Field Studies
A Prioritization Approach for Noise Risk Management in a Petrochemical Complex
2013 Volume 55 Issue 3 Pages 204-210
Details
2013 Volume 55 Issue 3 Pages 204-210
Objectives: Noise is probably the most common occupational hazard facing workers today. This paper presents a model to evaluate and select a unit as the first priority for implementing noise control measures from the different parts of an industrial complex. Methods: For this purpose, a formula was proposed as a priority risk index (NCPI) by considering some criteria, including: the number of exposed workers in each part, their exposure times and the range of sound pressure level. The above criteria were weighed by appending weighting factors, which can be different based upon the environmental noise levels. The environmental noise measurements were also carried out by lattice method according to ISO 9612:2009. At the next stage, the numbers of stations that fell within the desired ranges in the weighting factors table were found. then the priorities were identified using the above criteria and the NCPI. Results: The results indicate that the compression unit of air plant had the highest score among the different departments of the company under study (NCPI=0.3). Using this model is easy and fast. It is applicable to similar industries and also offers valuable information for prioritizing noise control measures. Conclusions: Therefore, it will be possible to minimize worker exposure to noise in the most polluted places in any industry by identifying the main noise sources and implementing measures suitable for controlling the risk.
(J Occup Health 2013; 55: 204-210)
Noise exposure is probably the most frequent risk factor at industrial settings, it can result in increased nervous tension with associated psychological effects, and most particularly it causes damage to the audiosensory mechanism and leads to premature and permanent loss of hearing1–6). High levels of occupational noise become a problem in most petrochemical companies. Since they are open-air industrial plants for reasons of safety, they may need to consider more specific noise control measures. This is due to both environmental and the occupational exposure7–9). So many industries have a strong interest in finding economic and effective solutions for eliminating or reducing the risk arising from noise exposure to a minimum6, 10).
With respect to economic and operational issues, implementation of control methods for all sources of a company would not be practical and cost effective, especially at large-scale noisy industrial complexes like petrochemical plants. On the other hand, most research projects have the limited time, limited funding, and other limited research/development resources; thus it seems to be necessary to develop a prioritization process. So it is necessary to determine the ranks of noise pollution for all parts of a company and identify the main noise sources for them. This is similar to the risk management issues in which a risk priority number is computed using quantitative or qualitative assessments to provide a classification and decision-making criteria for reducing risks and improving outcomes11,12). A large number of studies have been performed over many years to assess, predict or control the noise pollution in oil refineries and petrochemical plants, but almost all of them did not consider the prioritization approach for noise risk management, despite its importance7–9, 13–16).The aim of this article is to present a model to evaluate and select a unit as the first priority for implementing noise control measures from the different parts of industrial complexes like petrochemical plants.
The survey was carried out at Fajr Petrochemical Company which provides utility services to other companies. It is located in Petrochemical Special Economic Zone in Mahshahr in south of Iran. This company is composed of three plants: water treatment plant, air plant and powerhouse (Fig. 1).
Satellite image for the petrochemical complex under study.
The main noise sources at the three sites were as follows:
Environmental noise measurements were performed within the different parts of the three plants according to international standards ISO 9612:2009 and the lattice method17). A B&K (Bruel & Kjaer, Naerum, Denmark) Type 2236 noise analyzer was used to measure existing sound levels. The B&K analyzer was equipped with a windscreen to eliminate noise associated with wind blowing across the microphone. The noise analyzer was calibrated with an acoustical calibrator before and after each measurement (B&K type 4231). All measurements were done during the morning shift and while the meteorological conditions were as follows:
Each department was divided into 10-meter by 10-meter stations, and the sound pressure levels measurements were taken at the center of each station. The microphone was located at a height of 1.6 m from the ground, 1 m from walls and 2 m from crossings to avoid reflection of the sound waves. The logarithmic average of three readings of noise level at each location was taken. In cases in which a station was located on a machine or in nonmeasurable places, the station was considered an inaccessible point and excluded.
In this research project, places like office buildings, control rooms, workshops and nonmeasurable places in each unit were regarded as blind points. After collecting the measurements, noise maps and noise contours were prepared (Fig. 2).
Noise contour of Powerhouse plant.
In general, one of the most important steps in the project prioritization process is to establish the ranking criteria. After explaining what criteria to use for ranking parallel projects, you are able to rank each project quantitatively and determine its level of priority18, 19).
In present work, the priorities were developed based on the following criteria:
A formula was proposed using the mentioned criteria to get a simple and efficient approach that can prioritize the various parts of a company in order to find the first priority for adopting noise control measures.
In the first step, the criteria above were weighted by appending the weighting factors obtained using the method recommended by Bies and Hansen to calculate noise impact20). The numerical value of these weighting factors (Wi) can differ based on the sound pressure level range (Table 1). It is worth mentioning that some extrapolation was performed for sound pressures levels of 90-95 onwards.
| Range of SPL dB(A) | Wi | Range of SPL dB(A) | Wi |
|---|---|---|---|
| 35-40 | 0.01 | 70-75 | 0.83 |
| 40-45 | 0.02 | 75-80 | 1.2 |
| 45-50 | 0.05 | 80-85 | 1.7 |
| 50-55 | 0.09 | 85-90 | 2.31 |
| 55-60 | 0.18 | 90-95 | 3 |
| 60-65 | 0.32 | 95-100 | 5 |
| 65-70 | 0.54 | 100-105 | 8 |
W: weighting factor, SPL: sound pressure level.
In the second step, the magnitude of sound pressure levels was measured in each industrial sector; first, the numbers of stations that fell within the desired ranges, given in Table 1, were found. The relevant information is presented in Table 2. Then the numbers of employees who work in these sections and their numbers of working hours were determined. In this research, a priority risk index was proposed based on basic acoustic knowledge as follows:
 | (Eqn 1) |
| Plant | Part | Number of measurement stations | Number of nonaccessible point | Number of measurement stations in the mentioned ranges of sound pressure level (dB(A)) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 50-55 | 55-60 | 60-65 | 65-70 | 70-75 | 75-80 | 80-85 | 85-90 | 90-95 | 95-100 | >100 | ||||
| Water | RO domain | 166 | 138 | – | – | – | 9 | 71 | 57 | 24 | 5 | – | – | – |
| DM Domain | 100 | 30 | – | – | – | – | 18 | 32 | 30 | 18 | 2 | – | – | |
| Water of firefighting domain | 73 | 47 | – | – | 12 | 38 | 10 | 10 | 3 | – | – | – | – | |
| Pretreatment domain | 186 | 194 | 30 | 40 | 55 | 42 | 12 | 4 | 3 | – | – | – | – | |
| Water tanks domain | 58 | 12 | – | 8 | 38 | 12 | – | – | – | – | – | – | – | |
| Air | Separation | 96 | 4 | – | – | – | 1 | 3 | 35 | 19 | 17 | 19 | 2 | – |
| Compression | 22 | 2 | – | – | – | – | – | 3 | 7 | 4 | 4 | 2 | 2 | |
| Other | 17 | 9 | – | – | – | – | 4 | 8 | 3 | 2 | – | – | – | |
| Power-house | GT11-GT15 | 140 | 40 | – | – | – | – | 25 | 50 | 57 | 8 | – | – | – |
| GT16-17 | 131 | 24 | – | – | – | – | 22 | 94 | 13 | 2 | – | – | – | |
| Heat recovery generator | 46 | 8 | – | – | – | – | 1 | 34 | 11 | – | – | – | – | |
| Steam distribution system | 25 | – | – | – | – | – | – | 14 | 6 | 5 | – | – | – | |
| Auxiliary boilers | 35 | – | – | – | – | – | 1 | 12 | 18 | 3 | 1 | – | – | |
| Total | 1,095 | 508 | 30 | 48 | 105 | 102 | 167 | 353 | 194 | 64 | 26 | 4 | 2 | |
RO: reverse osmose, GT: gas turbine, DM: demineralized water.
where NCPI is the Noise Control Priority Index, wi is the weighting factors, pi is the number of exposed workers, ti is their exposure time in hours, P is the total number of exposed workers and T is the total exposure time in hours. According to the above equation, the weighting factors are multiplied by the number of exposed workers and their exposure time. In order to normalize it, the numerator is divided by the product of multiplying the total number of workers in the three plants by their total exposure time. The unit with the highest NCPI will have the highest priority for adopting noise abatement measures. This index is a new version of the NII (noise impact index) in which we added a time factor to both the numerator and denominator. The NCPI can utilize the parameters of noise that affect its health damage potential such as noise level and exposure duration. These factors alone will cause health damage, but all factors are integrated into the NCPI.
Fajr Petrochemical Company operates 24 hours a day, seven days a week. There are three shifts (morning, afternoon, nights). Workers (site-men) work four days on one shift and after take four days off after working the night shift21). Table 3 shows the number of workers and approximate area of each plant.
| Plant | Area (m2) | Number of site-man per shift |
|---|---|---|
| Water | 100,400 | 4 |
| Air | 15,000 | 4 |
| Powerhouse | 44,900 | 7 |
| Total* | 160,300 | 15 |
Site-men working here are mobile workers who do not have permanent workstations and inspect everywhere or everything of their plant. Considering the working conditions, site- men jobs by default involve working on site for 4 hours per day. They usually spend four hours per day in the resting room.
The sound pressure levels at 1,095 stations were measured across the company, and there were 508 non-measurable stations. The results of the calculations for determining the range of sound pressure level (SPL) for each different unit are shown in Table 2. The number of measurement stations, number of inaccessible points and number of measurement stations in the mentioned ranges of sound pressure level in all parts of company can be obtained from Table 3. for example, the compression part of the air plant has 22 noise measurement stations with the SPLs of two stations being over 100 dBA, and 54.6% of measurement stations of the compression unit of air plant had SPLs over 85 dBA. Separation, DM domain and steam distribution system were ranked next highest in this area, with 39.6, 20 and 20% of their measurement stations having SPLs over 85 dBA.
In general, the air plant, which had 52 measurement stations with SPLs greater than the exposure standard, was the most polluted plant at this company. Since the minimum pressure level recorded was 51.2 dBA, data classification was started with the range of 50-55. The air and water treatment plants have the same numbers of site-men for three shifts (12 people), but just three workers work in the steam distribution system unit of the powerhouse plant over 3 shifts, and other units of this plant have 18 site-men.
Below is an example of the NCPI calculation for the RO domain:
 |
NCPI=[(12 × (4÷5) × 0.54)+(12 × (4÷5) × 0.83)+(12 × (4÷5) × 1.2)+(12 × (4÷5) × 1.7)+(12 × (4÷5) × 2.31)]÷[(12 × 4)+(12 × 4)+(12 × 4)+(12 × 4)+(12 × 4)+(12 × 4)+(12 × 4)+(12 × 4)+(18 × 4)+(18 × 4)+(18 × 4)+(3 × 4)+(18 × 4)]=0.092
As mentioned earlier, each worker is exposed to noise on site for 4 hours per day, and the exact exposure time for each range of sound pressure level could not be decided. Therefore, in this case, the exposure time of 4 hours should be divided by the number of ranges of sound pressure level of each part. Otherwise the worker’s exposure time would become longer in the larger number of ranges of sound pressure level in some parts.
The results of NCPI calculation for each unit are shown in Table 4 According to Table 4, the compression unit of the air plant had the highest score among the different parts of the company under study (NCPI=0.24), and the next highest priority was the auxiliary boilers of powerhouse.
| Plant | Part |  |
NCPI | Rank |
|---|---|---|---|---|
| Water | RO domain | 63.17 | 0.092 | 8 |
| DM domain | 86.78 | 0.12 | 6 | |
| Water of firefighting domain | 44.06 | 0.06 | 9 | |
| Pretreatment domain | 33.32 | 0.04 | 10 | |
| Water tanks domain | 16.64 | 0.02 | 12 | |
| Air | Separation | 99.98 | 0.14 | 4 |
| Compression | 169.68 | 0.24 | 1 | |
| Other | 72.48 | 0.10 | 7 | |
| Powerhouse | GT11-GT15 | 108.72 | 0.15 | 3 |
| GT16-17 | 108.72 | 0.15 | 3 | |
| Heat recovery generator | 89.52 | 0.13 | 5 | |
| Steam distribution system | 20.84 | 0.03 | 11 | |
| Auxiliary boilers | 130.18 | 0.19 | 2 |
RO: reverse osmose, DM: demineralized water, GT: gas turbine.
The air plant is located in the northern half of the Fajr complex and consists of two departments: separation and compression. However, although acoustic enclosures for all types of its compressors have been supplied, the air plant is still most risky plant in connection with occupational exposure to high level sound. This is due to noise arising from a variety of sources and special operating conditions including: suction and blowing of air, water pumping, refining operations, drying processes and fluid motion in pipes. About 38.5% of the measurement stations in the air plant (area of 1,350 m2) had a noise level above 85 dBA (eight-hour time weighted average limit recommended by the American Conference of Industrial Hygienists) and workers were exposed dBA at 4% of the measurement stations (600 m2). Despite Powerhouse area is three times greater than that of the air plant, only 4.2% (1,900 m2) of its measurement stations had a noise level above the threshold limit value and 2,500 m2 of the water treatment plant, which is an area about 4.5 times larger than that of the air plant, was found in the danger zone. As a result, it may be possible to predict that the air plant will be given first priority for noise abatement measures. Besides, the results of NCPI calculation proved it. Among the different units at the Fajr petrochemical complex, the compression unit was in the first rank (NCPI=0.24). In a simple comparison between NCPI scores of different units, it can be concluded that the score for the compression unit is 12 times more than that of the unit with the lowest priority, i.e., the water tanks domain (NCPI=0.02). The second rank, auxiliary boilers, is found in the powerhouse, and there is a difference of 0.05 between their scores. If we exclude the number of workers and weighting factors for prioritizing, our results will vary. Under the area of the separation unit with a noise level above 85 dBA (area of 3,800 m2) would be the first priority, and the DM domain would be in the second rank. The compression unit would not be considered because of its small area. On the other hand, using the present approach, the compression unit, which has fewer workers and a small area, would be in the top rank due to the weighting ranges of the sound level.
The index presents a practical, scientific, simple and fast solution for the occupational hygienist and other relevant specialists. The results will be unreliable if all the important influential factors involved in selecting the first priority are not considered. It can be a debatable index, and everyone can look at it from a different perspective. Nevertheless, the NCPI is a new step in this issue, and its elements can change depending upon the industrial complex and conditions.
Finally, it could be possible to prioritize the different parts of an industrial complex for controlling noise risk by a simple and practical model, and this approach can also be implemented at many industrial complexes. Note that the NCPI is only a decisionmaking criterion that was formed based on the Fajr Petrochemical Company work environment. But it can be used at other industrial complexes with few changes. Anyone could add other factors influencing the noise control priority index depending on their working condition. It is our hope that such a model can help to reduce worker exposure to noise in the most polluted places by identifying main sources so that engineering or administrative controls can be implemented for them and help organizations decrease costs by solving their most dominant problems concerning noise pollution. It is just like the case of risk management.
The present work was part of an M.S. hesis supported by Tehran University of Medical Sciences. The authors would like to take this opportunity to thank HSE Services of the National Petrochemical Company of Iran for their financial and other support.
