2009/08/24
Noise at a supermarket: case-study of a workplace with medium levels of sound exposure
Student: Marta Barceló Villalobos Master’s Thesis in Applied Environmental Science at Halmstad University (15 ECTs) Supervisor: Eja Pedersen Teacher: Stefan Weisner
Index Summary
Page 2
1. Introduction
Page 3
2. Methods
Page 4
2.1 Systematical literature review
Page 4
2.2 Pre-study
Page 4
2.3 Noise measures assessment
Page 4
2.4 Statistical analyses
Page 5
3. Results
Page 6
3.1 Sound levels at the supermarket
Page 6
3.2 Literature review
Page 10
4. Discussion
Page 12
5. Conclusion
Page 12
6. References
Page 13
7. Appendix
Page 15
1
Summary Noise pollution is a serious environmental problem. People are often exposed to environmental noise everywhere: at home, in the workplace, on the streets and so on. Nowadays, noise has become a large problem in the developed countries as noise levels are increasing and noise exposure has been linked with a range of negative effects to health. The purpose of this case-study is to predict possible adverse health effects from noise among employees at a supermarket. The case-study consisted of measurements of noise levels at ICA-Maxi Högskolan supermarket in Halmstad (Sweden), during two weeks and a literature review. The measurements were analyzed statistically, and the results were linked to the literature review search. The results showed a range of 59 to 80 dB in LC eq. values and a range of 53 to 72 dB in LA eq. values. The values found are risky for the health of the employees. Hearing loss, blood pressure, stress, risk of accident and sleep disturbance, were studied. After that, a risk of suffering more stress because of the levels of noise in the supermarket was found. It is shown that the cashier area was the noisiest place evaluated. During a normal week, there is no large difference of decibels between the times evaluated. Furthermore, there was not low frequency noise levels found. The case-study has demonstrated that there are moderate levels of noise in a workplace at a supermarket. The results obtained could be useful in making decisions and to take care of the health of employees. Future studies should include questionnaires, to measure the perception of noise among the employees, and timetable from the employees’ schedule to make a deeper study.
I want to say thanks to Helen and Rickard, the bosses of ICA MAXI Högskolan, for allowing me to take the data of sound levels at their supermarket during the period of two weeks. Thank you to Stefan Weisner and Eja Pedersen for giving me the opportunity to make this study.
2
1. Introduction Noise has become a large problem nowadays for the health of the people. Elevated sound levels cause adverse effects on health, for example hearing impairment, hypertension, ischemic heart disease, irritation, sleep disturbance, and decreased school performance (Passchier-Vermeer and Passchier, 2000). The effects of noise in industrial and other workplaces that are known for high levels of noise are well studied. Supermarkets, where many people work, have not been studied yet. Even though the levels supposedly are not so high, there could be a risk of health effects due to noise exposure. The aim of this study was to predict possible adverse health effects from noise among employees at a supermarket. Below, the study goals of this case study are shown: - To know how the sound levels vary over a day and between different places in the same workplace. - To be able to know which places are more exposed in the supermarket, during Easter and in a normal week. - To find out the risks of getting a disease because of the exposure. This case-study has some limitations: this report is focused on the ICA Maxi Högskolan supermarket at Halmstad, the data was taken during two weeks, the measurements were taken at five different times and a small number of places at the supermarket were evaluated. Future studies should measure the sound from 08:00 to 22:00 hours and should design some questionnaires for the employees to make a deeper study.
3
2. Methods 2.1 Systematic literature review The first step was to do a systematic literature search in scientific databases on the health effects of the medium levels of noise at a workplace like a supermarket. The databases used were PubMed and ISI WEB Science, using the keywords “noise” and “workplace”. As a result, 488 articles were found. Of those, only 33 were available as full text papers at Halmstad University. Table 1. Results from the literature search.
Source PubMed ISI WEB Science
Total articles found 313 175
Total articles used 27 6
The articles were classified due to their topic. All of them were read. After that, I realized that not all articles from the searches were useful for this case-study and so I worked with those that were. Furthermore, some other interesting articles were found in the references of the articles searched. Notes from these articles were taken, which are shown in the chapter literature review. 2.2 Pre-study The second step was to learn what noise means, how to measure noise, how the equipment used works and how to use it. For this purpose, I found some information about what the noise means and how to measure it and so I looked up some information on Internet. I then read the manual from the device that I intended to use which explains how it is works and how to record the data correctly. A pre-study was carried out at the Students’ Union restaurant in order to test the sound level meter and the software. 2.3 Noise measures assessment Sound pressure measurements were conducted with a sound pressure level meter (model BZ-7130, Brüel and Kjaer). The level of the sound at the supermarket was measured five times during the day at the time: 10:00 to 10:30, 12:00 to 12:30, 14:00 to 14:30, 17:00 to 17:30 and 19:00 to 19:30 hours, in different places in the same supermarket (the clothing department, the chilled and frozen food area, the central corridor and the cashier area) (Appendix) during a period of fifteen days (including Easter). Five routine monitoring times were measured for noise in each place on each day. At each monitoring time, five measurements of 1 minute were carried out, in total 1500 measurements. Notes of extreme noise incidents were taken.
4
Table2. Overview of the measurements during one day
Places Clothing department Chilled and frozen food area Central corridor Cashier area Total measurements
Monitoring time 10:00-10:30 12:00-12:30 14:00-14:30 17:00-17:30 19:00-19:30 5 times*1 5 times*1 5 times*1 5 times*1 5 times*1 minute minute minute minute minute 5 times*1 5 times*1 5 times*1 5 times*1 5 times*1 minute minute minute minute minute 5 times*1 minute 5 times*1 minute 20
5 times*1 minute 5 times*1 minute 20
5 times*1 minute 5 times*1 minute 20
5 times*1 minute 5 times*1 minute 20
5 times*1 minute 5 times*1 minute 20
In order to establish the levels of sound, two variables were studied: LA eq. and LC eq. LA eq. and LC eq. are noise parameters that calculate a constant level for noise with the same energy content as the varying acoustic noise signal being measured. The letter “A” denotes that the A-weighting has been included and “eq.” indicates that an equivalent level has been calculated. An A-weighting filter is commonly used to emphasize frequencies around 3–6 kHz where the human ear is most sensitive, while attenuating very high and very low frequencies to which the ear is insensitive. The “Cweighting” filter is particularly used in the assessment of very loud or low-frequency sounds. In summary, LA eq. is the A-weighted equivalent continuous noise level and LC eq. is the C-weighted equivalent continuous noise level. To reach some conclusions about the measures, the collected data was included in an Excel document using the adequate software. Measurements with extreme noise incidents were excluded. After that, 14894 measurement values remained. The assessment of occupational exposure to noise was based on the noise exposure level averaged over 5 or less (after removing measurement with extreme noise incidents) repeated measurements at each monitoring time at each place. (Table 2) Then, the data was summarized by ranges, to determine which places are more exposed in the supermarket. 2.4 Statistical analysis The purpose of using statistical analysis is to answer the study goals from the thesis plan. The data processing and analyses were performed in SPSS 16.0 for Windows and Microsoft Excel 2003 software. Data from the measurements were organized in an Excel file which can be used in the program SPSS. The statistical analysis was based on the following tests: at first, the KolmogorovSmirnov Test was used to identify whether or not the data looked like a normal distribution. The significance level of the test was set at 0.05. Then, the one way analysis of variance (ANOVA) was used because I was working with more than two samples. Finally, a variance Post-Hoc Test (Least Significance Difference) was used to establish the significance of the observed differences between the places in the different variables (LA eq. and LC eq.). The differences between the LC eq. and LA eq. were calculated in order to detect possible low frequency noise. Differences higher than 15 dB are commonly considered an indication of low frequency noise.
5
3. Results 3.1 Sound levels at the supermarket The range of values obtained (Table 3) shows the most exposed area in the supermarket. Table 3 below, also shows the size of the range of the variables using breaks. The highest levels of LA eq. and LC eq are on the cashier area. Moreover, if you pay attention to the breaks, the comparison between both variables shows that the LC eq. range is high in the chilled and frozen food area and in the cashier area and that the LA eq. range is high in the cashier area. The noisiest place for the human ear is the cashier area. This is due to the sound of the money falling in to small bottles by cashier, people speaking, trolleys moving, and so on. Table3. Range of the mean of the sound pressure values in LC eq. (dB) and in LA eq. (dB) at places evaluated. The size of the ranges are shown using breaks.
Place Clothing department Chilled and frozen food area Central corridor Cashier area
Variables (Range) LC eq. (dB) LA eq. (dB) 59-61 (2) 53-58 (5) 61-74 (13) 58-66 (8) 64-73 (9) 63-80 (17)
54-65 (11) 59-72 (13)
The results from the Kolmogorov-Smirnov Test show a p-value higher than 0.05 (Table 4). This means that there are not significant differences with respect to a normal distribution, so parametric tests could be used. There is no indication of low frequency; all values are below 15 dB (Table 5). The one-way ANOVA test table shows significant differences between the places for both variables with significance less than 0.001(Table 6). Moreover the high F values were used to decide whether the sample means are within sampling variability of each other. The F values got high values which indicated clear differences between LA eq. and LC eq. means. Next, Post-Hoc Test (LSD) was used to identify the relationship between each place in the different variables (LA eq. and LC eq.). The results show significant differences (pvalue 0.05.
6
Table 5. Mean, Total measurements (N) and Standard Desviation
Place
LC eq.
LA eq.
Clothing department
61.59 68 1.88 70.43 68 1.37 66.75 68 1.30 66.29 67 2.85
56.62 68 2.78 61.21 68 1.56 58.89 68 2.03 62.39 67 2.18
Mean N Std. Deviation Chilled and Mean frozen food N area Std. Deviation Central Mean corridor N Std. Deviation Cashier area Mean N Std. Deviation
Low frequency noise 4.96 68 1.40 9.22 68 1.18 7.86 68 1.39 3.89 67 1.14
Table 6. Differences in mean LC eq. and LA eq. levels, respectively, between the four places. One-way ANOVA Test. Use to identify significant differences between LC eq. and LA eq. values.
F 235.613 92.870
LC eq. LA eq.
Significance
