ABSTRACT
In this pandemic period, pneumonia is often found in various cases. In many cases, COVID-19 has an adverse effect on people with pneumonia. Early detection of pneumonia can help health institutions map pneumonia in the community. However, pneumonia detection still uses conventional methods and takes a long time. This study detects pneumonia bacteria consisting of Acinetobacter baumannii and Pseudomonas aeruginosa. This study uses the DIBaS database, which consists of several bacterial images. This database is used to compare two classes, namely pneumonia and non-pneumonia. Detection is carried out using an artificial intelligence approach using the DenseNet121 and DenseNet169 methods. This study also uses the Genetic Algorithm optimization method to increase the accuracy of detecting pneumonia bacterial cells. The Genetic Algorithm provides random values for the last two DenseNet121 and DenseNet169 training layers. As a result, the accuracy of the DenseNet121 and DenseNet169 methods reached 95% and 96.7%, respectively. The optimization method intervention gave an increase of 5.2% and 3.4% over the original method, respectively. The best model results from this method are used as a reference model in making applications for the rapid detection of pneumonia with an average speed of accuracy reaching 4.25s. This computer-based application provides promising results for the future to be applied to the broader community. © 2022, Editura Universitatii din Oradea. All rights reserved.
ABSTRACT
In this pandemic period, pneumonia is often found in various cases. In many cases, COVID-19 has an adverse effect on people with pneumonia. Early detection of pneumonia can help health institutions map pneumonia in the community. However, pneumonia detection still uses conventional methods and takes a long time. This study detects pneumonia bacteria consisting of Acinetobacter baumannii and Pseudomonas aeruginosa. This study uses the DIBaS database, which consists of several bacterial images. This database is used to compare two classes, namely pneumonia and non-pneumonia. Detection is carried out using an artificial intelligence approach using the DenseNet121 and DenseNet169 methods. This study also uses the Genetic Algorithm optimization method to increase the accuracy of detecting pneumonia bacterial cells. The Genetic Algorithm provides random values for the last two DenseNet121 and DenseNet169 training layers. As a result, the accuracy of the DenseNet121 and DenseNet169 methods reached 95% and 96.7%, respectively. The optimization method intervention gave an increase of 5.2% and 3.4% over the original method, respectively. The best model results from this method are used as a reference model in making applications for the rapid detection of pneumonia with an average speed of accuracy reaching 4.25s. This computer-based application provides promising results for the future to be applied to the broader community. © 2022, Editura Universitatii din Oradea. All rights reserved.
ABSTRACT
Indonesia is often referred to as the Ring of Fire so that there are many natural disasters, especially earthquakes, volcanic eruptions, tsunamis and landslides. Based on the Environmental Profile, geography and population can increase the risk of natural disasters in Indonesia. To overcome these losses, it is necessary to mitigate natural disaster losses. The efforts of the Indonesian government as regulated in Law No. 24 of 2007, have allocated a budget for disaster management in the APBN/APBD. The government needs to reform natural disaster risk financing policies to meet the need for large amounts of disaster funding, timely and targeted, more planned, sustainable and transparent to reduce economic losses and the burden on the state budget. This disaster risk financing must also be able to answer the financing needs when there is no disaster in the context of disaster risk mitigation and transfer, financing when a disaster occurs (emergency response) and financing after a disaster occurs (rehabilitation and reconstruction). One way of financing disaster risk that can be done is with insurance services or insurance companies. The insurance company is referred to as the individual risk insurer who follows the insurance. In insurance there is a contractual agreement between the insured and the insurer called the policy. The agreement is regarding the insurer who is willing to bear a number of risks that may arise in the future in exchange for certain payments from the insured. Payments made by the insured to the insurer are called premiums.
ABSTRACT
This study empirically measures and analyzes determinants of productivity changes of the co-operatives across all 34 provinces in Indonesia over the 2015-2020 period using a-two stage approach. In the first stage, the study measures the productivity of the co-operatives using Data Envelopment Analysis (DEA). Meanwhile, in the second stage, the study utilizes a panel regression model to measure and analyze the determinants of productivity of the co-operatives in Indonesia. The study recorded that the co-operatives in Indonesia have experienced a 9.7% increase in their Total Factor Productivity (TFP), contributed mainly by the technical efficiency progress. Furthermore, the study found that the business volume has contributed to the improvement of the co-operatives' TFP. Meanwhile, the co-operatives' members, liquidity, and indebtedness are found to deteriorate the TFP growth. Profitability, however, is found to have an insignificant effect on TFP growth due to the non-profit orientation of the co-operatives. These findings suggest the need for cooperatives to diversify their business activities, supported by the adoption of relevant advanced technologies, particularly the use of online marketing. In addition, the co-operatives should improve their financial performances to maximize the use of capital by restricting liquidity and indebtedness. Finally, the government support to enhance financial and managerial aspects is essential to realize the co-operatives as the pillar of Indonesia's economy, as mandated by the 1945 Indonesian constitution.
ABSTRACT
PETRONAS completed well H-X on B field in Malaysia with a digital intelligent artificial lift (DIAL) gas lift production optimization system. This DIAL installation represents the first ever successful installation of the technology in an Offshore well for Dual String production. This paper provides complete details of the installation planning and operational process undertaken to achieve this milestone. DIAL is a unique technology that enhances the efficiency of gas lift production. Downhole monitoring of production parameters informs remote surface-controlled adjustment of gas lift valves. This enables automation of production optimization removing the need for well intervention. This paper focusses on a well completed in November 2020, the fourth well to be installed with the DIAL technology across PETRONAS Assets. The authors will provide details of the well and the installation phases: system design, pre-job preparations, improvements implementation, run in hole and surface hookup. For each phase, challenges encountered, and lessons learned will be listed together with observed benefits. DIAL introduces a paradigm shift in design, installation and operation of gas lifted wells. This paper will briefly highlight the justifications of this digital technology in comparison with conventional gas lift techniques. It will consider value added from the design stage, through installation operations, to production optimization. This DIAL system installation confirms the ability to be implement the technology in a challenging dual string completion design to enable deeper injection while avoiding interventions on a well with a greater than 60-degree deviation. With remotely operated, non-pressure dependent multi-valve in-well gas lift units, the technology removes the challenges normally associated with gas-injected production operation in a dual completion well - gas robbing and multi-pointing. Despite the additional operational & planning complications due to COVID-19 restrictions, the well was completed with zero NPT and LTI. Once brought online, this DIAL-assisted production well will be remotely monitored and controlled ensuring continuous production optimization, part of PETRONAS' upstream digitization strategic vision. © Copyright 2021, Society of Petroleum Engineers
ABSTRACT
The demand of face masks had increased tremendously due to pandemic outbreak of COVID-19, leading to the increment production rate of face masks in Malaysia. Waste is also produced at the same time, resulting impacts towards the environment. Due to the land scarcity issue in Malaysia, the end of life treatment for the waste is taken into consideration. The study tools used in this study is life cycle assessment (LCA) to identify the significant potential environmental impact produced during the life cycle stages for distribution and disposal through GaBi Education Software. The disposal stage between landfill and incineration of the 3-Ply face masks is studied to determine the end of life treatment for it. The impact assessment method selected in this study is CML 2001-Jan 2016 with the environmental indicator of Global Warming Potential (GWP), Acidification Potential (AP), Eutrophication Potential (EP) and Ozone Depletion Potential (ODP). GWP results in producing highest impact to the environment during both distribution and disposal stages. The impact of GWP also relates to the climate change. Modern incineration is recommended to overcome the issue of land scarcity in Malaysia as the amount of waste by 3-Ply face masks are increasing due to the pandemic COVID-19, reducing the impacts towards the environment. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
The biggest challenge facing the world in 2020 was the pandemic of the coronavirus disease (COVID-19). Since the start of 2020, COVID-19 has invaded the world, causing death to people and economic damage, which is cause for sadness and anxiety. Since the world has passed from the first peak with relative success, this should be evaluated by statistical analysis in preparation for potential further waves. Artificial neural networks and logistic regression models were used in this study, and some statistical indicators were extracted to shed light on this pandemic. WHO website data for 32 European countries from 11th of January 2020 to 29th of May 2020 was utilized. The rationale for choosing the stated methodological tools is that the classification accuracy rate of artificial neural networks is 85.6% while the classification accuracy rate of logistic regression models 80.8%.