Health GIS Conference Proceedings 2020

P a g e | 124 Students will fill in a consent form fourteen (14) Figure 2b: section c (disease of the students) days before they come to the university. The questionnaire consists of three (3) sections of 4.2 TEMPERATURE questions. One of the screening techniques for individuals with COVID-19 is by using temperature measurement. In Section A is for demographic data. The students this paper, we are proposing an Artificial are required to fill in their name, faculty, home Intelligence (AI) algorithm for the temperature address, rental address, age, user id and password. screening process using an Infrared Thermal camera All demographic data will be kept in the database device. The development of AI algorithm involves and need to be cross referenced with the student automated multiple measurements of body database at the university. Then, the students will temperature for rapid and high degree of screening log in the user id and password from section A. accuracy. An industrial standard thermal camera Section B is to assess the students’ current health will be connected to our AI engine to establish an status i.e. the presence of fever, sore throat, cough, algorithm by implementing a real-time deep or flu. Section C is to assess the students’ learning detection in which the positioning of bare underlying medical conditions such as hypertension, human body can be easily measured. The asthma, diabetes mellitus, heart disease, and parameters such as the distance of an individual, pregnancy. Each question from the questionnaire human body physical condition, the surrounding carries a different weightage scoring and the temperature will allow a simultaneous processing of cumulative marks will stratify the students in four a large number of individuals. Any object with a risk categories (Figure 2a-b). They are high, temperature above absolute zero emits a detectable medium and low as shown in Table 2. For example, number of radiations. The thermal camera converts if the student’s home address is in the red zone and IR radiations into grey value and establishes the he has cough and underlying diabetes mellitus, he accurate corresponding relation between grey value will be classified as High Risk. and temperature through the temperature measurement algorithm model. The model Table 2: Description of risk categories (Temperature Gray Level Curve) is obtained by black-body calibration. Figure 3 illustrates the Risk Description Marks thermal detection temperature screening. Category High Red zone area and 80 and above pre-existing illness 40 and below 80 Medium Red zone area and Below 40 Low none pre-existing illness Non-red-zone and none pre-existing illness Figure 2a: section b (status of the students) International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 125 b. For each cluster, calculate the new mean c. Calculate the junction among clusters to keep the k number of clusters Until convergence criteria are met. Algorithm1 : KNN Psuedo code. The KNN equation is as below: 4 3| 1 |\" 56 56 Equation 1 Figure 3: Temperature Screening Study In Figure 4 show the example of the output capturing from MI band 4. The advantages of these high-speed data processing capabilities are: • High Efficiency: It takes only one second for the thermal camera to measure the temperature for each person. Thus, no congestion will be made when passing through the temperature screening site. • Safety: Thermal camera supports non-contact temperature measurement which can accurately measure the temperature from 2.5 meters away. This eliminates the risk of direct contact transmission. 4.3 Heart Rate Figure 4: The temperature and heart rate A heart rate monitoring by using the bracelet will be added along with temperature measurement in high 4.4 Movement risk students. We will use KNN algorithm method. The students movement data and location will be The author (Helmi et al., 2020) has developed the tracked from their smartphone using the university heart rate measurement using K-Means for Wi-Fi network (Narzullaev et al., 2020). This monitoring ISA students in USIM. He classified it information also allows us to identify their daily and showed the student activities during classes. physical activities in order to exclude the The k-nearest neighbours (KNN) is one of the physiological increased of heart rate especially furthermost simple classification methods especially when they are engaged in physical activities such as in analyzing a huge matrix of features or providing jogging and playing sports. Figure 5 shows the recommendation (Tarus et al., 2018). Pseudocode illustration of the connectivity in the tutorial provides the normal stages of applying the KNN. classroom. For example in tutorial room BR 1019, the red color shows the student with the symptom. Input: The color in blue shows the student that might be infected because of sharing the same Wi-Fi devices. + 1, 2, … . , . //set of n data items But another color is normal color because it connected to difference Wi-Fi devices. This show 0 //number of desired clusters the location of the individual students based on the nearest Wi-Fi devices. Output: A set of k clusters Steps: 1. Arbitrarily choose k data items from D as initial centroids; 2. Repeat a. Allocate each item d1 to the cluster with the nearest centroids (Equation (1)); International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 126 4.5 Risk Assessment requires the usage of four applications. Table 3 In the University, we develop a Covid19 risk presented each application for students to install. assessment and mitigations for students registration. Risk mitigations for each risk category when the Risk assessment is one of the major developments students are already in campus or in their housing for Covid-19 to ensure the University health care is are show in Table 3. prepared to handle the infection. The risk mitigatios Table 3: Risk Mitigations for Each Risk Category High 1. Temperature check ASAP at University Health Center. 2. If High temperature. Mitigation:  Quarantine  Wear bracelet  Download and activate Contact Trace Apps.  Check temperature every day 3. If Normal temperature: Mitigations:  No quarantine  Wear bracelet  Download and activate Contact Trace Apps.  Check temperature every day Medium Low 1. Temperature check 1. No temperature check at ASAP at University Health University Health Center Center 2. If High temperature. Mitigations: Mitigations:  No quarantine  Quarantine  Wear bracelet  Wear bracelet  Download and activate  Download and activate Contact Trace Apps Contact Trace Apps.  Check temperature every  Check temperature week. every day  Educate on emergence of 3. If Normal temperature: symptoms and the need to Mitigations: get medical treatment  No quarantine ASAP  Wear bracelet  Download and activate Contact Trace Apps.  Check temperature every week International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 127 Figure 5: Illustration of the Classroom REFERENCES 5. CONCLUSION Cowling, B. J., Lau, L. L., Wu, P., Wong, H. W., Students are potential carriers of COVID-19 and Fang, V. J., Riley, S. and Nishiura, H., 2010, universities should develop an effective screening Entry Screening to Delay Local Transmission of framework to minimize the viral transmission. Our 2009 Pandemic Influenza A (H1N1). BMC framework consists of 4 components i.e. Infectious Diseases, Vol. 10(1), 82. temperature measurement, heart rate monitoring, movement tracking and risk assessment for the Gregoire, A., Dolan, R., Birmingham, L., Mullee, mitigation guideline. M. and Coulson, D., 2010, The Mental Health and Treatment Needs of Imprisoned Mothers of ACKNOWLEDGEMENT Young Children. The Journal of Forensic The authors wish to send his/her appreciation to the Psychiatry & Psychology, Vol. 21(3), 378-392. editor and anonymous referees for their constructive comments and criticism. This work is supported by Helmi H., Ismail, W., Hendradi, R. and ustitia, A., the School of Medical Sciences, Universiti Sains 2020, Students Activity Recognition by Heart Malaysia (USM), Malaysia. Rate Monitoring in Classroom using K-Means Classification. Journal of Information Systems Engineering and Business Intelligence. Vol 6(1). https://e-journal.unair.ac.id/JISEBI/article/view- /18166. Khan, K., Eckhardt, R., Brownstein, J. S., Naqvi, R., Hu, W., Kossowsky, D. and Sears, J., 2013, Entry and Exit Screening of Airline Travellers During the A (H1N1) 2009 Pandemic: A Retrospective Evaluation. Bulletin of the World Health Organization, Vol. 91, 368-376. Liu, Y. C., Kuo, R. L. and Shih, S. R., 2020, COVID-19: The first Documented Coronavirus Pandemic in History. Biomedical Journal, Vol. 43(4), 328-333. https://doi.org/10.1016/j.bj.20- 20.04.007. Narzullaev, A., Muminov, Z. and Narzullaev, M., 2020, Contact Tracing of Infectious Diseases Using Wi-Fi Signals and Machine Learning Classification. IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET), Kota Kinabalu, Malaysia, 2020, 1-5. Nishiga, M, Wang, D. W., Han, Y., Lewis, D. B. and Wu, J. C., 2020, COVID-19 and Cardiovascular Disease: from Basic Mechanisms to Clinical Perspectives. Nat Rev Cardiol, Vol. 17(9), 543-558. doi:10.1038/s41569-020-0413- 9. Tarus, J. K., Niu, Z. and Mustafa, G., 2018, Knowledge-Based Recommendation: A Review of Ontology-Based Recommender Systems for E-Learning. Artif. Intell. Rev., Vol. 50(1), 21–48. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 128 IMPACT OF BEHAVIOR, HEALTHCARE AND SOCIOECONOMIC FACTORS ON COVID-19 SITUATION OF ASEAN COUNTRIES Htet Yamin Ko Ko, Nitin Kumar Tripathi and Ranadheer Mandadi Department of Remote Sensing and Geographic Information System, Asian Institute of Technology, Thailand E-mail: [email protected], [email protected], [email protected] ABSTRACT Fundamental of detailed and comprehensive epidemiology study is focused on understanding of spatial and temporal patterns of disease. Main factors in the formation of Covid-19 across different spatial and temporal ranges plus understanding on these participants can manipulate for modelling. Numerical studies of pandemic patterns have been concentrated on spatial regression methods to test hypotheses related with factors to understand about the physical mechanism that drive precipitation. Ordinary least square method has been determined to yield biased results in the spatially correlated data. This study focus on spatial regression analysis of ASEAN Countries will perform by focusing on total number of Covid-19confirmed cases, total death cases of Covid-19, Socioeconomic Factors (Human Development Index, Life Expectancy Index, International Inbound Tourists Index, Gross capital formation (% of GDP), Population Density, Population at Urban, Behaviour Factors (Male Smoker, Female Smoker) and Healthcare Factors (Cardiovascular Death Rate and Diabetes Prevalence) by using “Ordinary Least Square”, “Spatial Lag” and “Spatial Error” Models. According to the comparison of regression outputs, Total Population and Male Smoker Ratio factors are stated to be most influenced on Covid-19 incidents and fatalities. In the comparison of R-squared and Akaike Information Criterion (AIC) values, Spatial Lag Model outperform among three regression models. KEY WORDS: spatial regression, spatial lag, spatial error model, ordinary least square, spatial analysis 1. INTRODUCTION control the disease spreading by intensive lock Since Covid-19 pandemic has significant impact on down and medical facilities at the whole district. health, economic, politic and education factors on Apparently, Chinese Government could not control global scales, various researches focused on various the disease spreading and then on January 2020, study fields have been performed to analyze about “World Health Organization” (WHO) its impacts, the factors that leads to increase the acknowledged the eruption as “Public Health Covid-19 infectious rate and the solutions to reduce Emergency of International Concern” and later the infectious rate. Traditional statistical analysis declared as a pandemic in March, 2020. According and spatial analysis methods are also applied to to the pandemic conditions, a lot of changes that understand the correlation of the factors and the never happen in human history emerged such as Covid-19 situation, to estimate the Covid-19 factories and airlines are closed, all the public infectious rate and to check the most impact factors events are postponed or cancelled, schools and that may lead to high transmission and morality rate. universities are fully/partially closed or changed as This study is focused on performing spatial analysis online classes, supply shortages because of at country level on Asean countries and find out the transportation between countries are closed and this major factors on the situation of Covid-19 pandemic lead to panic buying behaviours, etc. There were in ASEAN regions. also positive things happened during pandemic era such as reduction of air pollution index and 2020 was the special year for everybody not greenhouse gases and reduction of water pollution. because of this happens only once in a century (first two digits match the second two digits) but because 1.1 Study Area of the global pandemic caused by SARS (Several The “Association of Southeast Asian Nations” Acute Respiratory Syndrome)-Cov2 Virus, called as (ASEAN) is an intergovernmental organization Covid-19. It was initiated at the end of 2019 in Wuhan District, China. Chinese Government tried to International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

intended to promote economic development and P a g e | 129 regional constancy among its members. Currently, 10 members included: Indonesia, Malaysia, spatial correlation between topographic factors and Philippines, Singapore, Thailand, Brunei, Laos, monthly rainfall in East Africa. According to four Myanmar, Cambodia, and Vietnam. ASEAN fitness indexes (Akaike Information Criterion countries have a total population of 650 million and (AIC), log likelihood ration test for spatial it has 2040697 Covid-19 Cases and 44782 Covid-19 dependence and Schwarz criterion), Spatial Lag deaths on 1st February 2021. The motto of ASEAN model has better fit result than Ordinary Least is “One Vision, One Identity, One Community”. Square model. The purpose of selecting ASEAN countries as study region is ASEAN countries require no visa to travel Spatial regression models are also applied in within ASEAN regions therefore the roaming rate social science in (Voss et al., 2006) to analyze among ASEAN countries is higher than other variation of pre-teen poverty levels in US regions. countries and therefore, Covid-19 impact of Moran scatterplot is used to test spatial dependency ASEAN countries is selected to research. According among the dependent variable and the bivariate to the Covid-19 daily data, 2040697 Covid-19 dependency among each variable with dependent confirmed cases and 44782 deaths in ASEAN variable. Ordinary Least Square model is used as region in January 2021 (Figure 1). diagnostic tool for the presence of Spatial Error model and Spatial Lag model. Then “Spatial Lag” 1.3 Literature Review and “Spatial Error” models are built by using (Putra et al., 2020) have researched about the spatial different analysis tools and then log-likelihood and autocorrelation of food expenditure with Socio- AIC values are computed to check the fitness of Economic Factors of Indonesia at province level. each model. In this research study, authors stated Ordinary Least Square Regression model is that “Spatial Error” model surpasses “Spatial Lag” computed to run robust diagnostics which will select model. Another application area that spatial significant spatial regression model. Four measures regression models are actively participated is Spatial of fit statistics (“Pseudo R-Squared”, “maximized Demography (Chi and Zhu, 2008). Ordinary log-likelihood”, “Akaike Information Criterion”, regression and different spatial regression models to “Schwartz Criterion”) are used to select the best fit population growth rate in relationship with high way model. “Spatial Lag” Model outperform than expansion in 1980-1990, demographic factors, “Spatial Error” model. As we observed that spatial socioeconomic conditions, transport availability, regression methods are applied in different study land conversion, natural amenities and growth area, (Hession and Moore, 2010) analyzed the factors (total 37 important factors) are analyzed. “Spatial Lag”, “Spatial Error” and “Spatial Autoregressive Moving Average” Models are used on highway expansion dataset. Figure 1: Total Confirmed Covid-19 Cases per 100,000 populations, January 2021 International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

4-Nearest Neighbor weight method is used for P a g e | 130 spatial weight matrix and Moran’s I statistic is computed to check fitness of standard linear 1.6 Methodology regression model. Authors concluded that SARMA Standard deviation normalization method is applied model has clear advantages compared with other on raw data to adjust the range of input data to avoid models by evaluating over AIC and BIC values. the bias caused by raw values. Before preforming spatial analysis, the existence of spatial correlation Spatial regression models also participated in must check prior by using “Local Moran’s I” Covid-19 research area. (Vimal Kumar et al., 2020) statistics. There are other kind of statistics to check performed percentage of covid-19 incidence cases spatial correlation but Local Moran’s I statistics is on age group (less than 25, 26-34, 35-44, 45-54, 55- applied in this study. In order to perform “Local 64) from the records released by World Health Moran’s I” statistic computation, spatial weight Organization of all affected countries. Statewide matrix is required. K-Nearest Neighbour Weighting Gender analysis on Covid-19 Incidence cases of method is applied to compute spatial weight matrix. India was carried out and the results state that every K-Nearest neighbour method was selected because 2 men 1 women is affected by the virus. Age-wise there are isolated regions (Singapore and analysis based on symptoms (fever, cough, throat Philippines). Local Moran’s I statistic is computed pain, chills, breathiness, join pain, cold, fatigue, and plotted to check the spatial dependency among diarrhea, pneumonia, headache, sputum and the study regions. After spatial dependency is found malaise) are performed on age groups (0-19, 20-39, out, non-spatial regression method (Ordinary Least 40-59, 60+) and age group 60+ is more affected Square method) and spatial regression methods when it compares to other age group. And mortality (“Spatial Lag” and “Spatial Error” Model) are rate of Covid-19 infected people with diabetes are computed. The difference of non-spatial and spatial about threefold of Covid-19 infected people with no regression is non-spatial way is the traditional diabetes. regression method which do not consider the spatial dependency. Spatial lag model is computed by 1.4 Hypotheses considering the spatial dependency between The tested hypotheses of this study are as followed: dependent variables and spatial error model is countries with high population factors may have computed by considering spatial error dependency. high Covid-19 confirmed cases, countries with high GDP, HDI, LEI may have low Covid-19 cumulative Figure 2: System flow diagram cases, regions with high International Inbound Tourist Index may have high Covid-19 occurrence rate and countries with low healthcare factors and high smoker ration may have high Covid-19 death cases. 1.5 Data The required data such as shape file of ASEAN countries, required statistical indexes and covid-19 data are collected. Before further processing, required shape file is created by using ASEAN countries shape file and collected information. The required shape files are downloaded from https://www.diva-gis.org/gdata. Required Covid-19 Data from December 2019 to January 2021 are collected from ourworldindata website (https://ourworldindata.org/coronavirus-testing) and checked with data from World Health Organization website. Life Expectancy Index, Human Development Index, Gross Domestic Profit, Total Population, Urban Population Percentage data are collected from United Nations Development Programme. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 131 1.6.1 Standard Deviation Normalization 1.6.3 Moran’s I Statistics Data normalization process is a mandatory step to Spatial correlation analysis is performed by Moran’s alter the values of statistical columns in the data to a I statistic. The results can be ranged (-1 to 1). fixed scale, without changing variations in the value Positive results indicate positive spatial correlation ranges (Bidyuk et al., 2020 ). To be cleared, input while negative correlation is indicated by negative feature that has value ranges from 0 to 100 will put results. Zero result indicates that there is no spatial more weight than input features that has [-1,1] range. autocorrelation or no linear correlation. The basic Proceeding to further analysis without data formula of Moran’s I is as below: normalization will create bias which will give the large range weight in features with wide range. DEFG HI J ∑456 ∑K56 * \" \"K \" \"K ⁄∑456 ∑K56 * ∑456 \" \"K ⁄ Normalization: 789:;<8 =>894 789:;<8 Equation 2 ?:94@9<@ A8B 9: C4 789:;<8 where xi is the value at ith location, xj is the value at Equation 1 jth location, which is the neighbour of xi, xm is the mean value of the variable x, wij is weighted 1.6.2 K-Nearest Neighbor Weighted Method coefficient value at (i,j) location of weight matrix In this study, K-Nearest Neighbour method (k=4) is applied to construct the neighbourhood matrix and and n, m are the row and column lengths of x. each region will have four neighborhood values with different weights based on their distance. 1.6.4 “Local Indicators of Spatial Association” (LISA) by Moran’s I Indicator “Local Moran’s I” statistic is employed to detect neighbouring groups and spatial outliers. As shown in the LISA map of Figure (4), only one country (Philippines) has statistically significant spatial clustering and others have non-significant spatial correlation. The results indicate that spatial correlation is exist between ASEAN countries. 1.6.5 Ordinary Least Square Method Standard Linear Regression method consider error terms are identically, normally, and independently distributed and dependent variables. Figure 3: Distance Connectivity Graph for K- Y=A+Bx Equation 3 Nearest Neighbor (k=9) Method in Study Area here y is predicted value, B is slope of the line, A is Once the spatial neighbourhood matrix is calculated, intercept of the points and X is predictor values. weighted matrix of input data can be calculated for further processing. Figure 4: Local Moran's I map for each indicator applied for regression models International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 132 Figure 5: Basic Concept for Linear Regression Figure 7: Basic Concept for Spatial Error Model 1.6.6 Spatial Lag Model 2. Experimental Resuls Total confirmed cases ~ population density + Y= ρwy+xβ+ε population total + population urban Equation 6 Equation 4 Where y is dependent variable vector, ρ is the Table 1 is showing the result of regression models autoregressive coefficients, W is a weight matrix for total confirmed cases and population factors. which required to compute Wy, Wy is a spatially Population density has negative correlation in all lagged dependent variables, β is a vector of models while other two factors are stated as positive coefficients of the regression model, and ε is a correlation. Population total is the most significant vector of independent and identically distributed factor for all three models, and it has significant p- error terms (Moore, 2010). value (<0.05) in OLS and SLM models. To choose the best fit model for population factors, R-Squared Figure 6: Basic Concept for Spatial Lag Model and AIC values are considered. For R-Squared values, SLM model has highest value (0.9677) 1.6.7 Spatial Error Model compared to other two models and for AIC values, SLM model has the lowest value (9.00565) among Y= β 0+Xβ+ ρw ε + ξ three regression models. Theory stated that model having highest R-Squared and lowest AIC values Equation 5 are the best fit model. Therefore, SLM model is the best fit model for total confirmed cases vs Where Y denotes response variables vector, X population factors. denotes independent variables matrix, W is a Total confirmed cases ~ gross domestic profit per capital + human development index + life weighted matrix which is used to compute spatial expectancy index + international inbound tourism lagged error terms Wℇ, β means regression index coefficients of explanatory variable and ξ denotes Equation 7 the error terms vector that are distributed Correlation results in Table 2 stated that GDP has independently but not identically. β0 is the intercept negative correlation on confirmed covid-19 cases in all models which means the higher the GDP, the of the regression model, ρ denotes autoregressive lower the covid-19 cases. Human development index has positive correlation on total confirmed coefficients for the error terms, Wℇ is lagged error covid-19 cases in OLS and SEM models but vector of independent error terms (Moore, 2010). positive correlation in SLM model. Also, life expectancy index is indicating negative correlations in OLS and SEM whereas it has positive correlation in SLM model. International inbound tourism index are showing negative correlations only. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 133 Table 1: Comparison of Regression Results on Total Confirmed Cases ~ Population Factors OLS Model Spatial Lag Model Spatial Error Model Variables Coefficient P Coefficient P Coefficient P Population Density -0.06318 0.78099 -0.09212 0.728 -0.104018 0.1834 Population Total 0.91999 0.00136 0.90281 0.006 1.034856 0 ** ** Population in Urban 0.22940 0.32311 0.12427 0.694 0.343457 0 Constant 0.02203 0.369 -0.608086 0 R-Squared 0.8487 0.30959 P-Value 0.007139 0.9677 0.93777 AIC 18.44313 0.02445 0.017694 9.00565 18.443 Table 2: Comparison of Regression results on Total Covid-19 Cases and Socioeconomic Factors OLS Model Spatial Lag Model Spatial Error Model Coefficient P Variables Coefficient P Coefficient P -3.5679 0. 565 GDP per capita -1.580 0.167 -4.0918 0.887 -1.760267 0.0078232 7.6986 0.823 Human Development Index 2.428 0.114 -3.4002 0.807 2.662190 0.0008771 -0.2595 0.832 Life Expectancy Index -1.0044 0.34 0.8511 -1.015610 0.0866924 0.7292 International Inbound Tourist -0.4582 0.481 28.27805 -0.466270 0.2402152 Constant 1.710 0.394 -0.37464 0.6805287 R-Squared 0.453 0.640884 P-Value 0.472 0.00034183 AIC 33.29263 33.293 Table 3: Comparison of Regression results on Total Covid-19 Deaths ~ Healthcare factors OLS Model Spatial Lag Model Spatial Error Model Coefficient P Coefficient P Variables Coefficient P Cardiovascular Death Rate 0.47416 0.214 0.3575 0.2001 0.21214 0.239778 Diabetes Prevalence Rate -0.02808 0.938 -0.7451 0.0491** -0.79179 0.004125 0.757 0.024253 Constant -0.25077 0.455 0.4809 1.06442 R-Squared 0.2337 P-Value 0.394 0.819 0.386051 32.66366 0.0425 0.037966 AIC 22.23467 32.664 Table 4: Comparison of Regression results on Total Death ~ Behavior Factors OLS Model Spatial Lag Model Spatial Error Model Variables Coefficient P Coefficient P Coefficient P Female Smoker Ratio 0.1492 0.49565 0.17874 0.35541 0.091559 0.568 R-Squared 0.6983 0.8741 0.699202 P-Value 0.01508 0.02849 0.045836 AIC 23.34192 20.54391 23.342 International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 134 Human development index is the most influenced 3. CONCLUSION AND FURTHER STUDIES factor in OLS and SEM models and life expectancy The experiment results of “Ordinary Least Square”, index is the most influenced one in SML model. “Spatial Lag” and “Spatial Error” Model Regression SLM has the highest R-squared and also has lowest methods support the stated hypothesis in the factors AIC scores. Therefore, SLM model will be except International Inbound Tourism Index and considered as the best fit model. Diabetes Prevalence Rate which are showing negative correlation instead of positive correlation.. Total Death ~ Cardiovascular Death Rate + Instead of using raw data collected, some indexes Diabetes Prevalence Rate that can be calculated based on the raw data (such as Equation 8 age-group, urban population ratio) can be added to get better experimental results. Since the study Table 3 is showing regression results of Total concentrate on the confirmed Covid-19 incidents Covid-19 Death Cases and Healthcare factors and deceases, the study lacks information about the (Equation 8). Healthcare Factors (Cardiovascular number of tested cases per day. It is very cleared Death Rate and Diabetes Prevalence Rate) are used that if the number of Covid-19 tested case is low, to find correlation with total Covid-19 deaths the number of confirmed Covid-19 cases will be because people with underlying Cardiovascular and low. If the number of tested cases per day data will Diabetes disease may suffer severe symptoms and be available and then the ratio of confirmed cases this may lead to high mortality rate. Correlation and tested cases can also consider as one factor. Results stated that Cardiovascular Death Rate has positive correlation on Total Covid-19 Death while REFERENCES Diabetes Prevalence Rate is showing negative correlations. Cardiovascular Death Rate is the most Bidyuk, P., Gozhyj, A., Kalinina, I., Vysotska, V., influenced factor in OLS model, and Diabetes Vasilev, M. and Malets, R., 2020, Forecasting Prevalence Rate is the most influenced for in Spatial Nonlinear Nonstationary Processes in Machine Error Model. Diabetes Prevalence Rate has Learning Task. IEEE Third International significant effect on Spatial Lag Model. Spatial Lag Model is chosen as fittest model for Total Covid-19 Conference on Data Stream Mining & Deaths versus healthcare factors. Processing (DSMP), Lviv, Ukraine. DOI: 10.1109/DSMP47368.2020.9204077 Total Death ~ Female Smoker Ratio + Male Smoker Chi, G. and Zhu, J., 2008, Spatial Regression Ratio Models for Demographic Analysis. Springer, Equation 9 Vol. 27, 17–42, DOI 10.1007/s11113-007-9051- 8. Table 4 is showing results of regression model for Hession, S. L. and Moore, N., 2010, A Spatial Equation 9. Behavior factors (Female smoker ratio Regression Analysis of the Influence of and Male Smoker Ratio) are used to find correlation Topography on Monthly Rainfall in East Africa. with Covid-19 deaths because John Hopkin International Journal of Climatology, Vol. University stated people with smoking habit may 31(10), 1440–1456, DOI: 10.1002/joc.2174. suffer severe symptoms and this may cause high Putra, A. S., Guangji Tong and Pribadi, D. O., 2020, mortality rate. Correlation Results of Table 5 stated Spatial Analysis of Socio-Economic Driving both factors have positive correlation on Total Factors of Food Expenditure Variation between Covid-19 Death which support the hypothesis. Male Provinces in Indonesia. MDP Substainability, Smoker Ratio is the most influenced factor in all Vol. 2(4), https://doi.org/10.3390/su12041638. models and it also has significant effect (p- Voss,P. R., Long, D. D., Hammer, R. B. and value<0.05) in OLS and SLM Models. Diabetes Friedman, S., 2006, County Child Poverty Rats Prevalence Rate has significant effect on Spatial in the US: A Spatial Regression Approach. Lag Model. Popul Res Policy Rev, Vol. 25, 369–391, DOI 10.1007/s11113-006-9007-4. Vimal Kumar, M. N., Jaya, R., Rubesh, C. M. and Aakash Ram, S., 2020, Statistical Analysis on Novel Corona Virus: COVID-19. European Journal of Molecular & Clinical Medicine, Vol. 7(1), 95-103. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 135 STATISTICAL AND SPATIAL ANALYSIS ON OBESITY PREVALENCE IN UNITED STATES Ranadheer Mandadi1 and Nitin Tripathi2 Asian Institute of Technology, Thailand E-mail: [email protected],1 [email protected] ABSTRACT This study provides an overview and spatial analysis of obesity incidence at the state level in the United States of America using spatial analysis. Analysis indicated that obesity rates are higher in southeast states of the U.S. and confirmed that there is a statistical and spatial relationship between socio-economic and demographic to obesity. Behavioral factors and health variables have a positive relationship, however, few specific factors have low correlation mainly due to the outliers. This study is focused on analyzing the spatial incidence of obesity and the spatial relation between factors influencing obesity. The hotspot analysis revealed that the patterns of obesity prevalence in United States (US) at a state level are observable in 11 states located in southeast of United States with 99% confidence interval while 6 states are with low obesity and are mostly clustered in the northeast part of the US with an uneven confidence interval. The Moran’s I statistical analysis revealed that the obesity prevalence in the different states of US in not random, rather it is clustered. Results illustrate that obesity prevalence is high in black race with correlation coefficient 0.327* at 95% significance followed by white race with a correlation coefficient of 0.125. On the other hand, Hispanic, Asians and Multiracial have negative correlation with obesity. There is a relationship between economic, disparity and obesity prevalence in United States. Although research has linked obesity prevalence to different economic factors, other variables are often excluded; hence this study will incorporate factors that are often omitted such as behaviors factors, literacy level and health outcomes. Data such as demographic, economic, behavioral and health were collected for the year 2017 from the USA Census bureau datasets, health and medical data from USA Centres for Disease Control and Behavioral Risk Factor Surveillance System (BRFSS). These data were analyzed using cluster Hot spot analysis tool (Getis-Ord GI), Spatial Autocorrelation (Global Moran's I) to assess the role of location in health analysis. Pearson correlation two- tailed test and Bivariate Local Indicator of Spatial Association (LISA) were used to determine statistical correlation and spatial relationships between variables and location. KEYWORDS: Obesity, Spatial Autocorrelation, SPSS Pearson Correlation, Hot Spot Analysis and Bivariate LISA 1. INTRODUCTION population such as age, race/ethnicity and The word \"epidemic\" is often overused, but there is gender(Anderson and Butcher, 2019). It is higher in no better way to describe the explosion of obesity in women than men are (35 versus 33%) are, like other America. According to the latest numbers from the regions of the world. In addition, US has a smaller Center for Disease Control and Prevention (CDC), population than China and India, the United States an astonishing 68 percent of American adults are had the greatest number of obese adults followed by overweight (meaning they have a body mass index, China (Chooi and Ding, 2019). or BMI, of 25 or more) or obese. About 18 percent of children and adolescents are seriously overweight Obesity epidemic cannot be explained because as well (Chooi and Ding, 2019). The obesity rates in of a single cause or factor; therefore, it cannot be the USA have doubled in the last four decades; tackled with one single intervention. This epidemic more than two-thirds of Americans were classified needs to incorporate all the complex network of as overweight in 2015-2016 and during the same factors such as policy, economics, environment, year, obesity prevalence in all states had exceeded social influences, behaviour and physiology. To 20%. According to CDC-USA, the prevalence of understand and address complex problem obesity, a obesity among USA adults was 39.8%(Hales, 2017). unified, strategic approach is required (Lee et al., USA has the highest rates of obesity among 2017). Evidence indicates that spatially describing developed countries (32%); obesity varies in obesity inequalities at the state level is appropriate International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 136 because health and economic policies implemented to (1) investigate the prevalence of adult obesity at this geography are potentially influential (Jia et patterns in U.S and (2) to investigate the statistical al., 2019). and spatial relation between socio-economic, demographic, behavioural factors and health It was not until the year 2000 though, that the outcomes with adult obesity in U.S. number of people who were overweight or obese was greater than the number of those who were 2. METHODOLOGY underweight (Eknoyan, 2006). The obesity trend in the United States of America may actually have 2.1 Study Area originated in the early 20th century, during which it United States is a country with a population to be was discovered that poor children's health improved 327,167,434 as of July 1, 2018. It is the 3rd most tremendously when their malnutrition was corrected populated nation after China and India. Its by providing them access to more calories, namely population has diversity mainly Asian, White, from inexpensive sugars and fats (Popkin, 2010). Black, Hispanic and Multiracial. United States has a Low-cost foods provided to the working class life expectancy of 78.6 in 2017, which was the third improved overall industrial, and, subsequently, year of life expectancy following decades of economic productivity. Improved industrial continuous increase. Increasing obesity in the technology created ways in which producing cheap, United States and health improvements elsewhere high-calorie foods became even easier. This was contributed to lowering the country’s rank in life coupled with the development of technology that expectancy from 11th in the world in 1987, to 42nd in made life more sedentary such as cars, dishwashers 2007. Obesity rates have more than doubled in the and washing machines and created a situation where last 30 years, are the highest in the industrialized it was easy to consume an excess of calories world, and are among the highest anywhere. (OECD/EU, 2017). In the United States, the Approximately one-third of the adult population is prevalence of obesity barely changed during the obese and an additional third is overweight. 1960s and 70s, but escalated sharply starting in the 1980s. In 1980, the obesity rate was 13.4 percent, 2.2 Data Used but skyrocketed to 34.9 percent as per the 2011 to State-level demographic data, health, socio- 2012 National Health and Nutrition Examination economic and behavioural information were Survey, which was reported in a 2012 issue of the obtained from various sources. State level shape Journal of the American Medical Association files for the United States of America were obtained (Popkin, 2010). Early in the 20th century, obesity from US Census Bureau TIGER products that was mostly a problem in first world countries of contain spatial data for use in GIS. There are 51 Europe and the United States. In 1997, though, the states in U.S (Table 1). World Health organization recognized obesity as a global epidemic as rates rose in countries such as 2.3 Variables in the Study Mexico, Brazil, China and Thailand. The study finds the statistical and spatial relation between socio-economic, demographic, and Thus, this study focuses on the importance of behavioural and health factors with obesity (Pasco, location in understanding obesity. The overarching 2012). Based on research gap found in the previous goal is to offer a place-based approach to studies these variables considered. This study understanding obesity in United States. This study focuses on the importance of location in aimed to review the findings and methodology of understanding obesity (Table 2). studies on the prevalence, risk factors of obesity in the United States of America. Specifically, it aimed Table 1: Spatial Data (https://www.census.gov/geo/maps-data/data/tiger.html) Spatial – data Source State level shape file US Census Bureau TIGER products Table 2: Variables and source of data Variable Source of Data Health Data CDC - Centre for Disease Control and Prevention Behavioural Risk Data Behavioural Risk Factor Surveillance System Population (race/ethnicity) United States Census Bureau Household Income United States Census Bureau Educational Level United States Census Bureau International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 137 2.4 Software Used neighbouring states with high obesity prevalence at In this study, we performed statistical and spatial a 95% confidence and is observed in 3 states. On the analysis to investigate the location specific obesity other hand, there are 6 states with low obesity and prevalence and find the relation between obesity and are mostly clustered in the northeast part of the US associated factors. To perform this analysis, we with an uneven confidence interval. The output for need powerful Statistical software; hence we take hotspot analysis is a map indicating where high and the help of SPSS and Excel. Spatial analysis is low adult obesity are clustered. The results (Figure performed by using ArcMap and GeoDa, which 1) indicate that 11 states are hotspots with 99% helps in visualising and analysing the patterns confidence, 3 states hotspots with 95% confidence (Table 3). and 2 states with 90% confidence. On the other hand, results also show that there are 13 cold spot 3. Results states (low obesity prevalence) in the United States. The hotspot analysis explains the patterns of obesity The presence of high and low clusters indicates that prevalence in United States (US) at a state level. adult obesity in United States is not randomly Obesity hotspots are observable in 11 states located distributed. Considering the results, the adult obesity in southeast of United States with 99% confidence within US is clustered and most of the states are in interval. This confidence interval explains that the hotspot zones. states with high obesity prevalence have Table 3: Software used Statistical Analysis Spatial Analysis SPSS ArcMap 10.5 Excel GeoDa Figure 1: Obesity Hot Spot Analysis Map International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 138 3.1 Spatial Autocorrelation Analysis analysis is adopted to find the spatial relation with The spatial autocorrelation is measured of how obesity using Bivariate LISA. This analysis would much close objects are in comparison with other be useful for to understand relation between obesity close objects, and the results can be classified into and black / white race. The map explains the positive, negative and no spatial autocorrelation. locations where obesity prevalence is high in Positive spatial autocorrelation is when similar relation with race (black / white). The map explains values are clustered together on a map. Negative that obesity prevalence is not random, and it is spatial autocorrelation is when dissimilar values are clustered in southeast states (Table 4). clustered together on a map. These tools calculate the Moran’s I index value, a z-score and p-value to 3.3 Correlation between Obesity and Household evaluate the significance of the index. For this test, the results of Moran’s I Index is 0.496, a z-score of Income 4. 956 at a p-value of 0.000001. According to the There is a relationship between economic, disparity Spatial (Moran’s I) analysis, if the Z test statistics and obesity prevalence in United States. From the >1.96 (or < -1.96) the null hypothesis is rejected. two-tailed Pearson Correlation analysis in SPSS According to the results, the Z score is greater than exhibits that there is a negative correlation between 1.96, so the null hypothesis is rejected. This low income and healthy -0.365 with 95% of indicates that adult obesity is not random in United confidence. On the other hand, there is a positive States. correlation between low income and obesity. Income influences the choices and variation of food. 3.2 Correlation between Obesity and Low income limits the choices of healthy options and this is seen in families with household income Race/Ethnicity level of 15k-34.9k income level with 95% As per the data race is classified in to 4 categories. confidence. High income level income (50k+) is The results illustrate that obesity prevalence is high positively correlated (99% confidence) with healthy in black race with correlation coefficient 0.327* because the choices, options and capability to access with 95% significance and second place is whites healthy food is higher. The spatial analysis gives a with correlation coefficient 0.125. Whereas better understanding about obesity and household Hispanic, Asians and Multiracial have negative income as we see that household income has a correlation with obesity and their significance is not positive correlation with obesity. The Bivariate- the same (-0.305* with, -0.367** with 99% and - LISA explains that the states with high obesity and 0.152). surprisingly white race has a low positive income variation are been explained by spatial correlation 0.125. The results exhibit that Black and analysis (Table 5). white race have positive correlation. Further spatial Table 4: Statistical correlation between obesity and Race/Ethnicity WHITE BLACK HISPANIC ASIANS MULTIRACIAL .125 .327* OBESE Pearson -.305* -.367** -.152 Correlation .381 .019 51 51 .030 .008 .288 Sig. (2-tailed) 51 51 51 N Table 5: Pearson-Correlation between House Hold Income and Obesity < 15 k $ 15K - 24.9K 25K - 34.9K 35K - 49.9K 50K + -.365** -.452** .526** HEALTH Pearson Correlation .008 .001 -.410** -.275 .000 Y 51 51 51 .003 .051 Sig. (2-tailed) .466** .537** -.624** N .001 .000 51 51 .000 OBESE Pearson Correlation 51 51 .453** .309* 51 Sig. (2-tailed) .001 .027 N 51 51 International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 139 3.4 Correlation between Obesity and Education 3.5 Behavioural Risk Factors - Correlation Education has a high influence on health outcomes as discussed in literature review, literacy of a person between Obesity and Smoking brings out the health awareness, hygiene, sense of A causal effect of smoking on weight is possible control and empowerment. Statistical analysis because nicotine is both an appetite suppressant and shows that obesity prevalence is likely to happen in metabolic stimulant. Result shows that people who people with low education level as indicated in the are smoking have a negative correlation with Pearson-Correlation coefficient of 0.539 and 0.528 obesity while those who have quitted smoking has a in respondents with less than high school and high positive correlation. This is because the sudden school with 99% confidence in support to analysis absence of nicotine increases the appetite, thus, healthy BMI has a negative correlation with people people tend to gain weight after quitting smoking. who have an education less than high school and However, statistical analysis also shows that among high school (-0.470 and -0.511) with 99% the classification of smokers, those who smoke confidence. People who obtained college has a more intensively tend to weigh more. negative correlation with obesity because of awareness and exposure to health eating. On the Smoking every day and obesity has a negative other hand, people who obtained college have a correlation with 99% significance. However, former positive correlation with health BMI (0.586) with smoker has a positive correlation with obesity 99% confidence. In addition to this, obesity has a (0.631) with 95% confidence. A non-smoker person negative correlation with people who obtained has a negative correlation with obesity (-0.528) with college (-0.606) with 99% confidence. a 95% confidence. In the US, smoking is a factor that should not be neglected in consideration with The Pearson-Correlation analysis shows the obesity. Though smoking does not show direct relation between literacy levels and obesity relation with obesity, former smokers show a high prevalence to find the spatial relationship is positive correlation. Smoking in the US can be a obtained by performing Bivariate-LISA for the seasonal habit in response to seasonal changes. In literacy variables in relation to obesity. Results cold seasons people tend to smoke more, and when show that literacy is inversely correlated with they try to quit smoking they gain weight. Spatial obesity prevalence. As the literacy rate increases, analysis gives the picture of high obesity prevalence obesity prevalence in the state decreases (Table 6). and high smoking areas (Table 7). Table 6: Pearson correlation between literacy and obesity Less than High School High School COLLEGE HEALTHY Pearson Correlation -.470** -.511** .586** OBESE Sig. (2-tailed) .001 .000 .000 N 51 51 51 Pearson Correlation .539** .528** -.605** Sig. (2-tailed) N .000 .000 .000 51 51 51 Table 7: Pearson-Correlation between smoking patterns and obesity SMOKE EVERY DAY FORMER SMOKER NEVER SMOKE -.528* OBESE Pearson Correlation -.562** .631** .019 Sig. (2-tailed) .000 .018 51 N 51 51 Table 8: Pearson correlation with alcohol patterns and obesity OBESE Pearson Correlation HEAVY DRINKER BINGE DRINKER Sig. (2-tailed) -.488** -.397** .000 .004 N 51 51 International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 140 Table 9: Pearson-Correlation with health outcomes and obesity OBESE Pearson Correlation Coronary Stroke Diabetes Heart attack Sig. (2-tailed) .575** .544** .664** .593** .000 .000 .000 .000 N 51 51 51 51 3.6 Correlation between Obesity and Alcohol spatial analysis between variables and obesity, most of the patterns are clustered and observed in Drinking southeast region (Table 9). According to the literature, alcohol has a positive correlation with obesity. Heavy drinking and binge 4. Discussion drinking have been more consistently linked with The focus of the study was to understand the obesity adiposity. Recreational alcohol intake is a patterns in United States. The main assumption was widespread activity globally and alcohol energy can that, unlike the popular beliefs, incidence of obesity be a contributing factor to weight gain. However, varies by place and across different scales. The literature says that light-to-moderate alcohol intake overarching goal was to contribute to the existing is less likely to be a risk factor for obesity than body of knowledge on obesity and to help provide a heavy drinking. Statistical analysis shows that place-based approach applicable in formulating among the respondents, heavy and binge drinkers policy and usable in tailoring public health are negatively correlated with obesity. However, interventions directed towards management and adiposity is considered as a third grade of obesity prevention of obesity epidemics. According to the where as in the data, BMI greater than 30 – 100 is analysis, the obesity rate was identified to be categorized as obesity. Hence, data limitation clustered with Moran’s I value of 0.495712. The creates a possible situation to have a negative difference in the obesity rates spatially was correlation with alcohol (Table 8). determined using hotspot analysis indicating that 15 states are in hot spot zones out of which 10 are with 3.7 Correlation between Obesity and Health 99% confidence, 3 are with 95% and 2 are with 90% confidence. On the other hand, results also show Outcomes that there are 13 cold spot states (low obesity There are four health outcomes that are considered prevalence). Among these, 1 state has a 99% and are specified below. According to the Pearson- confidence, 8 states with 95% confidence and 4 Correlation analysis, obesity is positively correlated states with 90% confidence. The presence of high to the incidence of health outcomes at a 99% and low clusters indicates that adult obesity in confidence. People that are obese are more likely to United States is not randomly distributed. suffer from coronary artery diseases (0.575), stroke Considering the results, the null hypothesis is (0.544), diabetes (0.664) and heart attack (0.593). rejected – indicating that adult obesity within US is Among the health outcomes mentioned, diabetes clustered and most of the states are in hotspot zones. obtained the highest correlation value. Spatial analysis is used to map the states with high obesity The Pearson correlation analysis shows that with high diabetes, stroke, coronary artery diseases most of the variables are positively correlated with and heart attack. The result shows that spatial obesity. The race/ethnicity results illustrate that the patterns are clustered similarly to all health obesity prevalence is high in black race (0.327) with outcomes and concentrated in southeast states. 95% conference and observed in 8 states in down south east, these states are clustered. The US obesity The result of this study indicates that there is a epidemic is disproportionately affecting racial significant relationship between geographic location groups where more African American women are and incidence of obesity at the state level. As the more obese (50%) compare to only 33% of white results show that, the southeast states have high women (Kirby et al., 2012). Low household rates of obesity. Most of the states in hot spot zones incomes have a positive influence on obesity this is are in southwest regions while the coldspots are seen in income ranging from <15k to 34.9k with predominantly in the southwestern region. The 99% confidence. The higher income classes show result also indicates that there is a statistical inverse correlation with obesity (-0.624). It has been relationship between socio-economic, demographic, suggested that people who live in impoverished behavioural and health variables with obesity. Only condition have poor access to affordable healthy alcohol shows a negative correlation because of data food (Levine, 2011). In addition, (Ogden et al., limitation. On the other hand, smokers have a 2017) found similar result where obesity prevalence negative correlation with obesity while former smokers have positive correlation. According to the International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 141 was lower in high income group for both men and regions. Specifically, the south was identified as a women group. high obesity spatial regime, while the north and northwest were shown to be low obesity spatial There is an interesting result that was found that regimes. This is a unique contribution to the as the income level increases the states in hotspots literature because it shows that obesity in certain decreases from 11 states in south east to 2 states regions of the country is structurally different from which are distributed in east and south. Literacy rate obesity in other regions. One notable finding from has also shown a positive correlation with obesity, this research is that the greatest concentration of people who have education less than school and elevated adult obesity prevalence in the country was high school are directly correlated with obesity in a large contiguous region in the south that (0.539 and 0.528) with 95% confidence and spatial spanned from Arkansas, Mississippi, Louisiana, patterns also show the similar location with 10 hot Alabama, Tennessee, Georgia, Indiana and South spot states in south east. People who obtained Carolina. This underscores calls for special attention college have a negative correlation with obesity (- to the social, economic, political, and culture factors 0.605). In a study, result showed that obesity that are linked to poor population health in the prevalence was lower among women and men south. Additionally, two secondary notable (27.8% and 27.9%) who were college graduates concentrations of high adult obesity prevalence were compared to women and men (45.3%, 35.5%) who also shown in Kentucky/West Virginia and North were high school graduate or less (Ogden et al., Carolina/South Carolina. These two areas are also 2017). Behavioural risk factors have an interesting part of the U.S. South. This provides further correlation with obesity. Smoking every day has a evidence that in terms of concentrated obesity negative correlation with obesity (-0.562) as mainly prevalence the South needs to be a focal point for due to nicotine that supresses the appetite while research and public policy. To combat obesity, U.S former smoker has a positive correlation with is adopting the Supplemental Nutritional Assistance obesity (0.631). A popular belief is that smoking is Program (SNAP). It is a program that gives an efficient way to control body weight while incentives to buy healthier foods and providing quitting smoking can lead to gaining weight education to raise awareness about diet and nutrition (Chiolero et al., 2008). This is because the sudden to young children. The SNAP Program serves 47 absence of nicotine increases the appetite, thus, million people each month, nearly half of whom are people tend to gain weight after quitting smoking. children. Bivariate Spatial analysis gives a result that there are 2 states that are having high obesity prevalence 6. RECOMMENDATION after quitting smoking. This observation is The study finds a meaningful correlation and specifically found in east of U.S. Alcohol has a relation between obesity prevalence and factors negative correlation with obesity this mainly due to (race, household income, literacy disparities, the data limitation for heavy drinking and binge smoking, alcohol patterns, and health outcomes). drinking that is mostly linked with adiposity which Based on this finding, we understand that there are considers height and the hip circumference ratio. In several other things need to be explored and obesity the data, BMI greater than 30 – 100 is categorized data has to be more in detail and the environmental as obesity. According to the Pearson-Correlation factors, food consumption and physical activities analysis, obesity is positively correlated to the would be important factors. With the current data it incidence of health outcomes at a 99% confidence. is possible to answer the research questions but it is People that are obese are more likely to suffer from recommended that a place-based analysis for obesity coronary artery diseases (0.575), stroke (0.544), should consider sub-county level data for analysis. diabetes (0.664) and heart attack (0.593). Among The data at the zip code, census block or census the health outcomes mentioned, diabetes obtained tract level would have been more informative for the the highest correlation value. The result shows that analysis. spatial patterns are clustered similarly to all health outcomes and concentrated in southeast states. REFERENCE Among these diabetes and obesity is show high in 11 states in southeast. Anderson, P. M. and Butcher., 2019, Understanding Recent Trends in Childhood Obesity in the 5. CONCLUSION United States. Economics & Human Biology, 1– The current study aimed to identify significant 10. https://doi.org/10.1016/j.ehb.2019.02.002. regional differences in adult obesity prevalence in the U.S. our findings demonstrated the existence of spatial regimes of obesity prevalence across U.S. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 142 Chiolero, A., Faeh, D., Paccaud, F. and Cornuz, J., Kirby, J. B., Liang, L., Chen, H. J. and Wang, Y., 2008, Consequences of smoking for body 2012, Race, Place, And Obesity: The complex weight, body fat distribution, and insulin Relationships among Community Racial/Ethnic resistance. American Journal of Clinical Composition, Individual Race/Ethnicity, and Nutrition, Vol. 87(4), 801–809. Obesity in the United States. American Journal https://doi.org/10.1093/ajcn/87.4.801. of Public Health, Vol. 102(8), 1572–1578. Chooi, Y. C. and Ding, C., 2019, The Epidemiology https://doi.org/10.2105/AJPH.2011.300452. of Obesity. Metabolism: Clinical and Lee, B. Y., Bartsch, S. M., Mui, Y., Haidari, L. A., Experimental, Vol. 92, 6–10. Spiker, M. L. and Gittelsohn, J., 2017, A https://doi.org/10.1016/j.metabol.2018.09.005. Systems Approach to Obesity. Nutrition Eknoyan, G., 2006, A History of Obesity, or How Reviews, Vol. 75, 94–106. https://doi.org/- What Was Good Became Ugly and Then Bad. 10.1093/nutrit/nuw049. Advances in Chronic Kidney Disease, Vol. Levine, J. A., 2011, Poverty and Obesity in the U.S. 13(4), 421–427. https://doi.org/10.1053/j.ackd- .2006.07.002. Diabetes, Vol. 60(11), 2667–2668. https://doi.- Hales, C. M., 2017, Prevalence of Obesity Among Adults and Youth: United States, 2015–2016. org/10.2337/db11-1118. NCHS Data Brief, No 288. Hyattsville, MD: National Center for Health Statistics. NCHS OECD/EU, 2017, Obesity Update 2017. OECD Update Report, Vol. 13(5), 331–341. https://doi.org/10.1007/s11428-017-0241-7. Ogden, C. L., Fakhouri, T. H., Carroll, M. D., Hales, Data Brief, No 288. Hyattsville, MD: National C. M., Fryar, C. D., Li, X. and Freedman, D. S., Center for Health Statistics., Vol. 288, 2015– 2016. 2017, Prevalence of Obesity Among Adults, by Jia, P., Xue, H., Yin, L., Stein, A., Wang, M. and Household Income and Education — United States, 2011–2014. MMWR. Morbidity and Wang, Y., 2019, Spatial Technologies in Obesity Mortality Weekly Report, Vol. 66(50), 1369– Research: Current Applications and Future 1373. https://doi.org/10.15585/mmwr.mm- Promise. Trends in Endocrinology and 6650a1. Metabolism, Vol. 30(3), 211–223. https://doi.org/10.1016/j.tem.2018.12.003. Pasco, J. A., 2012, Prevalence of Obesity And The Relationship Between the Body Mass Index and Body Fat: Cross-Sectional, Population-Based Data. PLoS ONE, Vol. 7(1). https://doi.org/10.1371/journal.pone.0029580. Popkin, B. M., 2010, What’s Wrong with the U.S. Approach to Obesity? Virtual Mentor, Vol. 12(4), 316–320. https://doi.org/10.1001/virtual- mentor.2010.12.4.pfor2-1004. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 143 FACTORS RELATED TO FOOD CONSUMPTION BEHAVIOR FOR FRESHMEN OF THE COLLEGE OF ALLIED HEALTH SCIENCES, SUAN SUNANDHA RAJABHAT UNIVERSITY Phannee Rojanabenjakun, Jatuporn Ounprasertsuk, Tipvarin Benjanirat, Supaphorn Oasana, Pongsak Jareanngamsamear, Sasipen Krutchangthong, Sunatcha Choawai, Jirawat Sudsawad and Panupan Sripan [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected] ABSTRACT The purposes of this research were to study 1) level of food consumption behavior of the first-year students of College of Allied Health Sciences Suan Sunandha Rajabhat University, 2) level of knowledge on food consumption of the first-year students of College of Allied Health Sciences Suan Sunandha Rajabhat University, 3) the relationship between food consumption and behavior of food consumption, and 4) predisposing, enabling, and reinforcing factors related to the behavior of food consumption of the first-year students of College of Allied Health Sciences Suan Sunandha Rajabhat University. This research was quantitative research that samples were 112 undergraduate freshmen who studied at College of Allied Health Sciences Suan Sunandha Rajabhat University. A research tool was a survey with multiple questions proved by the descriptive statistics such as percentage, mean, standard deviation, Pearson’s correlation, and multiple regression analysis. The results found that 1) according to the frequency of food consumption behavior, the first-year undergraduate students were sometimes in overall consumption, 2) the students had good food consumption knowledge, 3) food consumption was related to the behavior of food consumption, and 4) enabling factors related to food consumption was “buying food from markets to prepare by self” at significance level .05. The research also recommended that the university should facilitate the students in food consumption, food knowledge, and nutrition. KEYWORDS: Behavior, Food Consumption, Nutrition 1. INTRODUCTION change such a behavior of the student in food Presently, adolescents' and undergraduate students' consumption which may help them to have better food consumption has been rapidly changing for health, and reduce the risk for obesity. The authors many reasons, such as socio-economic purposes and also believe that the results of this research will the environment, which affect the student's daily contribute to other universities and health promotion life. (Yimprasert, 2017). Nowadays, as our daily life center who concern on food consumption. is in a very urgent situation, people, including undergraduate students, choose a ready-to-eat or a 2. OBJECTIVES grab-and-go for their meals. They have not often considered food nutrition; most of them ignore its 1) To study level of food consumption behavior of importance and sometimes say no to breakfast. the first-year students of College of Allied Health Some students eat sticky rice with fatty pork grill Sciences Suan Sunandha Rajabhat University. for every breakfast. They think that this is an easy and quick choice. (Sudsaneha, 2013). Therefore, the 2) To study level of knowledge on food authors are interested in studying the factors related consumption of the first-year students of College of to food consumption behavior of the freshmen Allied Health Sciences Suan Sunandha Rajabhat undergraduate students who are studying at the University. College of Allied Health Sciences Suan Sunandha Rajabhat University, and seeking a direction to International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 144 3) To study the relationship between food Scope of the Research consumption and behavior of food consumption of This research focused on the factors related to food the first-year students of College of Allied Health consumption behavior of the first-year students of Sciences Suan Sunandha Rajabhat University. the College of Allied Health Sciences Suan Sunandha Rajabhat University. The researchers 4) To study predisposing, enabling, and reinforcing selected a quantitative research technique to find out factors related to the behavior of food consumption the results of the research. In this research, samples of the first-year students of College of Allied Health used were 112 first-year undergraduate students Sciences Suan Sunandha Rajabhat University. who were studying at the College of Allied Health Sciences Suan Sunandha Rajabhat University. The 3. LITERATURE REVIEWS research area covered the College of Allied Health For this research, the researchers reviewed the Sciences Suan Sunandha Rajabhat University in recent works from textbooks, academic articles, and Samut Songkhram. Data were collected from journals to design a conceptual framework. The January 2019 to October 2020. topics that all research had studied were (1) food consumption behavior and (2) knowledge of Hypothesis nutrition, and (3) factors related to food 1. Level of food consumption of the first-year consumption behavior as follows. student of the College of Allied Health Sciences Suan Sunandha Rajabhat University may relate to Food consumption behavior is a human behavior food consumption behavior. that responds to food in terms of physical and 2. The students may have good food consumption mental reactions (Mekwimon, 2016). Each human's knowledge. consumption behavior depends on many factors 3. Food consumption knowledge may be associate such as personal purposes, social culture, with food consumption behavior. neighboring people, and the environment 4. Predisposing factor, enabling factor, and (Steenkamp, 1993). However, as food consumption reinforcing factor affect consumption behavior of behavior affects human health, humans should first-year students of College of Allied Health select good food from good sources for their health. Sciences Suan Sunandha Rajabhat University. (Coon and Mitterer, 2013). According to food sources, good food should come from five food 4. METHODOLOGY groups consisting of protein, carbohydrate, fatty, This research was quantitative research that vitamin, and vegetables in order to follow a employed descriptive statistics like Mean, nutrition which is a critical part of health and Percentage, Standard Deviation, and Regression development. (World Health Organization, 2021). Correlation. The authors referred to academic Besides, the selected food should be produced with documents, textbooks, and related works as cleanliness to avoid any food poison and other risks. guidelines to initiate a conceptual framework for this study. A research instrument used for collecting Nutrition is a study of nutrients in food that data was a survey consisted of three parts of indicates how the body uses the nutrients. Nutrition multiple questions. The first part was an analysis of is related to diet, health, and disease. Nutritionists the level of consumption behavior, which had a always use molecular biology, biochemistry, and four-scale score. The second part was an analysis of genetics to understand how nutrients affect the the level of knowledge. The third part of the survey human body. Nutrition emphasizes how humans can was a Pearson’s analysis, and the fourth part was an use dietary choices to reduce the disease risk, what analysis of factors affecting consumption behavior will be if a human too much or too little of a and multiple regression analysis. nutrient, and how allergies work. Thus, the nutrients provide nourishment, proteins, carbohydrates, fat, 5. RESULTS vitamins, minerals, fibers, and water whether the The level of food consumption behavior of the human does not have the right balance of nutrients newcomers at the College of Allied Health Sciences in their diet, their risk of developing inevitable Suan Sunandha Rajabhat University. It was found health condition increase. Predisposing factors are that the students consumed “sometimes” with a the primary factors that persuade people to statistically significant difference of .05. By list, it consume, such as gender, height, weight, waistline, was found that most respondents reported income, etc. Enabling factors consist of environment “sometime” consumption of five-food groups, related to food consumption, such as sources of food eating three meals a day, and eating breakfast every and food safety. Reinforcing factors are the support day. However, the respondents reported “rare” things that influenced food consumption, such as family members, friends, and advertisements. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 145 consumption of eating on time every meal. The animal oil, and eating a large amount of each meal, students reported “sometimes” consumption of respectively. Table 1 showed food consumption vegetables, fruits, sweets, salty, spicy, and water 6-8 knowledge or predisposing faction related to food glasses a day regarding a kind of food. The consumption behavior of the first-year students of participants also answered “rare” consumption of the College of Allied Health Sciences, Suan coffee and tea. Sunandha Rajabhat University. The research found that the first-year undergraduate student at the Due to food seasoning, most respondents College of Allied Health Sciences, Suan Sunandha preferred “sometimes” ready meals than home- Rajabhat University had good food consumption cooked meals, fried dishes, add sugar or fish sauce knowledge (80.5%). Regarding enabling factors in to the meal, and grilled dishes.Besides, according to Table 2, the students reported “sometimes” fast food, the students reported “sometimes” consumption of sources of food. consumption of eating a ready-to-eat meal, ready-to- eat breakfast, foods use vegetable oil instead of Table 1: Food Consumption Knowledge or Predisposing Faction Related to Food Consumption Behavior of the First-year Students of the College of Allied Health Sciences, Suan Sunandha Rajabhat University (N=112) Food consumption knowledge Percentage Level of Correct Incorrect/No Knowledge and Predisposing factor responses information understanding Food consumption knowledge 47.8 52.2 Fair 1. Fats are a type of nutrient that gives high energy. 74.3 25.7 Good 2. Obese people are up to 10 times more likely to develop diabetes. 94.7 5.3 Excellent 3. Five food groups provide certain nutritional benefits. 65.5 34.5 Good 4. Gourd soup with minced pork is a low-calorie meal. 30.1 69.9 Fair 5. Fast food dishes are modern. 88.5 11.5 6. Fat foods are leading to obesity. 84.1 15.9 Excellent 7. Obesity is more likely to develop several serious health Excellent problems. 70.8 29.2 8. Custom dishes are high-fat food. Good Table 2: Enabling Factors Related to Food Consumption Behavior of the First-year Students of the College of Allied Health Sciences, Suan Sunandha Rajabhat University (N=112) Enabling factors related to food Percentage NM consumption behavior Always Sometimes Rarely Never frequency Enabling factors Sources of food 1. You buy food from the restaurants in 54.9 33.6 9.7 1.8 3.42 Always front of the dormitory. 2. You buy raw materials from the 11.5 54.0 26.5 8 2.69 Sometimes department store to cook by yourself. 3. You buy raw materials from the fresh 8 40.7 35.4 15.9 2.41 Rarely market to cook by yourself. Food expenditure 4. Earned income contributes to your 44.2 46 6.2 3.5 3.31 Always decisions on food consumption. Food safety 5. You choose clean and safe foods that 54.9 41.6 2.7 0.9 3.50 Always are not leading to health problems. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 146 Table 3: Reinforcing Factors Related to Food Consumption Behavior of the First-year Students of the College of Allied Health Sciences, Suan Sunandha Rajabhat University (N=112) Reinforcing factors related to food Percentage MN consumption behavior Always Sometimes Rarely Never frequency Reinforcing factors Family members 31.9 45.1 12.4 10.6 2.98 Sometimes 37.2 51.3 9.7 1.8 3.24 Sometimes 1. Dishes you cooked are from your family's 9.7 60.2 28.3 1.8 2.78 Sometimes recipe. 13.3 56.6 24.8 5.3 2.78 Sometimes 2. Family-style meals influence your eating 8.8 48.7 35.4 7.1 2.59 Sometimes Friends 3. Friends influence your eating. Advertisement 4. Dishes you cooked are from online advertisement. 5. You buy foods because of advertisement. Table 4: Predicted Variables Related to Food Consumption Behavior of the First-year Students of the College of Allied Health Sciences, Suan Sunandha Rajabhat University (N=112) Predicted variables MN SD Coefficients t P Ob Predisposing factors Enabling factors 2.62 .33 -.106 -.09 -1.15 0.25 Reinforcing factors 3.07 .39 .267 .21 2.90 0.00 2.87 .48 .113 .14 1.115 0.10 By list, according to sources of food, the research was .21, and its O equated to .267. T value in the found that the students reported “always” buy food table also showed that only enabling factors can from the restaurants in front of the dormitory while predict the dependent variables. they “sometimes” buy raw materials from the department store to cook by themselves. The 6. DISCUSSION students also “rarely” buy raw materials from the The first-year students of the College of Allied fresh market to cook by themselves. Following the Health Sciences Suan Sunandha Rajabhat food expenditure, the students reported income University had many factors associated with food “always” contributed to their food consumption consumption behavior that differed from decision. Besides, for food safety, the students predisposing, enabling, and reinforcing factors. reported “always” choose clean and safe foods that do not lead to health problems. Hypothesis 1 expected that the first-year food consumption level of the College of Allied Health Regarding Table 3, the students reported Sciences Suan Sunandha Rajabhat University might “sometimes” consumption by considering the relate to food consumption behavior. The students following reinforcing factors. By list of question, reported they “sometimes” consumed five-food due to family members, the respondents reported groups, ate three meals a day, ate breakfast every family-style meals “sometimes” influenced their day while they “rarely” ate on time every meal. eating, while the dishes they also cooked These results were consistent with Sripaoraya “sometimes” were from their family’s recipe. The Penpong (2016) and Klamsuwa (2011), who proved results of friends tended to be the same; the that adolescents' food consumption behavior was respondents reported their friends “sometimes” related to the amount of meal, and the consumers influenced their eating. In addition to the always chose foods as the result of five nutrients. advertisement, the respondents reported the dishes These results enhanced hypothesis 1. they cooked “sometimes” were from an online advertisement, and they “sometimes” bought foods Hypothesis 2 expected that the first-year food because of advertisement. Table 4 indicated the consumption level of the College of Allied Health regression coefficient (B) of independent variables Sciences Suan Sunandha Rajabhat University might that related to dependent variables. It was found that have good food consumption knowledge. Food the regression coefficient of the enabling factors International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

consumption knowledge according to the P a g e | 147 predisposing factors might influence food consumption behavior. About the research results, REFERENCES the students had good knowledge, which was consistent with Wanwimon, 2016, The study found Klamsuwan, B., 2011, School-Age Child Health that samples knew good food good health. Besides, it was coincident with Chaingkuntod (2012), Condition and Food and Water Sanitation in described most consumers had average scores for Khon-Khaen Health Promotion6Area. general knowledge of food consumption. The http://203.157.71.148/ hpc7 data/ Res/ResFile- findings supported hypothesis 2. /2560000401.pdf. Coon, D. and Mitterer, J. O., 2013, Introduction to Hypothesis 3 expected that food consumption knowledge might be associate with food Psychology: Gateways to Mind and Behavior. consumption behavior. The research found that (13 rd ed.). New Tech Park: WADSWORTH knowledge positively relates to food consumption CENGAGE Learning. behavior – kind of food, food seasoning, and fast Steenkamp, J. B., 1993, Food Consumption food. There were no positive indicators for the five- food groups. These results were related to Behavior. in E - European Advances in Mekwimon (2012), the relationship between Consumer Research Volume 1, eds. W. Fred knowledge and food consumption behavior as Van Raaij and Gary J. Bamossy, Provo, UT: effectively. It was also coincident with Steenkamp Association for http://uc.thailis.or.th/catalog/. (1993); consumers preferred to choose meals Sripaoraya Penpong, M., 2016, Food Consumption because of hungry than food satisfaction, so that fast Behavior of Students in Suratthani Province. food was more popular. The results strongly Journal of Management Sciences, Vol. 3 (1). supported hypothesis 3. Yimprasert, S., 2017, Food Consumption Behavior of Undergraduate Student Level 1 in Hypothesis 4 expected that predisposing, enabling, Rajamangala University of Technology Isan and reinforcing factors might affect consumption Nakhon Ratchasima. https://www.tcithaijo.org- behavior of first-year students of College of Allied /index.php/Ratchaphruekjournal/article/view/907 Health Sciences Suan Sunandha Rajabhat 83. University. The research results proved that the Sudsuntorn, D., 2003, Factors Affecting Health enabling factors were sometimes related to food Promotion Behavior of Health Professionals at consumption behavior due to different human Phamongkutklao Hospital. http://dric.nrct. behaviors, which was closed to Pensresirikul (2012) go.th/Search/SearchDetail/127007. who disclosed that workers in Bangkok had good Chaingkuntod, S., 2012, Knowledge and Behavior food consumption behavior; they selected the clean on Food Consumption of Pasi Charoen Persons. ready meals. Besides, the enabling factors of buying Nonthaburi. Srinagarind Medical Journal, Vol. raw food from the fresh markets were related to the 27(4): 347-53. first-year students' food consumption behavior at the Sudsaneha, T., 2013, Variables Affecting Food College of Allied Health Sciences Suan Sunandha Consumption Behavior of Metabolic Syndrome Rajabhat University. The findings supported Risk Group A Case Study: Phai-ngam Health hypothesis 4. Promotion Hospital, Khok Sung District, Sa - kaeo Province. Rajamangala University of 7. RECOMMENDATIONS Technology Phra Nakhon. http://newtdc.thailis. 1. The research suggests that the related or.th/docview.aspx?tdcid=40564 organization should educate students on food Mekwimon, W., 2012, Factors Related to Food knowledge, food consumption, and nutrition to better health and development. Consumption Behavior among elderly Samut 2. The research also suggests that future works Songkhram Province. Bangkok: Suan Sunandha should study more on behavior related to food Rajabhat University. consumption choice, employ the depth interview for Pensresirikul, W., 2012, Factors Related to Food the data collection and environment, social culture of food consumption and consider the influence of Consumption Behaviors of Working people in economic factors on food consumption. Bangkok Metropolis. http://tnrr.in.th/?page=re- sult_search&record_id=9939205 Konggit, W., 2005, Fast Food Consumption Behavior of Adolescents in Bangkok. Master’s Thesis. Srinakharinwirot University. http://thesi- s.swu.ac.th/swuthesis/Hea_Ed/Wilaiwon_K.pdf. World Health Organization, 2021, Nutrition. https://www.who.int/health-topics/nutrition. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 148 PREVALENCE AND ASSOCIATED FACTORS OF INSOMNIA AMONG THAI ADOLESCENTS IN SAMUT SONGKHRAM Rasornradee Pakpakorn, Chanokporn Panjinda, Papawee Sookdee and Wannee Promdao College of Allied Health Sciences, Suansunandha Rajabhat University Samut Songkhram, Thailand E-mail: [email protected] ABSTRACT The objectives of this research were to prevalence and factors that related to insomnia of Thai adolescents in Samut Songkhram province.The study subjects were male and female 173 students. Both of them age 10-18 years old. The data of study subjects was collected by general information questionnaire, Insomnia questionnaire, Suanprung stress test and Thai-PSQI questionnaire. The data were analyzed in term of percentage, mean and standard deviation. The results showed that 45.1 percent of male and 54.9 percent of female adolescents in Samut Songkhram province who had insomnia 8.7 percent. Factors related insomnia among Thai adolescents in Samut songkhram included; the personal, health, environmental, and social of adolescents; Height, Domicile, Psychiatric disorder and playing the game before going to bed. Factors mentioned above shows that related insomnia among Thai adolescents in Samut Songkhram were significantly at p-value 0.05. The result of Sleep quality showed that related insomnia among Thai adolescents in Samut Songkhram were significantly at p-value 0.01.The suggestions from these study results include 1. Could be collected data of the bigger sample size and more study the related other factors of adolescents’s insomnia in Samut Songkhram province. 2. Could be study in qualitative research to find out the direct factors of insomnia in adolescents. 3. Could be more study about information and method to decrease insomnia of adolescents in Samut Songkhram province. KEYWORDS: Insomnia, Adolescents, Prevalence, Samut Songkhram 1. INTRODUCTION 2017) causing psychological problems. The study Adolescents is an age that has many changes, both found that 10.7% of adolescents with insomnia have physically and mentally. Moreover, the current Thai a relationship with factors that cause mental health society has many external factors that are related, problems, emotions and negative behaviors (Telzer such as social, economic, cultural, technological, et al., 2013 and Soffer-Dudek et al., 2011). And family, friend community factors all these factors all expressed as a violent behavior with the use of affect sleep in adolescents. Which this insomnia drugs and alcohol (Roane BM and Taylor, 2008) on may all be ignored but the fact that insomnia is more games (Woods and Scott, 2006), the ability to common in children and adolescents, which are control emotions and low self-management. These studied in many countries, such as China, Japan, behaviors are very important issues. France and the United States, etc. The prevalence of insomnia is as high as 73% in France (Leger et al., In Thailand, studies on the prevalence and 2000) and 17% in the United States (Robert et al., factors associated with insomnia in adolescents still 2002). In China study (Liu et al., 2000), it was not many. The researcher therefore saw the reported that 16.9% of Chinese adolescents had importance of insomnia in adolescents. Therefore, insomnia symptoms. And in Japan study (Kaneita et interested in studying the prevalence and factors al., 2006), it was reported that 23.5% of Japanese associated with insomnia in adolescents to provide adolescents. Thongmuang and Suwannahong, basic information for caring for adolescents and (2015) study, it was reported that 19.4% of students those involved are aware of this condition in order in Thai Traditional medicine program was stress and to understand the problems that occur with insomnia. From the above figures, it is seen that adolescents. Including finding ways to prevent, insomnia is a problem in many countries. When adjust, solve problems and maintain properly so that adolescents have insomnia for a long time, it affects Thai adolescents grow with good mental health in the development of mental health (Raniti et al., the future. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

2. METHODS P a g e | 149 A cross-sectional descriptive study was carried out in Samut Songkhram Province located in the middle 3. STATISTIC PROCEDURES parts of Thailand. In the current study, we selected The researcher used the collected data to verify the one school three districts representative of the integrity. By using general data to analyze with province. The participants were the target descriptive statistics to describe general population of this study. A total of 173 randomly characteristics Personal factors and mental health selected adolescents were considered candidates for factors such as percentage, mean, standard deviation this study. and to find the relationship of factors related to insomnia by using Chi-square test statistic, with 2. MEASUREMENTS statistical significance at the level of less than 0.05. Data were collected by using self-report questionnaire, Insomnia questionnaire created by 4. RESULTS Napakkawat Buathong, number 8, which has a Prevalence of insomnia and sleep quality are reliability of Cronbach’s alpha more than 0.7, presented in Table 2. The study found that Suanprung stress test-20 created by Suwat adolescents in Samut Songkhram Province With Mahatniranakun, Wanida Poompaisachai, insomnia 8.7% and no insomnia 91.3% and have Phimmasta Panya, number ) 20SPST-(20, which has poor sleep quality Accounted for 64.7 % and had a reliability of Cronbach’s alpha more than 0.7, good sleep quality Accounted for 35.3 % . Factors Insomnia assessment is a translated and adapted related to insomnia among adolescents in Samut from The Pittsburgh Sleep Quality Index (PSQI) Songkhram province. When tested with Chi-square translated into Thai by Tawan. and Fisher's exact statistics, personal factors, health, Chiramongkhumpitak and Waran Chaiyawat gained environment and society of adolescents in Samut confidence by using the Cronbach’s alpha Songkhram province were high. The game before coefficients of 0.83. going to bed is related to the insomnia of adolescents in Samut Songkhram province. With statistical significance at the level of 0.05. Table 1: Sociodemographic data of the study population (n=173) Variable n % Sex 95 54.9 Female Height 20 12.2 71 43.3 <150 73 44.5 150-160 149 86.1 >160 24 13.9 Age (mean= 14.38, SD= 2.25: min= 10 max= 18) 37 21.4 Domicile 53 30.6 83 48.0 Samut songkhram province 50 28.9 Others 123 71.1 Educational level 79 45.7 Primary 94 54.3 Lower secondary Hight school Substance abuse Yes No Activities before going to bed 30 minutes Play game Yes No International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 150 Table 2: Mental health data of the study population (n=173) Variable n% Insomnia 15 8.7 Have insomnia 158 91.3 No insomnia Sleep quality 61 35.3 Good 112 64.7 Poor Table 3: The relationship of factors related to insomnia among adolescents in Samut Songkhram Province Variable Insomnia X2 p-value No Have n=158(%) n=15(%) Sleep quality Good 61(35.3) 0(0) 8.945 0.003** Poor 97(56.1) 15(8.7) Activities before going to bed 30 minutes Play game No 90(52.0) 4(2.3) 5.067 0.024* Yes 68(39.3) 11(6.4) Psychiatric disease No 157(90.8) 13(7.5) 0.02a Didn’t check 1(0.6) 2(1.2) Domicile Samut songkhram 139(80.3) 10(5.8) 0.039a Others 19(11.0) 5(2.9) Height <150 18(11.0) 2(1.2) 6.03 0.05* 150-160 63(38.4) 8(4.9) >160 72(43.9) 1(0.6) ** p<0.01 *p<0.05 a= Fisher’s exact Sleep quality factors from the study, it was found (Leger et al., 2000) and the lowest 17% in the that the quality of adolescents sleep in Samut Songkhram province was significantly related to United States (Robert et al., 2002). In China study insomnia. These are presented in Table 3. (Liu et al., 2000), it was reported that 16.9% of Chinese adolescents had insomnia symptoms. And 5. DISCUSSION in Japan study (Kaneita et al., 2006), it was reported Adolescents is an age that has many changes, both that 23.5% of Japanese adolescents. Epidemiology physically, mentally and mentally. The current Thai in Thailand is up to 25 percent, which may show society, there are many external factors involved that sleep problems are diagnosed with fewer such as social, economic, cultural, technology, problems than reality (Mindell and Owens, 2015). family, community, friends. All these factors all In Thailand, there are still a small number of people affect the sleeping in a lot of adolescents. Which who study insomnia in adolescents, so there should from this research study which is the study of be more education in the larger group. And studying prevalence and factors related to insomnia among factors related to insomnia in adolescents, found that Thai adolescents in Samut Songkhram province, the height was associated with insomnia Because of found that 8.7% which is different from research in this age, physical development will fully grow in China, Japan, France and the United States. The order to function. While the psychological prevalence of insomnia is as high as 73% in France development is not growing as expected the inconsistency of the two aspects of this development International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

therefore has resulted in children of this age having P a g e | 151 to face various problems easily which results in insomnia. 6. CONCLUSIONS This study is a study of specific symptoms. And Factors about domicile adolescents with limited only to adolescents in Samut Songkhram domicile from provinces other than Samut province, so it is not intended to spread the results to Songkhram Is related to insomnia because the general population. After doing research, the immigration affects the development of children researcher found that the population is still small. both physically and mentally, which can affect the And information about anxiety Some events that sleep of adolescents. By studying the research of occurred at that time and family relationships Which Aree Jampakon, Kanchana Tangchonthip, Carey is considered a factor that may be related to Richter and the faculty on the impact of domestic insomnia Which is an interesting way to study migration on health And early childhood research for those interested in the future. development as stated in the interview for qualitative research The main motivation of parents ACKNOWLEDGMENTS in migration is need for work to raise money to The present research was supported by Suan support the family This motivation reflects the lack Sunandha Rajabhat University. The author of opportunities in the domicile, which is a rural gratefully acknowledges the generous assistance of village, which makes migration seem inevitable. school directors and all the participants in the data Migration is therefore considered a good choice for collection in Samut Songkhram province children. Because it helps to have a better life but, migration has resulted in child side effects. And REFERENCES child development Children from households that both parents migrate during the survey period have Leger, D., C Guilleminault, J P Dreyfus, C delayed developmental levels (Jampakon et al., Delahaye, M Paillard, 2000, Prevalence of 2012). Insomnia in A Survey of 12 778 Adults in France. J Sleep Res, Vol. 9(1), 35-42. The perception that they do not know whether or not they have psychiatric disorders May cause Robert, E., Roberts, C. R. R. and Chen, I. G., 2002, adolescents to be more anxious which anxiety can Impact of Insomnia on Future Functioning of cause insomnia which is consistent with Monika Adolescentss. ELSEVIER, Vol. 53(1), 561-569. B.'s research and the faculty (Monika et al., 2016). Thongmuang, P. and Suwannahong, K., 2015, Playing games before going to bed and the Health Behaviours of Undergraduate Students in quality of sleep is associated with the insomnia of Suan Sunandha Rajabhat University. ELSEVIER, adolescents in Samut Songkhram Province Which Vol. 197, 973-976. corresponds to the research of Exelmans and Bulck (2015) and Turel and Morrison (2017) which found Liu, X., Uchiyama, M., Okawa, M. and Kurita, H., that poor sleep quality Is caused by playing games 2000, Prevalence and Correlates of Self- for more than 1 hour, which the amount of playing Reported Sleep Problems Among Chinese games affects sleep efficiency and the use of Adolescentss. Sleep, Vol. 23(1), 27-34. sleeping pills. Kaneita, Y., Osaki, Y., Tanihata, T., Minowa, M., Suggestion Suzuki, K., Wada, K., Kanda, H. and Hayashi, 1. should collect larger sample data and should K., 2006, Insomnia Among Japanese study factors related to insomnia that may be Adolescentss: A Nationwide Representative associated with adolescents in Samut Songkhram Survey. Sleep, Vol. 29(12), 1543-1550. Province. 2. Qualitative studies should also be conducted to Raniti, M. B., Allen, N. B., Schwartz, O., Waloszek, find relevant factors directly. And use smart watch J. M., Byrne, M. L. and Woods, M. J., 2017, for detect sleep quality because the data have Sleep Duration and Sleep Quality: Associations validity and reliability. with Depressive Symptoms Across Adolescents. 3. Should learn more about how to reduce, prevent, Behavioral Sleep Medicine, Vol. 15( 3), 198-215. adjust, solve problems, and maintain properly insomnia among adolescents in Samut Songkhram Telzer, E. H., Fuligni, A. J., Lieberman, M. D. and Province. Galvan, A., 2013, The Effects of Poor Quality Sleep on Brain Function and Risk Taking in Adolescents. NeuroImage, Vol. 71, 275-283. Soffer-Dudek, N, Sadeh, A, Dahl, R. E. and Rosenblat-Stein, S., 2011, Poor Sleep Quality Predicts Deficient Emotion Information Processing Over Time in Early Adolescents. Sleep, Vol. 34(11), 1499-508. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

Roane, B. M. and Taylor, D. J., 2008, Adolescents P a g e | 152 Insomnia as a Risk Factor for Early Adult Depression and Substance Abuse. Sleep, Vol. Jampakon, A., Tangchonthip, K. and Richter, C., 31(10), 1351-1356. 2012, The Impact of Domestic Migration on Health and Early Childhood De Jampakon Woods, H. C. and Scott, H., 2016, Sleepyteens: elopement. Results of Quantitative and Social Media Use In Adolescents i Associated Qualitative Basis Survey. 1-70. with Poor Sleep Quality, Anxiety, Depression and Low Self-Esteem. Journal of Adolescents, Monika, B., Raniti, N. B. A. and Schwartz, O., 2016, Vol. 51, 41-49. Sleep Duration and Sleep Quality: Associations with Depressive Symptoms Across Adolescents. Mindell, J. A., Owens, J. A., 2015, Insomnia. In: Behavioral Sleep Medicine, Vol. 15( 3), 198-215. Elfrank JM, Fischer A and Convery B, editors. A clinical guide to pediatric sleep diagnosis and Exelmans, L. and Bulck, V. D. B. J., 2015, Sleep management of sleep problems. 3rd ed. Quality is Negatively Related to Video Gaming Philadelphia, Wolters Kluwer, 200-207. Volume in Adults. J Sleep Res, Vol. 24(2), 189- 196. Turel, O. R. A. and Morrison, K. M., 2017, A Model Linking Video Gaming, Sleep Quality, Sweet Drinks Consumption and Obesity among Children and Youth. Clin Obes, Vol. 7(4), 191- 198. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 153 Development Analysis of Waste 4.0 Assessment Tool: Transforming Urban MSWM under Industrial Revolution 4.0 focusing Circular Economy Abhishek Kanojia Environment Engineering and Management, School of Engineering and Technology, Asian Institute of Technology, Thailand, E-mail: [email protected] ABSTRACT Rapid urbanization with a high rate of production and consumption is affecting the basic services related to human society today. Waste management is one such crucial sector that is now gaining a serious concern. The Urban dwellers and decision-makers are now working to keep the waste resource in the consumption cycle as much as possible. The major transformation tools proposed by the global community are, change in the economic model (Circular Economy, CE) and technological transition based on the internet (Industry 4.0, I4.0). Among all industrial approaches, MSWM services are kept at the lower priority for any such transformation. To overcome such gaps this research work on three major objectives focusing on such cities which are in transformation state for a smart city. The first research objective is about the development of the technological clusters of MSWM under the shadow of I4.0. The second objective analyzes the measures of the circular economy into the MSWM. The outcome of the research work is the development and analysis of the readiness index of the MSWM sector for transformation into Waste 4.0 focusing CE. This research evaluates the readiness of the urban local bodies (ULB), to upgrade its MSWM by I4.0 technologies and CE. The research work proposes a novel concept of Waste 4.0, which is an amalgam of I4.0 technical intervention followed by CE measures into the MSWM. Many published researches proposed that the I4.0 and CE concepts will be the key enablers of sustainability in various business and service industries. In this research, MSWM is considered as a potential segment of transformation. But to measure this transformation, ULBs lagging the assessment tools. Taking Waste 4.0 as an ultimate target, this research developed an assessment tool to measure the ULB readiness for the transformation of their MSWM system into Waste 4.0. This tool consists of eight determinants followed by different criteria elements to analyze the I4.0 and CE intervention into the MSWM system of ULB. With the help of this tool the ULBs of the cities can simultaneously assess the status of I4.0 and Ce into their MSWM system. The score obtained under the eight determinants are further supported with the assessment criteria associated with them. Based on the score the planning and implementation of MSWM system transformation can be done by the ULBs. KEYWORDS: Industry 4.0, circular economy, municipal solid waste management, urban local body. 1. INTRODUCTION that by 2035 EU member states will ensure that the In the era of rapid urbanization, high population, MSW reaching landfill will be reduced by 10% or and economic growth, municipal solid waste less of the total amount of MSW generated (by management (MSWM) has become a crucial issue weight). To achieve this set target all the member of global concern. The severity of the issue is states developed the roadmap of transforming their confirmed by the fact that in 2016 the world had MSWM under CE prospect. The focus was to set to produced 2 billion tons of municipal solid waste optimize the operations and performance of MSWM (MSW). The World Bank estimated that by 2030, under CE approach. (European Union, 2018). global MSW production amount will be 2.59 billion tons/ annum and by 2050 it will peak up to 3.40 On the other hand, with the growth of digital billion tons/annum (Kaza et al., 2018). To control technologies, intelligent robotics with Internet of the serious problem of MSW, the high economy things (IoT) emerge as a modern tool for countries of European Union (EU) set their focus on development for various industrial and service circular economy (CE). The EU circular economy sectors. To achieve the resource efficiency in waste package 2018 defines the required recycling rates management EU, decide to follow CE concept. The and landfilling rates for municipal waste. It stated novel I4.0 technologies are used to establish such International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

transformation. In this context, an innovative K- P a g e | 154 project: Recycling and Recovery of Waste 4.0 – “ReWaste4.0” was developed at Montaniniversitaet towards I4.0 technologies and CE perspective Leoben, Austria. For the first time in developed simultaneously in MSWM. (Anbumozhi and economic countries, I4.0 approaches were Kimura, 2018). elaborated for urban waste management sector to proceed towards CE. Under this project, digital Studies say that the I4.0 technologies have readiness of the waste management industry was potential to stimulate the CE interventions in discussed in detail for the first time. MSWM of developing countries. Both I4.0 and CE are the ‘leap-frog’ opportunity for MSWM system. ReWaste4.0 investigate the scope of digitization (ISWA, 2017). The global survey conducted by the and the use of robotic technologies in different ISWA on, “Impact of I4.0 in MSWM” highlights aspects of urban waste management system in the fact that only 14% of the participant experts context of CE. 28 members of EU participated in consider themselves as knowledgeable about I4.0 this project. Industry 4.0 has a great potential to opportunities. This shows that there is adequate serve as a tool for EU circular economy target. (Sarc need for Waste 4.0 concept development for the et al., 2019) (Mavropoulos and Nilsen, 2020). CE MSWM system. The key challenge for the and I4.0 were discussed for waste management development of “Waste 4.0 assessment tool” was upgradation in EU and such transformation is also the scarcity of relevant supporting literature. One of the need of developing countries. In developing the key discussion of the study shared under K- countries, the most popular approach to handling Project was that during 2001-2019 there were only MSWM is the ISWM framework. Which is indeed 85 relevant publications with keywords such as to be enhanced under I4.0 and CE transformation. ‘digitalization’, ‘robotics’, ‘smart waste’, ‘smart The urban administration commonly known as factory’, ‘Industry 4.0’, ‘internet of things’, ‘waste urban local body (ULB) is the key stakeholder for management’, and ‘circular economy’. Among the MSWM in the developing countries. The fourth published references many of them are published in industrial revolution Industry 4.0 (I4.0) and the last 3 years (2017-2029). Among the 85 relevant concept of Circular economy (CE) are the most literature references, only 18 publications have peer- recent and promising developments in technology reviewed. This information shared under project and policy level, respectively (Anbumozhi and document of ReWaste4.0, clearly indicates the Kimura, 2018). In developing countries MSWM reason for the lack of supporting literature for Waste system is majorly handled by the urban local bodies 4.0 concept development. (Sarc et al., 2019). (ULB). These bodies plan, execute and analyze the waste generation and its processing options. These In 2017, the International Solid Waste ULBs are using an integrated solid waste Association (ISWA) report reveals that only 14% of management framework (ISWM), which assists the participants consider themselves to be them to transform waste management into resource knowledgeable about I4.0 in waste management. It management. The ISWM approach was proposed by also confirms that SWM experts consider I4.0 to UN-Habitat back in 2010. Now ISWM framework stimulate the CE intervention in MSWM. Both of seeks the leap-frog opportunities to upgrade itself. these concepts bring the ‘Leapfrog Opportunity’ for (Wilson et al., 2015; Modak et al., 2017 and the MSWM sector of developing countries (ISWA, Esmaeilian, B., et al., 2018) 2017). There is a lack of a clear understanding of how to incorporate such changes (ISWA, 2019). To To enhance the waste data management for overcome these drawbacks this research paper efficient ISWM planning, monitoring execution I4.0 develops an assessment tool, Waste 4.0. This tool and CE emerge as a key factor. The ULB of assists ULBs to comprehensively upgrade their developing countries can also achieve high resource MSWM system. The research outcome of the EU efficiency in their MSWM system by incorporating project ReWaste4.0 as mention followed by the the modern trends of I4.0 and CE. Waste 4.0 an ERIA publication on, “Assessing the readiness of innovative waste management approach is Industry 4.0 and the circular economy” is the key conceptualized and discussed in detail in this basis of the selection of I4.0 technologies under research paper. Waste 4.0 is the amalgam of I4.0 Waste 4.0. (Ambumozi et al., 2020 and Sarc et al., technologies with circular economy initiatives for 2019). Figure 1 explains the components of I4.0 and the advancement of MSWM system. Waste 4.0 CE which intervene in the ISWM framework to concept is developed and discussed in this research develop Waste 4.0 system. The Waste 4.0 paper as a tool for assessing the readiness of ULB assessment tool is validated for the MSWM system of Indore and Sagar ULBs of India. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 155 Figure 1: Components of Waste 4.0 Figure 2: Abstract of the research work and its application The Waste 4.0 tool is developed to know the associated with Waste 4.0 and the flow of the following: a). The readiness index of ULB, for I4.0 research development. interventions. b) The readiness index of ULB, for CE interventions. c)The readiness rating of ULB for 2. ASSESSMENT OF THE STATUS OF CE focus in I4.0 readiness. These objectives are developed for ULBs to upgrade their current INDUSTRY 4.0 READINESS conditions of MSWM system. Further, the score Industry 4.0 concept was originally proposed by obtained in various determinants of Waste 4.0 Germany in 2011. It is the transformation of the analysis helps the ULBs to plan specific strategies system based on digital technological platforms. for transformations. As a case study, Waste 4.0 tool Among the various digital technologies which is applied in Indore and Sagar ULBs, to find out the contribute to the establishment of I4.0, development readiness of ULBs towards the intervention of I4.0 of the cyber-physical system plays a key role and CE. Figure 2 provides an abstract of the Waste (Forschungsunion and Acatech, 2013). As it 4.0 assessment. This figure explains the components connects the physical world with the virtual world. Other important technologies are Internet of things International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 156 (IoT), Cloud computing, Radio frequency of the system for, implementing I4.0 (Rajnai, 2018 identification (RFID), Big Data analysis, Artificial and Schumacher et al., 2016). The analysis of such intelligence, Enterprise resource planning (ERP), models suggests that although there is growth in the Mobile-based internet applications, Robotics and research trend on I4.0, there is also a research gap many more. Any organization which transforms its on the use of maturity model while implementing system by internet-based digital technology is I4.0 (Gokalp et al., 2017). The evaluation criteria, known as I4.0 enabler. This create the necessity for dimensions, and items are different for various the need of readiness scale that predicts or analyze models. At the same time there is no standard and the level of the organization for adopting digital well-accepted model yet developed (Akdil et al., technologies. 2018 and Schumacher et al., 2016). The domain of the determinants of most such models is focused on The I4.0 involves the connection and integration assessing the IT readiness only. Therefore, there is a of virtual and physical world through cyber physical need for understanding the key determinants to system and IoT through smart objects, which assess the readiness factor for implementing I4.0 continuously communicate and interact with each from a holistic perspective (Ramanathan, 2018). other (Oberg and Graham, 2016) to meet the Table 1 shows the various I4.0 maturity models predefined objective. Therefore, implementing I4.0 (MM) and their assessment approach. a crucial strategic decision and before taking such a decision, organizations have to assess the readiness Table 1: Industry 4.0 maturity models and their assessment approach Model Name Assessment approach Source The connected enterprise It has five stage processes for I4.0 implementation. Automation, 2014 maturity model (2014) In this model technology assessment have been conducted for four determinants. There are not Lichtbulau et al., 2015 IMPULS-Industrie 4.0 much detail illustrations about item and Readiness (2015) development. Lanza et al., 2016 It has 6 determinants for assessment. It has 18 items Waterhouse, 2016 Empowered and which measure the readiness in five levels. It implementation Strategy for discusses the obstacles and provides Schumacher et al., 2016 Industry 4.0 (2016) recommendations to overcome them. Menon et al., 2016 Industry 4.0/ Digital It is a process model for realization. It is also used Leyh et al., 2016 Operations Self-Assessment for the gap analyses. There are no details about Akdil et al., 2018 (2016) items and development process. It is an online self-assessment in six dimensions. Industry 4.0 readiness and The emphasis is on the digital maturity in four maturity of manufacturing levels in each phase. The application as a consulting enterprises tool required in three of the six dimensions. There Maturity model for are no details about the items and development Industrial Internet process offered. SIMMI 4.0 This model defines nine dimensions. The emphasis was on the extension of existing models and tools Maturity and Readiness through its strong focus on organizational aspects. Model for Industry 4.0 It is a preliminary study of assessing industrial Strategy internet maturity. The model has five maturity stages and three determinants of integration. Propose a model which considers the principles of real-time data management, interoperability, decentralized and service-oriented. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 157 The maturity models for measuring the I4.0 mention by (Lichtblau et al., 2015) proposed 6 dimensions in Table 1 are developed majorly for the production which are, Strategy and organization, Smart and supply chain management industries. These Factory, Smart Operations, Smart Products, Data- models focus only on the technological intervention driven Services, and Employees. These core of Industry 4.0 and it lag to cover CE intervention or dimensions are further subdivided. This model also its assessment. These models also do not focus on provides the rating level for assessing the status of the MSWM sector of urban administration. But at I4R. These levels are L0: Outsider, L1: Beginner, the same time these maturity models provide the L2: Intermediate, L3 Experienced, L4: Expert, L5: initial understanding of the assessment analysis of Top Performer. While insightful to experienced I4.0. Organizations should design their strategy practitioners, this is not an easy approach for a firm based on the anticipated changes in various to use it as a self-assessment tool. Similarly, the relationships due to the implementation of I4.0. WMG- The University of Warwick, proposed an These are the following six key ingredients which I4R assessment tool (2017) which have six are required by the organization for assessing their dimensions. These are as follows: a)Strategy and readiness for I4.0 (Sony and Naik, 2019): a)Top Organization b)Manufacturing and Operations management involvement b)Employee adaptability c)Supply Chain d)Products and Services e)Business with I4.0. c)Smart product and Services. d)Extent of Model f)Legal Considerations. It has the four rating digitization in supply chain. e)Level of digitization levels associated with these six determinants for of the organization. f)The readiness of assessing the state of I4R. These levels are L1: organizational strategy. One of the earliest studies Beginner, L2: Intermediate, L3: Experienced, L4: on Industry 4.0 readiness (I4R) is discussed Berger, Expert. This can be used as a self-assessment tool 2014. This study examined the I4R in Europe and by the firms, the limitation is that the Manufacturing highlighted challenges faced not just at the firm and Operation determinant is specifically for the level but within the business eco-system. Based on manufacturing technology perspective. Few aspects the analysis the report suggested that the different such as quality and energy consumption are not European nations could be classified as included in this tool. The dimension of human Frontrunners, Potentialists, Traditionalists, and resources is also not mentioned well in this tool. The Hesitators, concerning their transformation under Yanez maturity Index framework (Yanez, 2017) has I4.0. The IMPULS-Industrie 4.0 Readiness study eight determinants as mention in Table 2. Table 2: Core determinants of the selected I4.0 readiness assessment frameworks International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 158 The missing determinants of this framework used in Waste 4.0. The eight determinants (Table 4) ‘strategy and organization’ and ‘information are taken from the segments proposed for technology systems’. The (Akdil et al., 2018), sustainable MSWM mentioned in GWMO discuss ten core determinants to measure the I4R in analytical framework. These also align with the the firm. This model uses four stages to mention the determinants discussed in Table 2. Each of these maturity level of the firm, these four stages are determinants for Waste 4.0 assessment is the key ‘Absence’, ‘Existence’, ‘Survival’, and ‘Maturity’ to criteria for developing a sustainable waste determine the maturity level. Table 2 discussed the management system as per GWMO. selected four frameworks with their core determinants used for measuring the I4R in the firm 3. ASSESSING THE EXTENT OF CIRCULAR or organization. Based on the analysis of the various frameworks mention in Table 1 and Table 2 it is ECONOMY (CE) FOCUS IN INDUSTRY 4.0 concluded that these assessment determinants are focusing on manufacturing industries. There is lack READINESS (I4R) of such determinants which analyze the various Industry 4.0 has the potential to stimulate the CE segments of MSWM system. To develop the Waste and recycling markets. The concept of CE can be 4.0 determinants which align with MSWM sector, effectively employed in different organizations and the research follows the key elements of the systems with the help of I4.0. In MSWM system the analytical tool used in Global waste management measure of CE existence will reduce the quantity of outlook (GWMO)(United Nations Environment the MSW which need to be treated at the end. In Programme, 2016). Under the GWMO approach, addition to this it has the potential to reduce the there are three major segments that are considered overall financial expense of ULBs. The data for the development and analysis of a sustainable produced by the I4.0 technologies will assist the waste management system in the city. These ULBs to establish the CE measure scientifically. segments are further comprising of the sub- segments as shown in the Table 3. The I4.0 technologies will assist the CE establishment in MSWM system of ULBs. It is Considering the ULBs as a firm that manages proposed that it would be beneficial if a roadmap MSWM system of the city and seeking to perform could be developed to explicitly incorporates CE readiness assessment for I4.0 and CE interventions principles in I4.0 approaches (De Jesus et al., 2018). into MSWM system. The ULB is analyzed under In the same context it is said that it would be useful the proposed assessment framework of Waste 4.0. to examine how the six business actions proposed The determinants of Waste 4.0 are selected on by the Ellan MacArthur Foundation, which is behalf of Table 2 and Table 3. The key determinants referred as ReSOLVE Framework (Ellan MacArthur consider for Waste 4.0 assessment are shared in Foundation, 2015). The 6 principles of ReSolve Table 4. This table also mention the specific framework are guided by the 18 specific evaluating authority/ section of ULBs by whom the components advocating the CE establishment. The survey for Waste 4.0 assessment should be taken. components associated with each of the 6 principles The 3rd key information shared in this Table 4 is are as follows: Regenerate- a) Shift to renewable about the significance of each of the determinants energy and materials. b) Reclaim, retain, and restore health of ecosystems. c) return recovered biological resources to the biosphere. Table 3: Major and sub segments for sustainable waste management system Major segments for Physical elements Stakeholders Strategic aspects sustainable waste Waste generation and Municipalities/ Urban local Political management storage body Collection Regional and National Health Sub-segments government Transportation Formal and informal service Institutional provider. Recycling Local organizations Social Recovery material International agencies Financial Treatment process Environmental Disposal Technical International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 159 Table 4: Determinants for the proposed framework of assessment of Waste4.0 Sr. Determinants Evaluating Significance No. authority/section of ULBs This determinant assists the investigator to find out what D1 Strategy and Decision-making measures are taken at the decision-making level of ULB for organization authority implementing I4.0 and CE. These determinants will also measure the knowledge of decision-making authority for an improved MSWM system. This determinant focus on analyzing the various facility D2 MSWM Plants and MRF center and plants/units under MSWM system of the ULB. The status of equipment officials who handle equipment and machines employed for MSWM, its it maintenance, and operations are also analyzed on behalf of I4.0 and CE. IT system and data management peruse crucial role in the D3 IT system and data Back-office IT efficient working of MSWM operations. This lead the management support system for Waste 4.0 assessment to analyze the readiness of IT system MSWM and data management in terms of digital infrastructure available in the ULB. This determinant excess the stakeholders who execute the D4 Human Resources Both on-field and off- MSWM system both at front and back office of ULB. The field human resource awareness towards I4.0 and CE among the on-field and off- field officials are measured with the help of this determinant. This determinant is selected for Waste 4.0 assessment to find out the ULB administration approach of handling D5 Resource Material Staff processing secondary resource material in context of I4.0 and CE. The Recovery secondary material waste material which is produced daily comprises from MSW secondary materials both organic and inorganic form. The status of ULB applying I4.0 and CE perspectives for such material is the key finding of this determinant. D6 Managing Staff handling The ULB consumes huge energy for efficient operation and Operations: Energy material recovery maintenance of its MSWM system. This led the research to consumption facility center (MRF) incorporate this determinant which eventually find out the management and WtE (Waste to ULB status in context of I4.0 and CE. Energy) unit Both determinants (D7 and D8) are representing the most Managing Collection and crucial and expensive operation of the MSWM system. D7 Operations: Waste transportation These are included for Waste 4.0 assessment, to understand collection officials how ULB use I4.0 and CE for the efficient operations of collection and transportation of MSW in the city. Share- a) Share assets b) Reuse/secondhand c) This finally assists to develop 13 assessment criteria Prolong life through maintenance, design for that collectively determine the extent of CE focus in durability, upgradability. Optimize- a) Increase MSWM system of the ULB. The assessment performance/efficiency of product. b) Remove framework Waste 4.0 is specifically developed to waste in production and supply chain. c) Leverage measure the readiness of the ULB of developing big data, automation, remote sensing, and steering. countries to adopt I4.0 technologies and CE Loop- a) Remanufacture products or components. b) prospect into MSWM system only. This leads to Recycle materials. c) Digest anaerobically. d) support the fact that all the determinants and Extract biochemicals from organic waste. assessment criteria selected under Waste 4.0 are Virtualize- a) Dematerialize directly. b) associated with MSW. At the same time emphasis Dematerialize indirectly. Exchange- a) Replace old was given to keep all those factors that play key with advanced non-renewable materials. b) Apply crucial role under MSWM. The determinants such new technologies. c) Choose new product/service. as D1 (Strategy and organization), D3 (IT system and data management), and D4 (Human resource) The above mentioned 18 components associated are not directly related to the common waste with 6 principles of ReSolve framework are management hierarchy still they play crucial role in associated with the 8 determinants of proposed the efficient handling of MSW. concept of Waste 4.0 assessment. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

These principles are incorporated into actioning of P a g e | 160 the eight determinants in the proposed Waste 4.0 framework. It is the framework for assessing the to the investigator. Ones all the assessment criteria extent of the CE focus in I4R. The eight of the determinants are covered along with the determinants are the same as in the I4R framework evidence, the final score of the assessment should be (Table 4) to ensure compatibility between the two filled in a separate sheet. The quantity of the frameworks. Each of these determinants consists of participants for the interview and assessment 13 assessment criteria which collectively, will feedback may vary as per the size of the ULB of the determine the extent of CE focus concerning each city and the number of officials engage in the determinant. The CE based elements for each of MSWM system. The decision of selecting the these determinants are synthesized from (DeJesus et participant should be made by the chief investigator. al., 2018, Jovanovic et al., 2017, Lieder and Rashid, The concern during this assessment should be that 2016, Malinauskaite et al., 2017 and SITRA, 2016). the managers and officials who participated in the assessment should be biased, as it will reflect in the 4. PROCEDURE OF THE ASSESSMENT OF rating they provided. If there exist drastic variation among the rating provided by different official for a I4R AND CE FOCUS I4R, FOR MSWM particular determinant, then the chief investigator must take the concern and find out the reason for SYSTEM OF ULBS USING THE PROPOSED such variation. FRAMEWORKS: Step 3: Assessing the CE Focus in Industry 4.0 This research developed such framework which aims to find out the I4R and CE focus under I4R of Readiness ULBs for MSWM system. Both of the framework In this step, the chief investigator will collect the together form the Waste 4.0 assessment tool for the inputs from the same officials who participated for ULB. To execute this framework tool, firstly ULB step 2. The same procedure must be followed as in administration needs to appoint ‘chief investigator’ step 2, but this time the focus of the investigation who have access to all the direct and indirect must be on the CE intervention in the MSWM contacts of the city MSWM system. The assessment system of the city. The rating once collected must be has the following steps: properly organized in a separate sheet. This will help the investigator to keep a check on the data Step 1: Obtaining the Background Information of collection for every element of the respective determinants. the Selected ULB and its MSWM System. Before performing the assessment on the MSWM, it Step 4: Interpretation of the Findings and Data is necessary to first get the general information in the form of a discussion. This should be done with Analysis the administrative authorities of the ULB of the city, There are total of 32 assessment criteria associated particularly those who handle the MSWM section. with the eight determinants of the I4R framework, This initial exercise will assist the chief investigator each of the assessment criteria is associated with 5 to plan the meetings and visits at various sections of levels for rating. Each level has been assigned with MSWM of the selected city. The formal discussion the score. So, the maximum and minimum score should comprise of the introductory information achievable by 32 elements are 128 and 0. Since regarding the current procedure of MSWM, the there are 32 elements used for assessing I4R in hierarchy of the concern officials, what are the MSWM system of the city, the score obtained on general facts and figure for the urban waste behalf of these elements are assigned to the status collection and disposal, administrative hierarchy of mention in Table 5b. Similarly, for the CE focus the ULB, future strategies for MSWM, problem and I4R readiness rating framework, there are a total of challenges of the system. 13 assessment criteria associated with 8 determinants. Each of them is assigned with 5 Step 2: Assessing Industry 4.0 Readiness levels. The maximum and minimum scores for this This step aims for the rating of assessment criteria framework analysis are 52 and 0. The interpretation associated with the proposed determinants, for both of the obtained score is assigned with the status as I4.0 and CE readiness. For this purpose, the mentioned in table 6. This status will depict the CE information collected at step one will guide the focus I4R rating in MSWM of the ULB. investigator to fix the meeting with the appropriate officials and concerns of ULB. The official must be Step 5: Development of the Readiness Index and requested to provide with the rationale of their rating. This may be in the form of the field visits, Managerial Implications audio-video recordings, or authentic data provided The calculation of the readiness index for I4.0 and CE focus in I4.0 in MSWM system and the position International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 161 of the ULBs concerning Waste 4.0 has been Automation, R., 2014, The Connected Enterprise mentioned in the case study below. The managerial implications based on the result of Waste 4.0 Maturity Model. Rockwell Automation, 1-12. assessment are further developed considering the following suggestions: Berger, R., 2014, INDUSTRY 4.0–The New  Use both (of I4.0 and CE focus I4.0) the Industrial Revolution| Maschinenbau| assessment outcome to reach the consensus on immediate feasible actions that can be Engineered Products/High Tech| taken. Branchenexpertise| Expertise| Roland Berger.  Use the assessment of Waste 4.0 results to develop the long term and short term vision De Jesus, A., Antunes, P., Santos, R. and of ULBs for their MSWM system. Mendonça, S., 2018, Eco-Innovation in the  Identify the key sectors of the MSWM for transformation with the help of the score Transition to a Circular Economy: An Analytical obtained by ULBs for each determinant. Develop strategy for both prospective, waste Literature Review. Journal of Cleaner products as well as waste handlers. Production, Vol. 172, 2999-3018.  Create a steering committee to review the implementations of the actions, initiatives for Ellen MacArthur Foundation, 2015, Delivering the Waste 4.0 Circular Economy-A Toolkit for  Prioritize the infrastructure required for Policymakers.[online]https://www.ellenmacarth establishing Waste 4.0, as per the score urfoundation.org/assets/downloads/publications/ obtained for different determinants. EllenMacArthurFoundation_PolicymakerToolkit .pdf (Accessed 10 May 2020)  Strive for the growth of Waste 4.0 through Esmaeilian, B., Wang, B., Lewis, K., Duarte, F. continuous radical improvements supported Ratti, C. and Behdad, S., 2018, The future of by the policies and public participation. waste management in smart and sustainable cities: A review and concept paper, Waste 5. CONCLUSION Management, Volume 81, 2018, Pages 177-195, Based on the score obtained for I4.0 and CE ISSN 0956-053X, https://doi.org/10.1016/j.wa- readiness under Waste 4.0 the cities can easily find sman.2018.09.047. out the status of technological intervention and CE European Union, 2018, Directive (EU) 2018/850 of measures in their MSWM system. Each of the the European Parliament and of the Council of determinant value the ULBs get the base for 30 May 2018 amending Directive 1999/31/EC planning and design such strategies which can on the landfill of waste (14.6.2018). improve their MSWM system from business as http://data.europa.eu/eli/dir/2018/850/oj usual model to leapfrog position. Establishing a (accessed on 20 November 2020) sustainable MSWM system in a holistic manner Forschungsunion and Acatech, 2013, under the reign of Industry 4.0 and Circular economy. Recommendations for Implementing the REFERENCES Strategic Initiative INDUSRTIE 4.0. Final report of the Industrie 4.0 Working Group. [online] Akdil, K. Y., Ustundag, A. and Cevikcan, E., 2018, https://en.acatech.de/publication/recommendatio Maturity and Readiness Model for Industry 4.0 ns-for-implementing-the-strategic-initiative-ind Strategy. In Industry 4.0: Managing the Digital ustrie-4-0-final-report-of-the-industrie-4-0- Transformation, 61-94. DOI:10.1007/978-3-319- working-group/ (Accessed on 25 May 2020). 57870-5_4. Gökalp, E., Şener, U. and Eren, P. E., 2017, Development of an Assessment Model for Anbumozhi, V., Ramanathan, K. and Wyes, H, Industry 4.0: Industry 4.0-MM. International 2020, Assessing the readiness for Industry 4.0 and the Circular Economy. ERIA. ISBN: 978- Conference on Software Process Improvement 6-025460-28-9. and Capability Determination, 1-14. ISWA, 2017, The Impact of the 4th Industrial Anbumozhi, V. and Kimura, F., 2018, Industry 4.0: Empowering ASEAN for the Circular Economy. Revolution on the Waste Management Sector. Study Report Presented at the ISWA World Economic Research Institute for ASEAN and Congress, September 2017,Baltimore. [online] East Asia, 1-140. http:// www.iswa.org/index.php?id=1549 (Acce- ssed 20 April 2020). ISWA, 2019, How Industry 4.0 Transform the Waste Sector. Report Presented at the ISWA World Congress, October 2019,Bilbao, Spain. [online] https://www.iswa.org/media /publicatio- n/knowledge-base/ (Accessed 25 April 2020). International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 162 Jovanović, B., Filipović, J. and Bakić, V., 2017, http://www.rrcap.ait.ac.th/Publications/Asia%20 Energy Management System Implementation in Waste%20Management%20Outlook.pdf. Serbian Manufacturing–Plan-Do-Check-Act Cycle Approach. Journal of Cleaner Öberg, C. and Graham, G., 2016, How Smart Cities Production, Vol. 162, 1144-1156 will Change Supply Chain Management: a Kaza, S., Yao, L., Bhada-Tata, P. and Van Woerden, F., 2018, What a Waste 2.0: a Global Technical Viewpoint. Production Planning & Snapshot of Solid Waste Management to 2050. Control, Vol. 27(6), 529-538. The World Bank. DOI:10.1596/978-1-4648- Rajnai, Z. and Kocsis, I., 2018, Assessing Industry 1329-0. 4.0 Readiness of Enterprises. IEEE 16th World Lanza, G., Nyhuis, P., Ansari, S. M, Kuprat, T. and Symposium on Applied Machine Intelligence and Liebrecht, C., 2016, Empowerment and Informatics (SAMI), 000225-000230. Implementation Strategies for Industry 4.0. ZWF Ramanathan, K., 2018, Industry 4.0 Readiness with Journal for Economic Factory Operation, Vol. a Circular Economy Focus: An Integrated 111(1-2), 76-79. Assessment Approach, Second Technical Leyh, C., Bley, K., Schäffer, T. and Forstenhäusler, Workshop, Measuring Industry 4.0 Readiness S., 2016, SIMMI 4.0- A Maturity Model for (I4R) for Circular Economy. Bangkok, Thailand. Sarc, R., Curtis, A., Kandlbauer, L., Khodier, K., Classifying the Enterprise-Wide it and Software Lorber, K., & Pomberger, R. (2019). Landscape Focusing on Industry 4.0. In 2016 Digitalisation and intelligent robotics in value Federated Conference on Computer Science and chain of circular economy-oriented waste Information Systems (fedcsis). IEEE., 1297- management – A review. Waste 1302. Management, 95, 476-492. https://doi.org/10.10- Lichtblau, K., Stich, V., Bertenrath, R., Blum, M., 16/j.wasman.2019.06.035 Bleider, M., Millack, A., Schmitt, K., Schmitz, E. and Schröter, M., 2015, IMPULS-Industrie Schumacher, A., Erol, S. and Sihn, W., 2016, A 4.0-Readiness. Impuls-Stiftung des VDMA, Maturity Model for Assessing Industry 4.0 Aachen-Köln. Lieder, M. and Rashid, A., 2016, Towards Circular Readiness and Maturity Of Manufacturing Economy Implementation: a Comprehensive Review in Context of Manufacturing Enterprises. Procedia Cirp, Vol. 52(1), 161-166. Industry. Journal of Cleaner Production, Vol. SITRA, 2016, Leading the Cycle-Finnish Road Map 115, 36-51. Malinauskaite, J., Jouhara, H., Czajczyńska, D., to a Circular Economy 2016-2025. [online] Stanchev, P., Katsou, E., Rostkowski, P., https://media.sitra.fi/2017/02/28142644/Selvityk Thorne, R. J., Colon, J., Ponsá, S., Al-Mansour, sia121.pdf (Accessed on 15 May 2020), ISBN F. and Anguilano, L., 2017, Municipal Solid 978-951-563-978-3 (PDF). Waste Management and Waste-To-Energy in the Sony, M. and Naik, S., 2019, Key Ingredients for Context of a Circular Economy and Energy Evaluating Industry 4.0 Readiness for Recycling in Europe. Energy, Vol. 141, 2013- 2044. Organizations: a Literature Mavropoulos, A. and Nilsen, A. W., 2020, Industry Review. Benchmarking: An International Journal. DOI:10.1108/BIJ-09-2018-0284. United Nations Environment Programme, 2016, Global waste management outlook. United Nations. ISBN: 978-92-807-3479-9. https://www.unep.org/resources/report/global- waste management-outlook 4.0 and Circular Economy: Towards a Wasteless Waterhouse Coopers, P., 2016, Industry 4.0- Future or a Wasteful Planet?. John Wiley & Sons. 1-448. Enabling Digital Operations Self-Assessment. Menon, K., Kärkkäinen, H. and Lasrado, L.A., 2016, Towards a Maturity Modeling Approach Wilson, D. C., Rodic, L., Modak, P., Soos, R., for the Implementation of Industrial Internet. Carpintero, A., Velis, K., Iyer, M. and Simonett, O., 2015, Global Waste Management Outlook. UNEP. Towards a Maturity Modeling Approach for the WMG-The University of Warwick, 2017, An Implementation of Industrial Internet, Vol. 20, Industry 4 Readiness Assessment Tool. [online]https://warwick.ac.uk/fac/sci/wmg/resear 1-38. ch/scip/reports/final_version_of_i4_report_for_u se_on_websites.pdf (Accessed 15 May 2020) Modak, P., Pariatamby, A., Seadon, J., Bhada-Tata, Yanez, F., 2017, The Goal is Industry P. and International Solid Waste Association, 2017, Asia Waste Management Outlook. United Nations Environment Programme, 4.0:Technologies and Trends of the Fourth Industrial Revolution, San Bernardino,CA. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 163 AIR QUALITY AND NOISE IMPACT ASSESSMENT OF DUMPSITE MINNG FOR OCCUPATIONAL SAFETY OF WORKERS Pawan Kumar Srikanth1,2 1Department of Energy, Environment and Climate Change, Asian Institute of Technology, Thailand 2Centre for Environmental Studies, Anna University, Chennai, India ABSTRACT In India, disposal of solid wastes in low-lying areas is a common practice. Over the years of disposal, the practice has resulted in vast areas of heaps of wastes or dumpsites creating many environmental problems. To remediate or rehabilitate the old dumpsites, Dumpsite Biomining process is carried out to separate recyclables, refuse and inerts from the partially or fully decomposed solid waste. This process has resulted in release of particulate dust and harmful gases into the atmosphere causing health hazards to the workmen at dumpsite processing plant and surrounding population. The maximum PM2.5 concentration (72 µg/m3) was observed near the panel room and minimum PM2.5 concentration (41 µg/m3) was observed at Disc Screen Separator. The equivalent noise level calculated near the Trommel 2 is 92 dB which exceeds the maximum permissible noise level of 85 dB (Continuous or intermittent noise) from a Factory or Workshop, National Environment (Noise Standards and Control) Regulations, 2013. KEYWORDS: Dumpsite Mining, Municipal Solid Waste, Ambient Air Monitoring, Particulate Matter, Noise, Occupational safety 1. INTRODUCTION reclamation of old dump site by in-situ composting Municipal solid waste disposal is one of the major by formation of windrows at the top of the dumpsite threats to the environment.In India, out of the 2120 and their systematic turning and sieving. The dumpsites reported by the SPCBs/PCCs, only 21 garbage at the dumpsite may be loosened by a dumpsites have been converted into Sanitary tractor-cultivator in 15 cm layers and bulky waste Landfill Sites (CPCB, 2017). Mavropoulos et al., removed manually. The rotted garbage (quite (2014) mapped and profiled the 50 biggest smelly) may be sprayed with a bio-sanitiser plus dumpsites in the world. This set includes three composting bioculture and formed into windrows Indian dumpsites in Mumbai (Deonar, 132 ha), which could be turned periodically till complete Bengaluru (Mandur, 35 ha) and Delhi (Ghazipur, 30 stabilization of the waste. The Dumpsite Biomining ha). The report asserted that a population living process has cleared 75000 cubic tons of garbage and within a distance of 10 km from the dumpsites is at reclaimed 7.5 acres of land at Kumbakonam and potential risk of health hazard due to the exposure to 35000 tonnes of waste spread over 4.5 acres at either leachate (contaminated liquids leaching into Sembakkam (MoUD, India 2017). In future, the the groundwater) or dust from dumpsites. Leachate biomining process is set to take place at Pallavaram, can contain heavy metals, VOCs or hazardous Tambaram, Chidambaram, Poonamallee, Avadi, organic compounds. These pollutants are carried Erode, Trichy and around 90 dumpsite locations due into aquifers or surface waters. Dust from dumpsites to successful completed operations. The assessment may contain metals and human pathogens that come of air and noise quality during the biomining into contact with this pollution through operation has not been thoroughly done and contaminated groundwater and soil, or direct contact reported so far. Further, the effects of the air and with the waste site. noise pollution on the people working at site and the surrounding population have not been considered. Marvropoulos and Newman (2015) discussed the Zala and Jure (2011) provided a very good account possible pollutants in dumpsites which include of various sources of Particulate matter and the Particulate Matter, Methane and Carbon Dioxide, effects of PM on human health. Exposure to PM has SOx, NOx, CO, Ozone, Volatile Organic been linked to a number of different health Compounds), Persistent organic pollutants (e.g., dioxins) and Heavy metals. The process involves International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

outcomes, including lung inflammatory reactions, P a g e | 164 reduction in lung function, adverse effects on the cardiovascular system, visit to the hospital recommend mitigation measures to reduce the emergency department, admission to hospital, and adverse impacts on human health and environment. death. The study done by Kerns et al., (2018) estimated the prevalence of occupational noise 2. MATERIAL AND METHODS exposure, hearing difficulty and cardiovascular conditions within US industries and occupations, 2.1 Study Area and Process Description and to examine any associations of these outcomes The old dumpsite where the biomining process took with occupational noise exposure. Since the scalable place was in the neighborhood in Chennai potential of the process is high and biomining Metropolitan city of the state of Tamil Nadu, India. operations are set to take place in various part of the The Waste processing at site was contracted to country, the contribution of the air and noise excavate and process 80000 tons of waste at a rate pollution due to biomining process must be assessed of 200 tons per day. The wastes were turned, heaped and proper mitigation measures has to be given to and covered with tarpaulins the day before it was protect human health and the environment. This processed. The excavators fed the waste to the study addressed ambient air quality parameters trommel section through waste conveyor belts. The namely Particulate matter pollutants (PM2.5 and Trommel 1 separated the +20 mm and -20 mm PM10) and gaseous pollutants which included particles and larger particles like wood, stones were Hydrocarbons, Carbon Dioxide, Carbon Monoxide, manually removed. The -20 mm particles were Oxides of Nitrogen and Sulphur during the further separated by Vibro Screen. It sieved the biomining process at an old dumpsite near Chennai. particle into +10mm and -10mm particles which The ambient noise monitoring was also done and were used as filler material. The +20 mm particles compliance with CPCB standards was checked. were further processed by Disc Screen Separator Since the scalable potential of the process is high and Trommel 2 separated the +40 mm particles for and biomining operations are set to take place in using it as Refused Derived Fuel. The particles various part of the country, the contribution of the which were less than 40mm were separated using air and noise pollution due to biomining process Air classifier which blowed -20mm particles (Low must be assessed and proper mitigation measures Density Plastic) from +20 mm particles (High has to be given to protect human health and the Density Plastic). Figure 1 shows the dumpsite environment. The objectives of the study are to mining process description at the mining facility. assess the ambient air quality and noise at various locations of dumpsite during the biomining process, 2.2 Sampling Location and Instrumentation 2.2.1 Measurement of PM2.5 The sampling locations were at the working level approximately 1 m above the ground surface at a distance of 1-1.5m from the equipment. Figure 1: Dumpsite Mining Process Description International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

The measurement of ambient air pollutants and P a g e | 165 noise was carried at 6 different locations during the biomining process and radial study at different The maximum concentration of PM 2.5 was intervals from the site operational facility.Instrumex observed near Panel which is at 10 m from the IPM-FDS 2.5-10 µ was used for the measurement of Vibro screen and minimum concentration was PM2.5. The method of measurement of PM2.5 is observed near Disc Screen Seperator. The mean Gravimetric method and reference used is IS 5182 value of PM2.5 concentration was 54 µg/m3and these Part 23. particles tend to stay longer in the air than heavier particles. Hence, they were relatively more inhaled 2.2.2 Measurement of Ambient Noise by the workers at the site. The maximum Protocol for Ambient Noise Level Monitoring was concentration for a period of 24 hours or 8 hours or used to measure the ambient Noise as per Central 1 hour prescribed by CPCB, NAAQs 2009 specifies Pollution Board, India standards. AZ 8921/2 RS232 60 µg/m3. Therefore, it became mandatory to limit Sound Level Meter was used for the measurement the PM2.5 concentration as low as possible by of ambient noise. The background noise absorber providing suitable mitigation measures like was placed over the microphone and digital readings provision of nose masks and spraying of water jets were noted down in ambient noise monitoring to reduce resuspension of dust in the ambient air at datasheet for every 30 second interval. The total workplace. area of the dumpsite biomining process was divided into three zones. The zone 1 of size 30m *13m was 3.2 Ambient Noise Level Monitoring Data divided into six grids approximately 70 m2. The The noise monitoring at dumpsite was carried out zone 3 of size 18m*12.5m was divided into 2 grids for a duration of six hours including 1 hour of of approximately 110 m2. Two methods were control sampling when the site was not operational employed in finding out ambient noise levels. for every 30s interval. At various locations, the ambient equivalent noise levels exceeds the i) Stationary monitoring at a grid study maximum permissible noise level of 85 dB ii) Moving through the grids study. (Continuous or intermittent noise) from a Factory or Workshop for a period of 8 hours, National In the first method, ambient noise level was Environment (Noise Standards and Control) measured at the center of grid for a period of 15 Regulations, 2013. minutes at a 30 second interval. In the second method, ambient noise level was measured at the 3.2.1 Grid study of ambient noise levels center of the grids from moving from one grid to The equivalent noise levels at various grids were another at an interval of 30 seconds. The total given in Figure 2. The trial 3 of Figure 2 represents duration of measurement of noise for each cycle is the moving through the grids study while the other 10 minutes. The total duration of measurement in a two trials represent the stationary study. It was day was 7 hours. The zone 2 of size 15m*13m was inferred from the data that the workplace was divided into 2 grids of approximately 100 m2. hazardous for the workers with high level of noise and suitable Personal Protective Equipments (PPEs) 3. RESULTS AND DISCUSSION like Earmuffs and Ear plugs should be provided to reduce impact on workers. 3.1 Measurements of PM2.5 The Ambient PM2.5 Concentrations were also 4. CONCLUSION measured at five different locations during months Air quality and Noise impact assessment in a of February and March 2019 and presented in Table dumpsite near Chennai during the dumpsite 1. biomining process was conducted between September and March 2019. The process released quantifiable amount of particulate matter PM2.5. The Dumpsite biomining process also created noise pollution and its effect on surrounding area is measured by two methods. CPCB 2009 and Ambient noise levels exceeds the standards as well. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 166 Figure 2: Ambient noise levels during dumpsite mining process Hence the mitigation measures can be followed to REFERENCES reduce the human and environmental impacts. It includes provision of suitable air control barriers Central Pollution Control Board, Ministry for such as closures and protective sheets to reduce direct impact, provision of high quality nose masks Environment Forest and Climate Change, 2017, (Preferably 3M 99.5 Nose mask) to reduce the particulate matter exposure, hourly spraying of Consolidated Annual Report on Solid Waste water using hose pipes or spraying of water using sprinklers which reduce the ambient PM 10 Management Rules,New Delhi. concentration caused due to resuspension of dust, Routine shifting of workers within the workplace Kerns, E., Elizabeth, A. M., Christa, L. T. and area and the provision of Personal protective Equipment such as ear plugs, ear muffs, etc for the Geoffrey, M. C., 2018, Cardiovascular reduction of Ambient Noise during dumpsite biomining operations. Conditions, Hearing Difficulty, and Occupational Noise Exposure within US Industries and Occupations. American Journal of Industrial Research, Vol. 61(6), 477-491. Mavropoulus, A. and Newman, D., 2015, Wasted Health the Tragic case of Dumpsites ISWA’s Scientific and Technical Committee Work Program. https://www.iswa.org/fileadmin/galle- ries/Task_Forces/THE_TRAGIC_CASE_OF_D UMPSITES.pdf Mavropoulos, A., Koller, H., Leucke, M., Shrestha, S. and Tanaka, M., 2014, Waste Atlas Map D – Waste Environmental Consultants Ltd., Athens https://www.iswa.org/fileadmin/galleries/News/ WASTE_ATLAS_2013_REPORT.pdf Ministry of Urban Development, SwachhSurvekshan Guidebook 2017, Government of India. Zala, J. and Jure, P., 2011, The Effects of Particular Matter Pollution on Respiratory Health And Cardiovascular System. Slovenian Journal of Public Health, Vol. 51,190-199. International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 167 Abstract International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 168 Geographical Information of Malaria Infections in Balochistan Province of Pakistan; A Trend Analysis of Malaria Sarmad Saeed Khan,1 Chairat Uthaipibull,2 Ghani Baloch3 and Choosak Nithikathkul11 1Tropical and Parasitic Diseases Research Unit, Faculty of Medicine, Mahasarakham University, Thailand E-mail: [email protected] 2BIOTEC’s Protein-Ligand Engineering and Molecular Biology Laboratory, NSTDA, Thailand 3Malaria Control Program Quetta, Balochistan, Pakistan ABSTRACT Pakistan bears a high burden of malaria cases despite tremendous achievements in both the malaria case management and prevention activities over the period of time. The purpose of this study was to conduct a trend analysis of Plasmodium species prevalence, incidence of malaria cases and geographic distribution of malaria species in Balochistan. The aggregated, annual malaria data of seven years (2013 to 2019) were obtained for the 30 districts of Balochistan province of Pakistan from the provincial Directorate of Malaria Control Program Quetta through formal request. All these summery sheets of the last seven years were compiled in MS Excel 2013 sheet and imported in SPSS version 23 for analysis. Population data for the districts were extrapolated at a growth rate of 2.3% annually from the census of 1998 for the years 2013 to 2017 and annual growth rate of 3.37% from census 2017 for year 2018 and 2019. The cumulative mean test positivity rate was 10.76 in province. This varied from high endemic districts being 25.8 to 2.1 for the low endemic districts. The test positivity rate was consistent over the period of all seven years. Plasmodium falciparum and Plasmodium vivax were the most prevalent species. The mean falciparum rate was 29.56 over the year of seven years, with an increasing trend from 28.23 in 2013 to 42 in 2019. This trend was seen for most of the districts which had very high falciparum rate in 2018 and 2019 and this trend is consistently high (>50) for some of the districts. Similarly, annual parasite incidence rate had a slight increasing trend through the years, being highest (API=9) in 2019. The findings from this analysis showed high burden of disease and trend showed that disease burden is far high from national targets set for the year 2020. High burden of the most dangerous species falciparum and its increasing trend is alarming and reflects serious threat which needs strengthening the preventive strategies. KEYWORDS: Malaria, Burden, Balochistan International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 169 Mapping and Intervention of Malaria Situation in Sisaket Province Sutthisak Noradee,1 Chairat Uthaipibull2 and Choosak Nithikathkul1 1Tropical and Parasitic Diseases Research Unit, Faculty of Medicine, Mahasarakham University, Thailand, E- mail: [email protected] 2National Center of Genetic Engineering and Biotechnology, NSTDA, Thailand ABSTRACT Malaria is a major public health problem in the tropics. In Thailand especially the border between Thailand Myanmar, and Cambodia province is highly endemic for malaria. Knowing this, the local Health Department has placed a program to educate the local residents about health risk factors that people in this area are facing, in particular the dangers and symptoms of a malaria infection. This study was performed to evaluate these efforts by determining the number of malaria infections in a segment of the population. The majority are Lao speaking people. The total population was 1.473 million people. The malaria report already showed 444.16, 318.41, 153.89 per one hundred thousand cases in Khuhan, Phu Sing, Kanchanalak and Khu Kan districs respectively, compared to infection rates in similar studies, the results of this study indicate overall that the Health Department's efforts are meeting with relative success. The low percentage of infected individuals shows that the villagers are using the information that they have received to help combat infection. With this study, we hope to provide valuable information to the residents and local Health Department in Si Sa Ket Province in order to develop further prevention programs for both diseases. However, all human malaria is transmitted by Anopheles mosquito, and in the absence of immunologic interactions, this would lead to them occurring together more often than expected by chance. Since we did not directly investigate the movement of villagers in the area, we cannot rule out the possibility that they acquired infections outside the village, such as from the forest area. The geographic information (latitude and longitude) associated with the infection rate among susceptible species of malaria was recorded and used to build a geographical information system. A number of environmental parameters such as mean yearly temperature, rainfall level, and population density were imported to the system as well. The developed GIS can be useful in the establishing of a prevention strategy for transmission of malaria infection from Si Sa Ket province. KEYWORDS: Malaria, GIS, Population, Endemic Area International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 170 Trend and Spatial Analysis of Dengue Cases in Northeast Malaysia Afiqah Syamimi Masrani,1 Nik Rosmawati Nik Husain,1 Kamarul Imran Musa1 and Ahmad Syaarani Yasin2 1Universiti Sains Malaysia, Malaysia, E-mail: [email protected], [email protected], [email protected] 2Kelantan State Health Department, Malaysia, E-mail: [email protected] ABSTRACT Dengue remains hyperendemic in Malaysia despite extensive vector control activities. With the dynamic changes in land use, urbanization and population movement, periodic updates on the pattern of dengue transmission is crucial to ensure implementation of effective control strategies. We sought to assess the shift in the trend and spatial distribution of dengue in Kelantan, the north-eastern state of Malaysia (5°15′N 102°0′E). This study incorporated data from the national dengue monitoring system (eDengue system). Confirmed dengue cases that were registered in Kelantan with disease onset between 1st January 2016 to 31st December 2018 were included in the study. The yearly changes in dengue incidence was mapped by using ArcGIS. Hotspot analysis was also performed using Getis-Ord Gi to track changes in the trend of dengue spatial clustering. A total of 10,645 dengue cases were recorded in Kelantan between 2016 to 2018 with an average of ten dengue cases reported daily (11.02 SD). Areas with persistently high dengue incidence were seen mainly at the coastal region for the three years. However, the trend of hotspot has shifted between the years with gradual dispersion of hotspots to their adjacent districts. A notable shift in the spatial distribution of dengue was observed. We were able to glimpse the shift of dengue from an urban to peri-urban disease with the possible effect of a state-wide population movement that affects the transmission of dengue. KEYWORDS: Dengue, Spatial Distribution, Hotspot Analysis International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 171 Spatial Density of Aedes Distribution in Urban Areas: A Case Study of Breteau Index in Melaka, Malaysia Mohd Hazrin Hashim,1 Farihan Yatim,2 Mohd Amierul Fikri Mahmud,3 Faizul Akmal Bin Abdul Rahim3 and Mohd Hatta bin Abdul Mutalip3 1Survey Research Center, Institute for Public Health, Malaysia, E-mail: [email protected] 2Disesase Control Division, Ministry of Health, Malaysia, E-mail: [email protected] 3Centre for Communicable Disease Research, Institute for Public Health, Malaysia E-mail: [email protected] ABSTRACT This study aims to establish the spatial density of mosquito population or breteau index (BI) in the areas of Melaka using geographic information system (GIS) and spatial statistical tools. The present study performed distribution and spatial pattern analysis of dengue cases reported in the growing Melaka district in Malaysia in 2017 and 2018. Data related to dengue cases were gathered from the Disease Control Division, Ministry of Health Malaysia. Geospatial analysis was applied to further study the spatio-temporal patterns of dengue cases in data set, including hot spot/cold spot analysis and geographically weighted regression models. This study found that the distribution pattern for dengue cases is clustered. Spatial mean center and directional distribution for both sets of years have slight differences. Meanwhile, standard distance for dengue cases reported in the year 2017 is 16,438.2 m, which is bigger than dengue cases reported in 2018, showing a standard distance of 14,583.1 m. In the present study, several hotspots identified will be beneficial to assist the local health authorities to reduce and eradicate mosquitoes in these areas. These results will provide valuable information through the application of advanced tools in combating Aedes mosquitoes. KEYWORS: Spatial, Leprosy, GIS International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 172 Spatio-Temporal Pattern Analysis of Typhoid and Diarrhea Cases and Associated Factors in Drought Districts of Andhra Pradesh, India 2013-2017 S V S Aditya Bharadwaz Ganni, Indrajit Pal and Nitin Tripathi Asian Institute of Technology, Thailand, E-mail: [email protected] ABSTRACT The present study carries out a Spatio-temporal analysis of typhoid and diarrhea incidence cases from 2013 – 2017 in two districts present in the state of Andhra Pradesh. Also will find the relationship between rainfall and disease incidence. Monthly rainfall data is also collected from the year 2013 to 2017 to identify its relationship with both the diseases. Districts having a small population can display false observed incidence in reality since they are more variable than villages with a larger population in order to solve this problem Empirical Bayes Smoothing (EBS) method will be used. Then Spatial Autocorrelation analysis is carried out in Arc GIS software. KEYWORS: Health GIS, Typhoid, Diarrhoea, Spatio-Temporal Analysis International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International

P a g e | 173 Exploring Geo-spatial Analysis to Map Malaria Endemicity and Applied Interventions in a Rural District of Chhattisgarh, India Hitakshi and Sunil Vilasrao Gitte National Vector Borne Disease Control Programme, MoHFW, India E-mail: [email protected], [email protected] ABSTRACT Spatial analysis and mapping of Malaria endemicity and applied interventions was done for past 4 years using an open source QGIS Software for Kawardha District of Chhattisgarh state in India. The endemicity was mapped on the basis of case load, Annual Parasite Incidence and type of species. The effect of interventions - Long Lasting Insecticidal Nets (LLINs) and Indoor Residual Spray (IRS) on the burden of Malaria was also studied over last four years. KEYWORS: Malaria, India, LLIN, IRS, GIS International Journal of Geoinformatics, Conference Proceedings for 7th HealthGIS Conference ISBN No: 978-616-90698-5-0 © Geoinformatics International