ISD variable process layout Total ISD implementation cost

Impact toward safety

Impact toward economics

ISD implementation cost Moderately costly

ISD implementation cost 1.15E + 04

–

Low cost

6.26E + 02

1.44E + 03

–

–

–

Moderately costly

1.36E + 04

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

TIM tool

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Table 11 – Base-case cost. Cost

Value ($)

Piping cost Vaporizer skid Total capital cost Steam Raw material Total operating cost

3.46E + 03 1.31E + 04 1.65E + 04 2.51E + 03 3.26E + 04 3.51E + 04

Table 12 – Energy requirement for ammonia evaporation process. Ammonia concentration Anhydrous ammonia Aqueous ammonia (28 wt% NH3 ) Aqueous ammonia (19 5 wt% NH3 )

4.

kJ/kg NH3 1.37E + 03 7.35E + 03 1.10E + 04

Conclusion

Although the ISD philosophy itself is a methodology to achieve fundamentally safer plants, conflicts between principles can deviate from their original intention. This paper proposes a systematic procedure to review a process alternative using a combination of three-stage ISD matrix tool and guide word approach. Through the proposed procedure, the related parameter was identified systematically in hierarchical order: ISD heuristic, ISD guide word, ISD indicator, and ISD variable. The key of the review process using this tool are two questions that needed to be answered as the ISD variable deviated from the base case. This shows that this tool is capable of helping users understand the impact of the ISD trade-off and can be used as a compliment to more detailed hazard evaluation such as risk-based methodology. The documentation of the impact evaluation in this tool will ensure that analysts highlight their known assumptions so as to ensure a knowledge transfer within companies and future users. In this work, the issues related to the conflict between four ISD principles and their influence to the safety and cost has been demonstrated and discussed using an ammonia supply system alternative. It is worthy to note that ISD is not simply an isolated improvement, but it is a part of the total package toward having a more economically viable, safer, and environmentally benign aim of the CPI.

Acknowledgments The authors would like to thank Universiti Teknologi PETRONAS for providing the facilities and the Ministry of Higher Education Malaysia for their sponsorship under MyBrain15 program.

References Bayer Crop Science, 2012. The Use and Storage of Methyl Isocyanate (MIC) at Bayer CropScience. The National Academies Press, Washington, D.C. Bernechea, E.J., Viger, J.A., 2013. Design optimization of hazardous substance storage facilities to minimize project risk. Saf. Sci. 51, 49–62, http://dx.doi.org/10.1016/j.ssci.2012.06.007. Center for Chemical Process Safety, 2011. Guidelines for Hazard Evaluation Procedures, 3rd ed. John Wiley & Sons, New York. Center for Chemical Process Safety, 2010. Inherently Safer Chemical Processes: A Life Cycle Approach. Wiley, New York.

41

Crowl, D.A., Louvar, J.F., 2002. Chemical Process Safety: Fundamentals with Applications, 2nd ed. Prentice Hall PTR, New Jersey. Delvosalle, C., Fievez, C., Pipart, A., Debray, B., 2006. {ARAMIS} project: a comprehensive methodology for the identification of reference accident scenarios in process industries. J. Hazard. Mater. 130, 200–219, http://dx.doi.org/10.1016/j.jhazmat.2005.07.005. Edwards, D.W., 2005. Export inherent safety {NOT} risk. J. Loss Prev. Process Ind. 18, 254–260, http://dx.doi.org/10.1016/j.jlp.2005.06.014. Etchells, J.C., 2005. Process intensification: safety pros and cons. Process Saf. Environ. Prot. 83, 85–89, http://dx.doi.org/10.1205/psep.04241. Etowa, C.B., Amyotte, P.R., Pegg, M.J., Khan, F.I., 2002. Quantification of inherent safety aspects of the Dow indices. J. Loss Prev. Process Ind. 15, 477–487, http://dx.doi.org/10.1016/S0950-4230(02)00039-6. Gould, J., Glossop, M., Ioannides, A., 2000. Review of Hazard Identification Techniques. Health and Safety Laboratory, Sheffield. Gupta, J.P., 2002. The Bhopal gas tragedy: could it have happened in a developed country? J. Loss Prev. Process Ind. 15, 1–4, http://dx.doi.org/10.1016/S0950-4230(01)00025-0. Hale, A., Kirwan, B., Kjellén, U., 2007. Safe by design: where are we now? Saf. Sci. 45, 305–327, http://dx.doi.org/10.1016/j.ssci.2006.08.007. Hendershot, D.C., 1995. Conflicts and decisions in the search for inherently safer process options. Process Saf. Prog. 14, 52–56, http://dx.doi.org/10.1002/prs.680140109. Hendershot, D.C., 2006. An overview of inherently safer design. Process Saf. Prog. 25, 98–107, http://dx.doi.org/10.1002/prs.10121. Hendershot, D.C., 2011. Inherently safer design: an overview of key elements. Prof. Saf., 48–54. Khan, F.I., Abbasi, S.A., 1999. Major accidents in process industries and an analysis of causes and consequences. J. Loss Prev. Process Ind. 12, 361–378, http://dx.doi.org/10.1016/S0950-4230(98)00062-X. Khan, F.I., Amyotte, P.R., 2003. How to make inherent safety practice a reality. Can. J. Chem. Eng. 81, 2–16, http://dx.doi.org/10.1002/cjce.5450810101. Khan, F.I., Amyotte, P.R., 2004. Integrated inherent safety index (I2SI): a tool for inherent safety evaluation. Process Saf. Prog. 23, 136–148, http://dx.doi.org/10.1002/prs.10015. Khan, F.I., Amyotte, P.R., 2005. I2SI: A comprehensive quantitative tool for inherent safety and cost evaluation. J. Loss Prev. Process Ind. 18, 310–326, http://dx.doi.org/10.1016/j.jlp.2005.06.022. Kletz, T.A., 1999. The constraints on inherently safer design and other innovations. Process Saf. Prog. 18, 64–69, http://dx.doi.org/10.1002/prs.680180112. Kletz, T., 2001. Learning from Accidents, 3rd ed. Butterworth-Heinemann, Oxford. Kletz, T., Amyotte, P., 2010. Process Plants A Handbook for Inherently Safer Design, 2nd ed. Taylor & Francis Group, Boca Raton. Luyben, W.L., Hendershot, D.C., 2004. Dynamic disadvantages of intensification in inherently safer process design. Ind. Eng. Chem. Res. 43, 384–396, http://dx.doi.org/10.1021/ ie030266p. Mannan, S., 2004. Lees’ Loss Prevention in the Process Industries: Hazard Identification Assessment and Control, 3rd ed. Elsevier Science & Technology Books, Oxford. Moore, D.A., Hazzan, M., Rose, M., Heller, D., Hendershot, D.C., Dowell, A.M., 2008. Advances in inherent safety guidance. Process Saf. Prog. 27, 115–120, http://dx.doi.org/10.1002/prs.10238. Peters, M.S., Timmerhaus, K.D., West, R.E., 2004. Plant Design and Economics for Chemical Engineers. McGraw-Hill Education, New York. Rusli, R., Shariff, A.M., 2010. Qualitative assessment for inherently safer design (QAISD) at preliminary design stage. J.

42

Process Safety and Environmental Protection 9 9 ( 2 0 1 6 ) 30–42

Loss Prev. Process Ind. 23, 157–165, http://dx.doi.org/10.1016/j.jlp.2009.07.005. Sanders, R.E., 2003. Designs that lacked inherent safety: case histories. J. Hazard. Mater. 104, 149–161, http://dx.doi.org/10.1016/S0304-3894(03)00241-3. Srinivasan, R., Natarajan, S., 2012. Developments in inherent safety: a review of the progress during 2001–2011 and opportunities ahead. Process Saf. Environ. Prot. 90, 389–403, http://dx.doi.org/10.1016/j.psep.2012.06.001.

Study, K., 2006. A real-life example of choosing an inherently safer process option. Process Saf. Prog. 25, 274–279, http://dx.doi.org/10.1002/prs.10163. Wiley, R.J., Hendershot, D.C., Berger, S.,2006. The accident in Bhopal: observations 20 years later. In: 40th Annual Loss Prevention Symposium. American Institute of Chemical Engineers, Orlando, p. 15.