Enthusiastic : International Journal of Applied Statistics and Data Science

Implementation of Hotelling’s T2 Method in Quality and Capability Control of Newlab Collagen Production Processes

2025-03-18T05:50:34+00:00

Every company has quality standards that are determined for the production process. However, there are factors that occur in the production process that causes defects in the product. From these problems, this research was conducted to analyze the quality control, causal factors, and performance of the production process on Newlab Collagen products. The methods used in production quality control were Hotelling’s T² control chart, fishbone diagram, and process capability analysis. In the Hotelling’s T² control chart, the multivariate observation data was divided into two phases, with five quality indicators. The results of the first phase of the Hotelling’s T² control map showed that the quality indicators of the Newlab Collagen production were out of control, which caused by unstable machine factors. Based on control chart, the second phase showed that the quality indicators of the Newlab Collagen production process were still out of control. This condition was evidenced by the process capability value in phase I and phase II being less than one. These findings suggest that the company needs to make improvements, optimization, and quality control in the production.

Analysis of Factors that Influence Maternal Mortality Rates Using Generalized Poisson Regression

2025-06-17T09:38:17+00:00

Maternal Mortality Rate (MMR) is the number of deaths of women within 42 days after childbirth or during pregnancy. Objective: This study aims to identify factors affecting MMR in East Java and compare the performance of the Generalized Poisson Regression (GPR) model with Poisson regression. The method used is Generalized Poisson Regression, a regression model for count data, which extends Poisson regression to overcome the problem of overdispersion or underdispersion with data derived from the East Java Health Office, including MMR as the dependent variable, as well as five variables that are thought to affect it in 38 districts/cities. The GPR model proved superior to Poisson regression with an Akaike Information Criterion (AIC) value of 239.515 to identify factors affecting maternal mortality. Factors such as delivery handled by health workers, K6 visits by pregnant women, provision of diphtheria-tetanus immunization, and obstetric complications affect MMR in East Java in 2022.

Pension Funding Calculation Using the Benefit Prorate Method of the Constant Percent Type and Vasicek Interest Rates

2025-03-14T14:36:50+00:00

Pension funds are financial programs established by individuals or companies to secure the future of employees by providing benefits during their retirement years. Pension funds are built up through contributions made by both the participants (employees) and the employer. To calculate the amount of pension benefits, normal costs, and actuarial liabilities, various method are used, including the constant percent benefit prorate method. A key factor influencing these calculations is the interest rate. This study employs the Vasicek model, a stochastic interest rate model, to analyze the Unfunded Actuarial Liability (UAL). The analysisi reveals that the amount of normal cost (annual contributions) will vary, and both contributions and actuarial liabilities that are calculated using vasicek interest rate for each participant will adjust based on the interest rate during their retirement period. The amount of UAL is derived from the discrepancy between the total amount of actuarial liabilities from all the participant in certain period and the accumulated funds. The UAL is sufficient to cover future pension fund payments when calculated using the Vasicek interest rate model.

Spatio-Temporal Modeling of Crime Rates Using Geographically and Temporally Weighted Regression

2025-09-29T12:04:53+00:00

This study analyzes the spatio-temporal modeling of crime rates in 35 regencies and cities in Central Java using the geographically and temporally weighted regression (GTWR) method. The objective is to investigate how socio-economic factors, including the open unemployment rate, percentage of the poor population, population density, average years of schooling, job vacancies, labor force participation rate, and labor wage, influence crime rates across different regions and periods. The goodness-of-fit test results indicateed that the GTWR model had an R-squared value of 93.51%, higher than the 88.64% of the geographically weighted regression (GWR) model, demonstrating GTWR’s ability to explain crime data variations that were heterogeneous both spatially and temporally. Partial significance tests and mapping results showed that the influence of variables differed across years and regions, with population density and labor-related factors consistently being the main predictors. These findings highlight the importance of designing crime prevention policies that are locally tailored and based on spatio-temporal evidence.