BESTARI BPS Kalimantan Timur

CLUSTER ANALYSIS OF REGIONAL WASTE MANAGEMENT THROUGH WASTE BANKS IN KALIMANTAN USING FUZZY K-MEANS TO STRENGTHEN ENVIRONMENTAL POLICY

2025-12-15T07:27:06+00:00

The increase in waste generation in Kalimantan, along with population growth and economic activity, has become a complex environmental issue. The disparity in management capacity among regions has led to differences in the effectiveness of waste management systems, particularly in the implementation of waste banks as a form of community-based management. This study aims to cluster regencies/cities in Kalimantan based on waste management characteristics using the Fuzzy K-Means method to obtain a spatial mapping that supports the formulation of fairer and more efficient waste management policies. Secondary data were obtained from the National Waste Management Information System (SIPSN) and the Central Statistics Agency (BPS), covering variables such as waste generation, managed waste volume, land area, and population. The analysis results show that the optimal number of clusters is two (2): one cluster representing regions with high waste management performance dominated by major cities such as Balikpapan and Banjarmasin, and another cluster representing regions with low management performance due to limited infrastructure. These findings highlight spatial disparities in the effectiveness of waste bank programs across Kalimantan. The clustering results are expected to serve as a foundation for local governments in developing strategies to strengthen waste management policies, particularly through the implementation of the national program “1 RW 1 Waste Bank” and the adoption of sustainable circular economy principles.

The Influence of Economic and Environmental Factors on Deforestation in Indonesia

2025-12-15T06:57:41+00:00

Indonesia is one of the countries with the highest rate of deforestation in the world, influenced by various economic and environmental factors. This study aims to analyze the effect of Gross Regional Domestic Product (GRDP), forest area, population density, forest fires, and CO? emissions on deforestation in Indonesia during the 2018–2022 period using panel data regression analysis. Based on the Chow, Hausman, and Lagrange Multiplier tests, the Random Effects Model was identified as the most appropriate model to explain the relationships among variables. The results indicate that forest fires, forest area, and CO? emissions have a significant effect on deforestation, while GRDP and population density show no significant effect. Forest fires and CO? emissions positively influence deforestation, and forest area also shows a positive effect, indicating that provinces with larger forest coverage tend to record higher levels of deforestation. These findings suggest that environmental factors play a more dominant role than economic factors in determining the rate of deforestation in Indonesia.

Modeling MODELING AND FORECASTING RAINFALL IN BALIKPAPAN USING THE ARIMAX METHOD

2025-12-08T10:41:19+00:00

The city of Balikpapan experiences high and fluctuating rainfall intensity, rendering it vulnerable to hydrometeorological disasters such as floods and landslides. Accurate rainfall forecasting is crucial for risk mitigation. This study aims to model and forecast daily rainfall in Balikpapan using the Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) method, considering environmental factors. This study utilizes daily data from the January-December 2024 period sourced from NASA POWER, encompassing the variables of rainfall, average temperature, air humidity, and wind speed. The data was analyzed using the Box-Jenkins approach, which includes stationarity tests, parameter estimation, and diagnostic checks. The results indicated that the data was not stationary in variance, necessitating a logarithmic transformation. The best-fit model, identified by the lowest AIC value (582.05), was ARIMAX (1, 0, 1). Analysis of exogenous variables identified that Air Humidity and Wind Speed significantly influence rainfall, whereas Average Temperature does not. The Ljung-Box diagnostic test confirmed that the model's residuals behave as white noise (p-value 0.2662). The model's forecasting evaluation yielded an RMSE of 9.9617. The model proved reasonably effective in capturing the general rainfall patterns, despite limitations in predicting extreme spikes. These findings can contribute a scientific basis to support early warning systems and disaster mitigation policies in Balikpapan.

CLUSTERING ANALYSIS OF VILLAGES IN EACH PROVINCE OF INDONESIA BASED ON TYPES OF ENVIRONMENTAL POLLUTION IN 2024 USING PCA AND K-MEANS CLUSTERING

2025-12-07T11:43:31+00:00

Environmental pollution is a critical issue in Indonesia due to its impact on public health and ecosystem sustainability. Variations in pollution conditions across provinces indicate the need for analyses that can comprehensively describe spatial patterns. This study aims to classify 38 provinces in Indonesia based on the number of villages and urban villages according to types of environmental pollution, including water, soil, air pollution, and areas without pollution, in 2024. The data were obtained from official publications of Statistics Indonesia (BPS). The analysis employed Principal Component Analysis (PCA) as a dimensionality reduction technique, followed by K-Means Clustering to group provinces with similar pollution characteristics. The initial analysis was supported by descriptive statistical exploration and data standardization. The PCA results show that two principal components explain 89.41% of the total data variance. The optimal number of clusters was determined using the Elbow method and Silhouette coefficient, indicating that a two-cluster solution provides the most appropriate clustering structure (Silhouette score = 0.40). The clustering results reveal differences in environmental pollution characteristics between provinces in western and eastern Indonesia. These findings provide an initial, area-based descriptive overview of environmental pollution distribution in Indonesia and can support regional environmental management and more targeted policy formulation.

Air Quality Forecasting In Balikpapan City Based On PM2.5 Indicator Values Using The Autoregressive Integrated Moving Average (ARIMA) Method

2025-12-07T14:45:53+00:00

Air quality is an important indicator in assessing environmental conditions because it directly affects human health. The increase in industrial activities, transportation, and infrastructure development in Balikpapan City along with the development of the Nusantara Capital (IKN) has the potential to increase the concentration of air pollutants, particularly fine particulate matter PM2.5. This study aims to analyze patterns and forecast PM2.5 concentrations in Balikpapan City. This study uses the Autoregressive Integrated Moving Average (ARIMA) method to model daily PM2.5 time series data. The data used covers the period from January to September 2025 with a total of 273 observations, divided into 80% training data, which is 218 observations, and 20% testing data, which is 55 observations. The ARIMA method was chosen because of its ability to capture fluctuating patterns in time series data. The research results indicate that the ARIMA(2,0,1) model is the best model for forecasting PM2.5 concentration in Balikpapan City based on model selection criteria and forecast performance evaluation. This model is able to represent historical data patterns well and provides fairly accurate forecasting results on the test data. The conclusion of this study shows that the ARIMA(2,0,1) model can be used as an air quality forecasting tool, particularly for PM2.5 concentration in Balikpapan City, and has the potential to support policy-making in controlling air pollution in the area.

Analysis of Factors Affecting the Amount of Waste Generation in East Java Province Using the Random Forest Regression Method

2025-12-19T00:11:53+00:00

The issue of waste generation in East Java Province continues to rise in line with population growth and economic activities. This study aims to analyze socioeconomic factors influencing waste generation in 2024 using the Random Forest Regression (RFR) method. The dataset consists of secondary data from SIPSN and BPS, covering 38 districts and cities with five predictor variables: land area, population growth, population density, average expenditure per capita, and GRDP. The optimal model parameters were obtained at mtry = 2 and ntree = 500. The results indicate that GRDP and land area are the most influential factors based on the feature importance analysis. An R² value of 0.198 suggests that other unobserved variables also contribute to variations in waste generation. This study is expected to provide data-driven insights for local governments in formulating more effective and sustainable waste management policies.