Natural gas is a widely consumed source of energy whose application is rapidly growing. It is also a major source of urban energy. Natural gas supplies the energy required for heating and cooking in buildings. Therefore, the wide-ranging, expansive, and highly dispersed consumption of this energy source at the urban level is an environmental pollution challenge that has so far been neglected. Natural gas often comprises methane and small amounts of heavier hydrocarbons such as ethane, propane, butane, pentane, and C6 . In some states, natural gas contains small amounts of hydrogen, CO2, nitrogen, oxygen, sulfur compounds, and water. The thermal value of natural gas depends on the percentage of hydrocarbons and inert gases in it, and lies in the 800-1200 BTU/SCF range based on its quality. The quality of natural gas differs based on its constituents’ composition percentage [2, 3].
The quality and composition of fuels directly affect many combustion properties, the most important of which are the released heat rate, burning speed, excellent energy conservation tendency, and adiabatic flame temperature. The major contaminant sources of natural gas are heavy gases, hydrogen sulfide, hydrogen, CO, and CO2. Due to the combustion of natural gas, different pollutants are released depending on the quality and composition percentage of its constituents. The most important pollutant is NOx; the quality and composition percentage of fuel constituents significantly affect the emission of pollutants, especially NOx and SOx .
The consumption of natural gas in a metropolitan such as Tehran is considerable due to the density of buildings and large residing population. In 2018, the average daily consumption of gas in Tehran Province was 77 million m3, which is equivalent to the daily production of three phases of the South Pars/North Dome Gas-Condensate field. Between November and December 2018, a record of 122 million m3 gas consumption was recorded in Tehran Province, which equals to the daily production of five phases of said field. Based on the data of the Gas Company in Tehran province, 50% of the gas sales belong to six million residential units in cities and villages; 25% (7 billion m3) to power plants; 10% (2.5 billion m3) to industries, and 15% to other sectors [5, 6]. Based on the available reports, the National Iranian Oil Refining and Distribution Company consumed 3.6 billion L of liquid fuel last year, apart from transportation; of this, 4.7 (165 million L of liquid fuel) was consumed in different residential sectors, factories, etc. It should be mentioned that ,except for transportation, 60 million L of the of liquid fuel belongs to the consumption of 400 casting workshops Tehran Province [7, 8].
Qualitative risk assessment is a reasonable method to determination of the quantity and quality of hazards as well as examination of the potential outcomes related to individuals, materials, equipment, and the environment. In fact, through risk assessment, the efficiency of existing control methods is specified, and valuable data are provided for decision-making on risk and hazard reduction, control system optimization, and responding plans .
Risk assessment is a major activity for many processes, including urban gas distribution networks for consumers . In other words, the distribution network in the urban region forms a complex and sensitive network, and the accurate identification of the risks of this network is utmost importance. Awareness of these hazards as well as their risk assessment from the viewpoint of citizens provides an insight into urban residents’ degree of comfort with regard to this distribution network and gas consumption.
Pipelines are among the most practical and safe methods to transfer the hazardous substances such as natural gas, though, this transfer method has always been exposed to different risk and hazards over time. Natural gas pipelines are often designed, built, used, and managed in an open environment. They are, therefore, exposed to various threats and unknown situations. These lines finally are connected to buildings for different purposes. In buildings, gas is mainly consumed for heating and cooking, for which different facilities are used in buildings. The major consumption of gas is heating the building. Currently, based on the technologies available in Iran, one or multiple systems (e.g. home packages (HP), central heating boiler room (CHBR), water heaters, and gas heaters) are used to supply water and air heating in the building.
Most studies on risk assessment of gas distribution networks and transfer pipelines have been conducted in urban areas. Based on the degree of importance, the large number of the studies can be justified. Some of these studies are reviewed below:
Moradi et al. assessed the risk of urban gas distribution network using analytic hierarchy process (AHP) in Sanandaj, Iran . In the assessment, due to the lack of access to all the information and their weighting criteria, the AHP technique was used in order to do the risk management of urban gas distribution network. The results showed that the interference of real persons (i.e., the sub-category of causal indicators) and substance hazard (i.e., the sub-category of outcome indicator) had the maximum weight, while the pipeline pressure indicator had the minimum weight.
Ghandehari et al. identified and quantitatively assessed the risk of urban gas lines and determined the sensitive regions by proposing an integrated model . They segmented the region’s line network using appropriate indicators and possible incident scenarios about gas leakage using failure mode and effects analysis (FMEA) and event tree analysis (ETA) techniques. They estimated the final outcomes and probabilities of these scenarios by different human, financial, and social dimensions. In the study, the optimality functions of the outcomes were extracted using the concepts of optimality theory, as well. Accordingly, the risks of sections pertaining by each scenario were estimated using the functions and the calculated probabilities as well as equal units. Eventually, different parts of the region were ranked in terms of risk.
Few studies have been conducted on risk assessment of thermal and heating facilities in buildings. For instance, Karimi Asl assessed the risk of these facilities in religious buildings using FMEA . This study tried to identify the hazards and assess the risks using FMEA in the mentioned buildings to enhance the individual and visitors’ satisfaction and comfort.
Omidvari et al. assessed the risk of medical and treatment facilities in several hospital buildings resulting from flammable facilities . In the study, a combination of fuzzy theory, multiple-criteria and decision-making models, and FMEA risk assessment method were used. Based on the results, some recommendations were made to promote the staff’s safety, confidence, and comfort resulting from the hazards of CHBR and heating facilities.
Nouri et al. examined the accidents in academic buildings in Iran. One of the factors which was considered in the study was the gas heating facilities in the studied buildings . In the study, the FMEA method was adopted, in which a set of general indicators related to the environmental, safety, and health domains of these buildings was examined based on the status quo of the visited samples. The findings, however, do not enable us to do the precise and detailed evaluation of the heating and gas facilities in the examined buildings.
To date, diverse methods have been applied for risk assessment. Based on the reviewed literature, the FMEA method is among the most widely-used risk assessment method. The FMEA model can be combined with various theories and models such as the fuzzy theory to enhance flexibility, improve uncertainty, and approach reality . To date, various studies have been combined the methods, and their results showed that there is more precision compared to the classic FMEA model [9, 16, 17, 18].
The present study aimed to assess the risk of safety and environmental incidents resulting from building heating facilities during earthquakes from the viewpoints of the residents of district 9 in Tehran. This examination provides an insight into the sustainable development of heating and thermal facilities, including HP and CHBR in the buildings in Tehran. It also provides some information about the citizens’ trust of the natural gas-related facilities, especially after earthquakes.
2. Materials and Methods
The FMEA model along with fuzzy theory was adopted in order to do the risk assessment and comparison. Fifty buildings were selected for field visits and completing FMEA questionnaires. The information collected by the questionnaires and interviews with the residents were used in this study. The buildings were randomly selected from district 9 in Tehran. It should be mentioned that because some buildings used different heating systems such as gas heaters and water heaters, they were excluded. Finally, 50 buildings with CHBR or HP were chosen. The FMEA method and fuzzy theory adopted in this study. They have been explained below:
Failure mode and effects analysis (FMEA)
FMEA is one of the most well-known methods of risk identification. The FMEA technique is a preventive and systematic method mainly aiming to determine the points and paths in which the performance of a system can be disrupted under specific and pre-defined conditions, thereby disrupting the efficiency of the entire system. In this technique, after finding these points, the reasons for these damages are examined, and eventually, their prevention is studied. In fact, this technique is used to maintain the system’s health and safety .
FMEA is a systematic tool based on the working group employed for defining, evaluating, preventing, removing, or controlling the states, reasons, and effects of potential errors in a system, process, design, or service .
The Risk priority number (RPN) is composed of three values named severity (S), occurrence (O), and discovery probability (D) [21, 22] which is estimated using the Equation 1.
Hazard severity or the degree of novelty is the “potential hazard effect” on individuals. It is considered because of its “effect”. Hazard severity can be reduced only through applying the changes in the process and performance of activities. To do so, there are some quantitative indicators ranging from 1 to 10. Table 1 presents the hazard severity .
The occurrence probability specifies the frequency with which the cause of a potentially hazardous mechanism can be defined (Table 2).
The value of occurrence might be reduced only by eliminating or reducing the causes or mechanism of each hazard. The occurrence probability ranges from 1 to 10. The assessment of previous evidences is very useful. Assessment of control processes, standards, requirements, work regulations , and their application can greatly contribute to find the value .
Discovery probability is a type of capability assessment to identify a cause or a mechanism of hazard. In other words, the discovery probability is the ability to identify the hazard before its occurrence. Like previous probability, the examination of control processes, standards, requirements, work regulations, and their application can greatly contribute to find the value (Table 3) .
One of the problems of the FMEA method is to use of a crisp space for the scoring oriented from the type of human thinking system, the impossibility of error-free decision-making, and simplifying the interviews and questionnaires. Fuzzy theory is one of the most suitable methods for further flexibility of FMEA calculations and the uncertainty improvement of the questionnaires . In this study, FMEA calculations were performed in the fuzzy space. The details have been provided below.
According to Zadeh (1965, 1975), a fuzzy number X∼ is a fuzzy set (the membership function of is denoted by μX ) which is of a universe of discourse (the real line R) which is defined as :
The triangular fuzzy number was adopted in which three real numbers (i.e., l, m, u) were used in its definition. The l, m, and u are the lower, mean and upper limits of the triangular fuzzy number, respectively. Figure 1 shows a typical triangular fuzzy number in which the three numbers were required.
3. Results and Discussion
In this section we compare and assess the risk of two conventional heating systems in Iran, namely HP and CHBR. These systems are commonly used in the majority of administrative and residential buildings. Although they have similar functions, they have some differences in safety at the time of earthquakes. Thus, using the examinations performed in this study, a better insight into the risks of each system can be provided.
During the visits of 50 residential buildings in Tehran, the status of CHBR and HP was examined; interviews with the residents were conducted to identify the criteria and effective factors on the risk of these facilities during earthquakes. In addition, the FMEA questionnaires were completed. Eleven risks were identified for each studied facilities which were listed in the Tables 5 and 6.
After being transferred to the fuzzy space, the average results of 50 questionnaires and interviews after being transferred to the fuzzy space were presented for each system separately in Tables 4, 5, and 6. The RPN values in Tables 5 and 6 were calculated based on the multiplication operator in fuzzy equations.
During the visits, the residents expressed several concerns about the heating facilities of the buildings during earthquakes. Most cases can be attributed to gas and gas transfer pipes’ leakage or explosion. Although the main gas transfer stations in Tehran are equipped with valve shutdown facilities during natural disasters such as earthquakes, urban residents are highly concern about gas leakage and explosion. Some others are concern about other equipment such as tanks and pumps. Earthquakes can cause numerous accidents related to gas equipment in residential buildings.
Secondary gas pipes’ breakage in the building is a concern which is originated from two systems examined in this study. Of course, the path of secondary pipes is longer in buildings with HP than others which means that they run a higher risk of breakage in the buildings. It should be mentioned that the gas pipe junctions in the building are shorter in buildings with CHBR.
As HPs are usually installed at a high location and their connection is not robust against earthquakes, they may be separated from the wall and fall down. The distance between water or gas equipment and HPs are rarely long enough. Therefore, there is a risk of breakage or leakage, especially for gas. CHBRs do not have this problem, and the majority of burner and boiler facilities are installed on the ground, usually with strong and reliable connections.
The expansion tank is usually installed at a high location in CHBRs, which might be toppled down during earthquakes. In most of the visits, weak and unreliable connections were observed in the frameworks supporting these reservoirs. The volume of the expansion tank in HPs is much lower than that in CHBRs; as a result, it has a lower risk and causes less concern for the residents.
Risk of damage to the boiler during an earthquake was among the most important concern of the residents. Gas leakage and smoke are very common in case of boiler damage. Risk of damage to the burner during earthquakes is another important concern. Gas leakage from the connections in one hand and the existence of a flame in the other hand can lead to explosion. This can be observed in both studied systems. To ensure the quality of gas pipe-to-burner connections and lack of corrosion of electric wires and equipment can prevent the boiler and burner-related accidents.
Hot water tank, radiator, and water pumps exist in both systems and can be damaged during earthquakes. Destruction and malfunction of these pumps can disrupt the function of the heating system. Moreover, breakage and leakage of inlet and outlet connected to the pump, water tank, and radiators can lead to water leakage in the building. Damage to the boiler-to-chimney connection and the chimney’s outlet path can lead to gas toxicity, boiler and burner malfunction, and gas leakage in the building. It is usually difficult to identify a disruption in the path of chimneys and their connection. So, these problems are not likely to be discovered.
Another damage in heating equipment such as boiler or pump caused by earthquakes, is loud noise and noise pollution. This problem is easily identified and has no casualty, though, it can cause undesirable situations during and after earthquakes.
An important point was the large number of HPs compared to CHBRs in two buildings with a similar number of residential units. For instance, there was one CHBF in a five-story building with 2 residential units on each floor, but a similar building had 10 HPs. Overall, the probability of accidents is higher in buildings with HPs due to the larger number of the facilities. In the other hand, due to the larger volume and size of CHBRs, the extension of accidents can be larger. However, we are not able to claim which system causes more damage. The equipment and facilities status are different across buildings, and many different possibilities exist during earthquakes.
Finally, the mean values of each column have been provided in Tables 7 and 8.
For the final comparison, the fuzzy RPN values were defuzzified using the center of gravity method. The final values have been presented in Tables 7 and 8. Moreover, the values as well as mean value have been plotted in Figures 3 and 4.
Based on Figure 3, the risk of damage to radiators and pumps is lower than the other facilities.
In addition, they have the minimum RPN value. HPs exist in each apartment unit which increase the length of gas junction pipes in the buildings. Based on the interviews and questionnaires, they had the highest risk value with a RPN of 334. Other items had less risk values (i.e., close to the mean value) (131.42).
Figure 4 shows that the risk of damage to radiators and expansion tanks is lower than the other facilities, and they have the minimum RPN value.
Due to the concentration of the entire building’s heating system in one room as well as the larger size of the equipment compared to HPs, CHBRs can cause different and severe damages during earthquakes, such as damage to the boiler-chimney connection, damage to the chimney outlet, incomplete combustion, and smoke leakage, with a RPN of 222.00, 204.00, and 179.33, respectively. It should be mentioned that the mean RPN value was 128.73. We should also pay attention to the risk of the main gas pipe breakage which is connected to inside of the building and has a considerable risk value for both systems. By the interview method, the citizens mentioned two points which might have a large impact on the risk score.
The possibility of urban gas shutdown by automatic shutdown facilities after earthquakes is very high in urban areas, which can have a positive effect on this risk.
In case of accidents and gas leakage from the main pipe, a possible explosion may occur outside of the building. Finally, the RPN values of both systems were compared (Figure 5).
Based on the questionnaires, the risk of HP system (mean RPN) was higher than that of the CHBR system (mean RPN of 131.42 vs 128.73).
The mean risk of HP was about 2% higher than the mean risk of CHBR. Overall, the small difference between the mean RPNs of the two systems can be neglected. As mentioned, based on the results of questionnaires and the fuzzy FMEA model, no definitive superiority can be presumed for HP or CHBR during earthquakes. The major difference in the risk of each system can be separately evaluated based on the conditions in each building, confidence about the quality of installation, connections, pipe materials, electric wire corrosion, etc. Thus, it is suggested that the risk of CHBR and HP during earthquakes is the same, and some issues such as design, economy, and technical issues play a greater role in selection of the system by the contractors. The important point is that citizens should ensure gas equipment risk reduction to decrease the hazards before and during crises such as earthquakes by regular checkup and ensuring the factors discussed in this paper.
It should be noted that the use of fuzzy space for FMEA calculations enhances the flexibility of computations and improves the uncertainty in model’s estimations. In classic FMEA, the values are estimated by individuals’ point of view as the scores in a crisp manner, thereby having some errors. The concepts of membership function in fuzzy theory and their combination with crisp calculations in the FMEA model approximates the results of reality and make more flexible, computational processes.
Gas systems have been always paid attention during earthquakes due to the potential gas leakage and explosion. Gas heating equipment are commonly used in all residential buildings in Iran, and have a high potential risk during earthquakes. In this study, during the visits, we found that a considerable number of heating facilities, especially CHBRs, were old and worn out that need repairment. The issue can double the risk of accidents resulted from the damage to facilities during earthquakes.
On the other hand, HPs can cause numerous concerns for the residents during earthquakes. The large number of packages in the buildings, their installation at high level, and lack of regular visits were the most important concerns observed during visits and interviews.
Based on the results of FMEA model, three factors namely severity, occurrence rate, hazard discovery rate were evaluated in 50 residential buildings in Tehran (district 9). FMEA’s estimations were performed in a fuzzy space to improve uncertainty and flexibility, which can confer better results compared to the classic FMEA model.
Finally, the RPN values of the two systems were compared. Based on the results, the RPN of the HP system was about 2% higher than that of the CHBR system (mean RPN value of 131.42 vs 128.73). Overall, the small difference between the mean RPNs of the two systems can be neglected and the risk of both systems can be deemed equal. As mentioned, based on the results of the questionnaires and the fuzzy FMEA model, no definitive superiority can be presumed for HP or CHBR during earthquakes. The major difference in the risk of each system can be separately evaluated based on the conditions in each building, confidence about the quality of installation, connections, pipe materials, electric wire corrosion, etc. Thus, it is suggested that the hazard rate of CHBR and HP during earthquakes is the same, and some issues such as design, economy, and technical issues play an important role in selection of the systems by contractors. The important point is that citizens should ensure the gas equipment risk reduction in order to decrease the hazards before and during the crises such as earthquakes. They can be sure of the risk reduction by regular visits, high confidence about the factors discussed in this paper, and installation of a frame for HPs.
Compliance with ethical guidelines
There were no ethical considerations to be considered in this research.
The paper was extracted from the PhD. Dissertation of Reza Nessari, PhD candidate, School of Environment, College of Engineering, University of Tehran, Tehran, Iran.
All authors equally contributed to preparing this article.
Conflict of interest
The authors declare no conflict of interest.