Khouloud Mesloub
| Fares Innal* | Yves Ducq
© 2023 IIETA. This article is published by IIETA and is licensed under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/).
OPEN ACCESS
Emergency response plans play a key role in limiting the consequences of major accidents and consequently preventing them from causing domino effects. It is therefore crucial to efficiently design and implement emergency response plans according to the expected accidents. Within this framework, this paper is aiming to present a structured approach in order to model and evaluate the performance of such plans, based on IDEF0 and BPMN (Business Process Modeling Notation) methods. In fact, the IDEF0 allow a detailed functional and structural description of the emergency response plan, whereas the BPMN is used to clarify the relations between its different components and to simulate it. The simulation results give valuable information regarding the execution of the emergency response process, especially the required time to reach a safe situation. The proposed approach was illustrated on a special emergency plan called “Internal Intervention Plan: IIP” related to a gasoline storage leakage that may lead to a major accident scenario (fire) within an LNG facility.
emergency response plans, Internal Intervention Plan (IIP), IDEF0, BPMN, modeling, simulation
Major accidents in gas and oil industry, such as fire, explosion, release of hazardous materials, etc., cause huge losses for potential targets (people, property and the environment). This reality is attributable to the large quantities and dangerous properties of the materials used and to the severe operating conditions (high temperature, pressure, etc.). In addition to the inherent safety and supplementary safety barriers intended to prevent or control major accidents, effective emergency response plans play a key role in preventing a major accident from causing a secondary accident and thus potentially forming a domino effect. In order to ensure the rapid, orderly and effective implementation of emergency response procedures, emergency response plans are usually developed in advance in process plants. These plans arrange in prior the emergency organization’s structure, personnel, technology, equipments, materials, procedures, orders and coordination.
To help establishing such plans, many guidelines have been published to assist the work of emergency responders in handling emergency situations in the chemical industry, such as the Centre for Chemical Process Safety (CCPS) [1], the U.S Federal Emergency Management Agency (FEMA) [2], the Oil and Chemical Industries Safety Studies Group in France (GESIP) [3], UK Health and Safety Executive (HSG191) [4], the European Union ‘Seveso-III’ directive (Council Directive) [5], and Incident management system for oil and gas industry [6].
The evaluation of emergency response actions has become an urgent requirement for improving emergency preparedness, since emergency actions in response to an accident can significantly affect its evolution. Different patterns of an accident evolution may lead to different consequences. As a result, the efficacy of the emergency response actions and their impact on both the evolution of the accident and the likelihood of the ensuing consequences should be effectively taken into account in the risk analysis [7].
Some studies on the evaluation of emergency response plans have been carried out [8-13]. Most of these researches aimed to evaluate the emergency plans from robustness and integrity perspectives [7]. However, the relevant studies in the field of emergency response actions are still relatively limited. For example, regarding risk assessment, Antonioni et al. [14] applied recently developed equipment damage probability models to identify end scenarios and assess the likelihood of escalation. Khakzad et al. [15] introduced a Bayesian network-based methodology to model risk patterns with a domino effect in order to estimate the probability of its impact at different levels. The probabilities of events can be updated in light of new information. Based on probabilistic models and physical equations, Kadri et al. [16] presented a methodology for the quantitative assessment of the hazards and domino effects of fire and explosion in storage areas. A human exposure model to the effects of excess pressure and heat radiation was developed to estimate individual and societal risks. In addition, Khakzad et al. [15] introduced a dynamic outcome analysis approach to evaluating risks and managing their impacts and demonstrated the application of Bayesian networks and conflict analysis to allocate risk-based chemical inventories to minimize consequences and reduce the potential for escalation.
For the escalation thresholds, Cozzani and Salzano [17] studied the definition and assessment of overpressure threshold values for the damage to equipment caused by blast waves which are originated by primary accidental scenarios, and they proposed threshold values for different categories of process equipment, taking into account either damage levels or release intensities following the loss of containment. Cozzani et al. [18] further studied the revision and the improvement of criteria for escalation credibility, based on the modeling of explosion, fire or release of toxic gas damage to process equipment due to different escalation vectors, and proposed revised threshold values. Landucci et al. [19] developed an approach for the quantitative assessment of the risk caused by escalation scenarios triggered by major accidents like explosion, fire or release of toxic gas.
Major accident, toxic gas clouds, overpressure waves and heat radiation effects do not delay to claim their toll. Therefore, emergency planning as mitigation measure plays a key role in reducing the risk of accidents by avoiding fatalities and injuries, protecting the environment and accelerating the resumption of normal operations [20].
The occurrence of major accidents in the oil field around the world is a proof that the risk is omnipresent. Hence, the control of major accidents requires a global approach, allowing, on the basis of the analysis of the feedback, to enrich the reflection, already started on how to adopt major risks in the area of constituency and the most effective means to implement both in the prevention stage and in emergency management and contingency planning [21].
Internal Intervention Plan (IIP) is a specific emergency response plan and it is regarded as a major safety barrier in the emergency response stage. It is a grouping of all measures to be implemented in the event of a major industrialaccident in order to prevent or reduce the associated adverse effects in the shortest possible time [21].
There are a large body of published studies in terms of assessments of internal intervention plan and emergency response performance for chemical accidents. Some studies have used quantitative analytical methods. For example, Karagiannis [22] presented a function-interaction-structure (FIS) modelling approach to measure the robustness of an internal or external industrial emergency plan.
Thanh et al. [23] simulated and analyzed the tsunami emergency plans from a model perspective. An earthquake and tsunami resolution plan was represented as a Business Process Model and Notation (BPMN) approach.
In this paper, we carried out a structural/ functional modeling of the Internal Intervention Plan (IIP) adopted in the oil and gas industries in Algeria. The IIP is triggered in the event of a major accident detected such as fire, explosion and toxic release. The rest of this paper is organized as follows. Section 2 and 3 details the different aspects of an IIP in the context of Algerian oil and gas industry. section4 describes the proposed methodology based on a combination of IDEF0 and BPMN methods. An illustration of that approach is given in section 5 the structural modeling of the IIP is performed using the IDEF0 technique, whereas the functional modeling and simulation of the IIP is carried out thanks to the BPMN approach. A wrap up of this study is given in section 6.
The IIP is established to define the organization of relief and disaster response within a given industrial facility. It aims to protect the personnel, the public and the immediate environment and describes the measures to be taken to restore the safety of the facility and to prevent the disaster from becoming more widespread [24]. The IIP is based on the risk analysis. It defines the conditions for managing the accident and its consequences. It describes according to the major accident scenarios, the organization of the alert, the rescue and the response. More specifically, for each scenario, the IIP defines the strategy of the intervention, the first interventions, the course of the attack and the means required for each phase of the intervention. Risks of escalation of the disaster are identified, the action and precautions to be taken are reported.
2.1 Missions considered in the IIP
The IIP defines a 3-level emergency organization (the response teams as shown in Figure 1, an Operational Command Post, an Internal Operations Management Post) set up internally to establish the interface with the other plans (PAM, ORSEC, Crisis, etc.) and also the various third parties involved in the accident. It specifies the structure of this organization, the exact role of each stakeholder and the hierarchical links between the different entities involved [24].
Figure 1. Communications structure
2.2 The value of an IIP
This plan is a well-defined management of major emergencies. In addition to providing guidance during an emergency, the IIP has other benefits. The development of this plan can identify and eliminate undetected risks that could aggravate an emergency situation. The planning process identifies gaps, including a lack of resources (equipment, trained personnel, supplies) that can be addressed before a situation occurs. In addition, an IIP promotes security awareness and emphasizes the organization’s commitment to the safety of its employees.
2.3 Activating the IIP
Alerting is the most important function as it triggers the IIP [25]. The latter defines the alert chain that illustrates the logic used on the site regarding the organization of communications and the mobilization of the actors involved. This chain begins with the detection of an incident by automatic means or by a witness. After the incident detection, the alert is passed on to all IIP actors who are thus mobilized. Finally, when the situation is under control, the end of the incident is reported by the appropriate level of management.
There are three approaches to evaluating the performance of IIP, namely [25]:
3.1 Assessment by feedback
Feedback is important to improve the performance of emergency response in industrial operations. In order to assess the overall effectiveness of the different measures used in the IIP, simulation exercises are essential tools. Indeed, the simulated implementation of the plan will be an opportunity for the participants to put into practice the theoretical learning, to familiarize themselves with their roles and responsibilities in an emergency situation and to validate the different procedures established in the IIP.
Improvement of IIP through feedback is generally based on reports following simulation exercises and incidents requiring activation of the IIP. Although actual accidents are the only true test of the plan, simulation exercises can also assess some or all of the IIP [25]. At the end of the simulation, all the actions carried out are analyzed to identify the causes of any non-conformities. The reports shall contain a brief narrative of the operation, accompanied by a presentation on the elements of the action which have worked well or poorly and proposals for improvement.
Feedback process identifies failures that have already occurred but does not allow for a comprehensive IIP analysis [25]. Several authors have already highlighted the need for a systemic analysis of these plans.
3.2 Audit assessment
The audit approach to assessing IIP is to directly assess the performance of the plan using questions and factors identified by a large number of experts [25].
Performance is assessed by questions focused on the quality of the plan, the consideration of all indicators defined in the methodology, and the plan’s ability to meet regulatory requirements. Scores are assigned to the answers to these questions, and a final score is produced. Despite its pragmatic and relevant nature, this method lacks of structure. In fact, the identification of factors to be assessed is based on expertise and does not take into account the structure of the IIP. Moreover, the aggregation of scores is based on a sum of the points obtained by the different questions and does not take into account the relative importance of the different aspects of the plan [24].
3.3 Evaluation of IIP using the FIS model
The National School of Mines of Saint-Etienne (France) [24]. has developed a method for the analysis of the robustness of IIP using the FIS method. This research used the computer tool (XRisk) developed for structural-functional modeling and risk analysis of complex systems. The model was used to structure the analysis and to identify potential failures during crisis management.
Each type of resource is characterized by a set of attributes (including fault trees) that allow the probability of resource failure to be calculated. The probability of failure of the model functions is calculated from the associated resource failure probabilities across the fault tree of each function. The severity of each function is estimated according to the maximum consequences of its failure on the system and its environment. The combination of the probability and the severity of failure of each function results in its criticality. The robustness of the IIP is expressed in terms of its failure criticality, which is obtained from the aggregation of the criticalities of its functions.
However, the aggregation of function criticalities does not take into account the relative importance of the different plan functions and the performance is assessed qualitatively, hence the importance of evaluating the IIP quantitatively.
The modeling of a system provides a complete and coherent model for the system functions (activities, actions, processes and operations) required by a public information chain in the event of an emergency, as well as the relationships and data supporting the integration of this system [26]. In this paper we propose a methodology based on the combined use of the IDEF0 and BPMN approaches.
4.1 The IDEF0 approach
IDEF0 is a modeling language with graphical and textual features that provides a standard template for the decisions, functions and activities of an enterprise, process or system. It is based on the Analytical and Structured Design Technique developed in 1972 by Douglas (T. Ross) and (SofTech), Inc. (Marca and McGowan, 1988). IDEF0 is now an IEEE standard (KBSI, 1998). In each diagram, the activities, processes and transformations are represented by boxes describing their function. Function-related data and objects are represented by arrows. As shown in Figure 2, the role of an arrow is determined by its relationship to a function area:
Figure 2. Actigram of the IDEF0 model
IDEF0 models are created by a bottom-up analysis process, but they have a top-down representation and interpretation. A model includes a series of hierarchical diagrams and associated materials. A higher-level context diagram, called Diagram A-0, defines the context of the model and defines its purpose and perspective. Diagram A-0 can be broken down into sub-functions represented in a child diagram, and each sub-function can again be broken down into low-level child diagrams.
4.2 BPMN approach
The BPMN “Business Process Modeling Notation” language is a standard for the modeling of a company’s business processes, allowing defining a common graphical notation for all modeling tools.
BPMN is chosen in this study because it is a powerful language of modeling that describes the flow of the process by using the proper events, activities and gateways [26].
The main objective of BPMN is to provide a common framework for describing a process in a way that is understandable to all the actors of a company, from business analysts who create the initial drafts of the procedures, up to the developers responsible for setting up the technology thatwill execute these procedures [27].
BPMN decouples business information from technical information and provides correspondence to runtime languages.
4.2.1 BPMN graphical elements
A BPMN diagram consists of a set of graphical elements that allow modelling activities, flows, relationships, data and their interactions. It is structured around four categories of elements as shown in Figure 3
Figure 3. Basic elements of a BPMN
•Flow Objects: these are the main graphical elements that define the behaviour of a business process. There are three types of flow objects:
-Activities: activities that correspond to an action that can be performed by a human or machine. They have a beginning and an end and can only begin if the previous activities are completed. An activity can be atomic or composite.
-Events: events that correspond to an action that occurs during the process. They usually have a cause and a consequence. This makes it possible to change the course of a process when a particular event occurs during the execution of the process. There are three categories of events: Start, Intermediate and Stop.
-Gate-ways: which correspond to a connection in the process and allows representing an action in the progress of the latter.
•Connecting Objects: these are the graphical elements that allow to connect the flow objects to each other to represent the flow of the process. There are three kinds of object connector: the “Sequence flow” which indicates the execution order of the actions, the “Message flow” which correspond to a link between two separate processes and finally, the “Association” which allow to link data or documents to process objects.
•Swimlanes: these are the graphical elements used to structure and group the different elements that make up the BPMN diagram. These are the partitions (Pools) and the sub-partitions (Lanes) [28].
•Artifacts: are used to provide additional information about the process. There are three types of artifacts: “Data object” that show how the data are linked to a task (via an association), “Groups” which allow to group together tasks of the same category in order to better locate them visually and finally, “Annotations” which allow to add comments to facilitate the reading of the BPMN diagram [29].
The proposed approach based on IDEF0 and BPMN is illustrated on an IIP related to a gasoline storage tank fire. This major accident scenario was identified within an Algerian LNG facility.
5.1 Gazoline storage sphere 76-MD-03
The gasoline produced by the deisopentanizer is stored in the 76-MD03 sphere as shown in Figures 4 and 5, with a capacity of 3000 m³, before sending it to the refinery [30]. Normal operating pressure is 0.51bar g. The maximum operating temperature is +62°C and the minimum temperature is +1°C. Physical characteristics of gasoline are given in Table 1.
5.2 Scenario description
The internal intervention plan is executed after the detection of a leak in the gasoline storage that has led to a major accident scenario (fire) inside the LNG installation [30], through the successive stages below the plan modeling through the IDEF0 approach and each step is developed and detailed by BPMN:
Table 1. Physical characteristics of pentane (gasoline)
|
product |
liquid |
flash point C° |
boiling point C° |
vapor pressure at 38C° (kg/cm²) |
self-ignition point C° |
flammability limit % |
||
|
density at 15°c |
vapor |
lower |
upper |
|||||
|
Pentane C5+ |
0.63 |
2.49 |
-40 |
36 |
1.1 |
309 |
1.5 |
7.6 |
Figure 4. Gasoline storage sphere 76-MD-03 [30]
Figure 5. Location of flash fire/gasoline storage tank
5.3 Modeling of the IIP using the IDEF0 approach
First, we modeled the internal intervention plan using the IDEF0 approach, which allowed a functional analysis of the IIP that is top-down, modular, hierarchical and structured. From the modeling of the IIP events, i.e., activities carried out by men and machines, we have shown the relationships that exist between them and the way they fit into a hierarchical structure when there is a major accident (alert) and the IIP is triggered, including:
Level 01 «A0» Global vision of the IIP: it shows a global vision of the IIP and presents just the inputs, the outputs, as shown in Figure 6, the used resources and the possible constraints in a general way when the major accident is detected.
Level 02 «A1»: Communication of the IIP with the other safety barriers: this second established model depicts the global management process in the major incident, as shown in Figure 6. It shows how the three levels of safety barriers communicate with each other and how the first safety barrier (Level 1/IIP) communicates with the other levels during the exacerbation of their interference in the response.
Level 03 «A2»: Model to understand: represents a hierarchical structural modeling of the IIP triggering when an alert/alarm occurs indicating an incident detection. As shown in Figure 6
Level 04 «A3»: Top-down modeling of the IIP: after obtaining the general description of the IIP (A0 and A2) and the relation with the other levels (A1), A3 diagram gives a functional modeling of the IIP (decomposition of A0), which clearly shows the relationships between all the elements, as shown in Figure 7:
5.4 Modeling of the IIP using the IDEF0 approach
Figure 6. Modeling by the IDEF0 approach of the IIP applied to the scenario studied (fire gasoline storage tank)/ Global vision
Figure 7. Modeling by the IDEF0 approach of the IIP applied to the scenario studied (fire gasoline storage tank)/Detailed vision
5.5 The BPMN model for the IIP
We used the BPMN approach to obtain a more detailed structural and functional modelling of the IIP. Then, we simulate the resulting model using bizagi software [31], a shown in Figure 8 and we obtain the Table 2 and the graph below Figure 9, which gives all the details of this plan, in particular the time necessary to carry out the task/activity for each step (examples carried out), the type of communication between these activities, and the human and material resources used necessary for each task / activity, in addition to the time of realization (key situation under control) of which:
Figure 8. Modeling of the IIP applied to the scenario studied by the BPMN method
5.6 Results obtained by BPMN
Table 2. Time needed to complete the task/activity for each stage of the IIP
|
Name |
Type |
Duration of execution of each task of the internal intervention plan of a gasoline storage tank fir (Min) |
human resource |
material resource |
|
The internal response plan and its communication with the other levels in the event of a gasoline storage tank fire |
Process |
65 |
IIP |
- |
|
Incident detection |
Start event |
1 |
- |
Main room & HSE standby room |
|
Message received |
Gateway |
1 |
- |
talky walky/Main room & HSE standby room |
|
On-site operator/Surveillance video/Automatic detection |
Task |
1 |
operator |
surveillance video/phone dialing alert number |
|
Alarm |
Intermediate event |
1 |
0 |
Main room & HSE standby room |
|
General alert |
Task |
1 |
EHS department chief & informed the derogator of OCP to trigger the PII |
Manual incident alarm boxes (MAI)are equipped with a pull or push button system |
|
Confirm and assess the situation |
Task |
1 |
Team OCP |
Fire fighting equipment with coolant |
|
Master small fires/incident |
Task |
1 |
Team OCP |
FIRE truck (water curtain applied) |
|
Parallel Gateway |
1 |
- |
- |
|
|
End of incident |
End event |
1 |
Team OCP |
MAI |
|
Inform the leader of the OCP |
Intermediate event |
2 |
Team OCP |
Fire fighting equipment with coolant |
|
Gather the OCP personal group |
Task |
3 |
Team OCP |
Fire truck (water curtain applied) |
|
Activation order |
Intermediate event |
4 |
Leader OCP |
Fire fighting equipment with coolant |
|
Triggered the IIP |
Task |
10 |
Leader OCP |
Fire fighting equipment with coolant |
|
Inclusive Gateway |
Gateway |
10 |
- |
- |
|
The rescue and incident control team assembles |
Task |
13 |
Team of rescue and incident control |
Fire truck / VS truck (water curtain applied) |
|
Deployment of the medical team |
Task |
6 |
doctor and nurse |
ambulance |
|
Gathering of heavy equipment |
Task |
7 |
heavy equipment team |
reinforce FIR material |
|
Gathering done |
Gateway |
10 |
- |
- |
|
Check the availability of cooling means |
Intermediate event |
9 |
Team of rescue and incident control |
cooling product |
|
Isolate the affected area |
Task |
8 |
Team of rescue and incident control |
- |
|
Apply cooling means |
Task |
15 |
Team of rescue and incident control |
cooling product |
|
Evacuation of injuries off site |
Task |
8 |
Medical team |
ambulance |
|
Incident extinguished |
Task |
14 |
heavy equipment team |
|
|
Estimate the state |
Gateway |
14 |
- |
- |
|
End of incident major/a gasoline storage tank fire |
End event |
65 |
Team OCP |
MAI |
|
Re-ignition |
Task |
65 |
heavy equipment team |
cooling product |
|
Event evaluation |
Task |
60 |
Team of rescue and incident control |
- |
|
Situation report |
Intermediate event |
55 |
Coordinator I |
E-mail or telephone |
|
Inform the TCP |
Task |
26 |
Coordinator I |
E-mail or telephone |
|
TCP gathering |
Task |
33 |
Teams from neighboring areas |
FIRE truck / VS truck |
|
Detailed incident report |
Intermediate event |
32 |
Coordinator II |
E-mail or telephone |
|
SCP group |
Task |
42 |
Coordinator II |
E-mail or telephone |
|
Support and advice |
Intermediate event |
22 |
Civil protection and teams from neighboring areas |
FIRE truck / VS truck |
Figure 9. Average execution time of the IIP of a gasoline storage tank fire
In this paper we proposed a coherent approach, based on a joint use of the IDEF0 and the BPMN approaches, in order to model a specific emergency response plan called Internal Intervention Plan (IIP) that is adopted in the Algerian oil and gas industries. In fact, the modeling of the IIP allows its improvement before its execution in case of a major accident. Moreover, the proposed approach has been illustrated on a real major accident scenario related to a gasoline storage tank. Precisely, the IDEF0 approach has been used to model the structural aspects of the IIP. Based on the IDEF0 model, the BPMN one has been established in order to allow a more detailed functional modeling of the IIP. In addition, this later model has been simulated in order to estimate the different execution times corresponding to the different tasks to be performed.
However, the BPMN model does not allow considering and simulating some complicated situations especially those induced by the different dependencies that may exist during tasks execution. To overcome this limitation, more sophisticated simulation approaches such as Petri nets will be used in a future work.
|
IIP |
Internal Intervention Plan |
|
OCP |
Operational Command Post |
|
TCP |
Tactical Command Post |
|
SCP |
Strategic Command Post |
[1] Guideline for Technical Planning for on-Site Emergencies: Chapter 8. AlchE, New York https://lccn.loc.gov/2016044310.pdf.
[2] FEMA (Federal Emergency Management Agency). (1996). Guide for All-Hazard EmergencyOperations Planning, State and Local Guide (SLG), vol. 101, https://www.fema.gov/pdf/plan/slg101.pdf.
[3] Petroleum and Chemical Industries Safety Studies Group. (2001). GESIP Methodological Guide for the Development of the Internal Operation Plan of a Flammable Storage Facility (Depot) or a Small Industrial Facility – Report 96/02.
[4] Emergency Planning for major accidents. Control of major accident hazardsregulations. 1999 (COMAH). Health and Safety Executive (HSE). http://www.hse.gov.uk/pUbns/priced/hsg191.pdf.
[5] Directive, C. (2012). on the Control of Major Accident Hazards involving Dangerous Substances. Off. J. Eur. Communities. Official Journal of the European Union. L197/1.Special edition in Croatian: Chapter 15, 31: 77-113.
[6] Hosseinnia, B., Khakzad, N., Reniers, G. (2018). Multi-plant emergency response for tackling major accidents in chemical industrial areas. Safety Science, P: 275. https://doi.org/10.1016/j.ssci.2017.11.003
[7] Zhou, J., Reniers, G., Khakzad, N. (2016). Application of event sequence diagram to evaluate emergency response actions during fire-induced domino effects. Reliability Engineering & System Safety, P: 202. https://doi.org/10.1016/j.ress.2016.02.005
[8] Cheng, C.Y., Qian, X. (2010). Evaluation of emergency planning for water pollution incidents in reservoir based on fuzzy comprehensive assessment. Procedia Environmental Sciences, 2: 566-570. https://doi.org/10.1016/j.proenv.2010.10.061
[9] Chen, G., Zhang, X. (2009). Fuzzy-based methodology for performance assessment of emergency planning and its application. Journal of Loss Prevention in the Process Industries, 22(2): 125-132. https://doi.org/10.1016/j.jlp.2008.10.003
[10] Lumbroso, D., Stone, K., Vinet, F. (2011). An assessment of flood emergency plans in England and Wales, France and the Netherlands. Natural Hazards, 58: 341-363. https://doi.org/10.1007/s11069-010-9671-x
[11] Karagiannis, G.M., Piatyszek, E., Flaus, J.M. (2010). Industrial emergency planning modeling: A first step toward a robustness analysis tool. Journal of Hazardous Materials, P: 324. https://doi.org/10.1016/j.jhazmat.2010.05.014
[12] Piatyszek, E., Karagiannis, G.M. (2012). A model-based approach for a systematic risk analysis of local flood emergency operation plans: A first step toward a decision support system. Natural Hazards, 61: 1443-1462. https://doi.org/10.1007/s11069-011-0079-z
[13] Zhou, J., Reniers, G., Khakzad, N. (2016). Application of event sequence diagram to evaluate emergency response actions during fire-induced domino effects. Reliability Engineering & System Safety, P: 209 https://doi.org/10.1016/j.ress.2016.02.005
[14] Antonioni, G., Spadoni, G., Cozzani, V. (2009). Application of domino effect quantitative risk assessment to an extended industrial area. Journal of Loss Prevention in the Process Industries, 22(5): 614-624. https://doi.org/10.1016/j.jlp.2009.02.012
[15] Khakzad, N., Khan, F., Amyotte, P., Cozzani, V. (2013). Domino effect analysis using Bayesian networks. Risk Analysis: An International Journal, 33(2): 292-306. https://doi.org/10.1111/j.1539-6924.2012.01854.x
[16] Kadri, F., Châtelet, E., Chen, G. (2013). Method for quantitative assessment of the domino effect in industrial sites. Process Safety and Environmental Protection, 91(6): 452-462. https://doi.org/10.1016/j.psep.2012.10.010
[17] Cozzani, V., Salzano, E. (2004). Threshold values for domino effects caused by blast wave interaction with process equipment. Journal of Loss Prevention in the Process Industries, 17(6): 437-447. https://doi.org/10.1016/j.jlp.2004.08.003
[18] Cozzani, V., Gubinelli, G., Salzano, E. (2006). Escalation thresholds in the assessment of domino accidental events. Journal of hazardous materials, 129(1-3): 1-21. https://doi.org/10.1016/j.jhazmat.2005.08.012
[19] Landucci, G., Gubinelli, G., Antonioni, G., Cozzani, V. (2009). The assessment of the damage probability of storage tanks in domino events triggered by fire. Accident Analysis & Prevention, 41(6): 1206-1215. https://doi.org/10.1016/j.aap.2008.05.006
[20] Hosseinnia, B., Khakzad, N., Reniers, G. (2018). Multi-plant emergency response for tackling major accidents in chemical industrial areas. Safety science, P:289. https://doi.org/10.1016/j.ssci.2017.11.003
[21] Djeddi, C. (2015). Évaluation de la performance des plans internes d’intervention au moyen des réseaux de Petri. Doctoral dissertation, Université de Batna 2.
[22] Karagiannis, Industrial emergency planning modeling: A first step toward a robustness analysis tool. Journal of Hazardous Materials, P: 334. https://doi.org/10.1016/j.jhazmat.2010.05.014
[23] Thanh, L.N.T., Hanachi, C., Stinckwich, S., Vinh, H.T. (2013). Representing, simulating and analysing Ho Chi Minh City Tsunami Plan by means of process models. arXiv preprint arXiv:1312.4851. https://doi.org/10.48550/arXiv.1312.4851
[24] G.-M., Piatyszek, E., and Flaus, J.-M. Industrial emergency planning modeling: A first step toward a robustness analysis tool. Journal of Hazardous Materials 2011.
[25] thodology for the analysis of the robustness of industrial contingency plans. Doctoral thesis, Ecole Nationale Supérieure des Mines de Saint-Etienne.
[26] Kherbouche, M. (2013). Contribution to the management of the evolution of business processes. PhD dissertation, University of Littoral Côte d’Opale.
[27] Khelifaoui, M. (2016). L’élaboration d’un plan interne d’intervention au niveau de la station-service TAMEZGUIDA-NORD. Master thesis.
[28] Khalaf, A. (2008). Systèmes de contrôle de la qualité de production : Méthodologie de Modélisation, de Pilotage et d'optimisation des Systèmes de Production .doctoral thesis with a view to obtaining the degree of doctor in industrial engineering. University of Paul Verlaine de Metz.
[29] Rodrigues, D.P. (2017). Modelling business processes in BPMN inspired by the DEMO principles. Técnico Lisboa.
[30] Reference system for emergency and crisis management; standard internal organization plan-IOP; central management health,
[31] Bizagi software: https://www.bizagi.com.
