• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• The Journal of Society for e-Business Studies
• /
• v.15 no.1
• /
• pp.139-155
• /
• 2010

The purpose of this paper is to develop a framework of software cost estimation guidelines and to derive a mid-long term direction for development of the software cost estimation guidelines. In this paper, all the steps in the software life cycle are researched in the view of cost estimation, and current software cost estimation guidelines and models have been reviewed and analysed first. Second, a plan to separate unit cost per function point from standard procedure in current software cost estimation guidelines is presented to strengthen maket self-regulating function as a mid-long term developmental direction for software cost estimation guidelines. Third, construction of cost repository, making standard procedure for software cost estimation guidelines, development of various kinds of software cost estimation models, and a system for experts on software cost estimation are presented as the prerequisites for the future model framework of software cost estimation guidelines. Finally a roadmap for establishing the future model is proposed.

• Journal of the military operations research society of Korea
• /
• v.4 no.1
• /
• pp.91-109
• /
• 1978

The paper discusses Cost and Operational Effectiveness Analysis (COEA) in the context of a military study system. The objectives of a study system, development of study programs, and management of studies are addressed. Basic cost-effectiveness models are presented and the COEA planning steps, analytical processes, and implementation discussed.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2270-2273
• /
• 2001

This paper presents calibration and alignment algorithms for low-cost inertial sensors. The error models for gyro and accelerometer are presented with a study of their effects. A navigational Kalman Filter is derived based on those error models. Test results are presented, which shows the initial calibration and alignment scheme and the proposed filter configuration effectively reduce the drift of the sensors and provide improved accuracy for its practical use for navigation.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2018.05a
• /
• pp.294-295
• /
• 2018

Nowadays, the construction industry utilizes 3D models in the designing process, on which research is being conducted to establish an automated system for project cost estimation in connection with information related to construction such as material unit costs and wages, beyond the level of design interference review and construction quantity estimation. In this process, the project cost is estimated in connection with unit price data after takeoff the quantity based on the 3D model attributes and data types. A way to reduce cost and risk would be first developing prototypes of some of essential buildings and works, comparing and validating the outcomes, and then extending to the whole scope, because estimates differ on the basis of the scope and level of 3D design models as well as the data accuracy. This study analyzes case studies of project cost estimation by computing the quantity on the basis of 3D model in the construction industry and explores methodologies and management measures applicable for estimating nuclear power plant construction project costs.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.19 no.1
• /
• pp.87-111
• /
• 2021

Nuclear power plants (NPPs) produce radioactive waste and decommissioning this waste entails additional cost; determining these costs for various types and specifications of radioactive waste can be challenging. The purpose of this study is to identify major determinants of the decommissioning cost and their impact on NPPs. To this end, data from defunct NPPs were gathered and 2SLS (Two Stage Least Squares) regression models were developed to investigate the major contributors depending on the reactor types, viz. PWR (Pressurized Water Reactors) and BWR (Boiling Water Reactors). Additionally, cost estimations and the Monte Carlo simulation were performed as part of performance validation. Our study established that the decommissioning costs are primarily influenced by the level of radioactivity in the decommissioned waste, which can be realized from operational factors like operation period, overall efficiency, and plant capacity, as well as from duration of decommissioning and labour cost. While our study provides an improved statistical approach to recognize these factors, we acknowledge that our models have limitations in forecasting accurately which we envisage to bolster in future studies by identifying more substantive factors.

• Proceedings of the KSR Conference
• /
• 2004.06a
• /
• pp.392-396
• /
• 2004

The number of parameters and the number of samples are the key point of trans-log cost function which studied recently. Most of models show that the number of parameters is more than the number of samples. Therefore, these studies gave unreliability of the estimation results. First, we surveyed theoretical cases and researches for the formulation of cost function of railroad industry. Second, we will suggest trans-log cost function by analyzing cost data of KNR. And, the cumulative data will be needed for the confidence of estimation results.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2004.10a
• /
• pp.157-162
• /
• 2004

Recently, Life Cycle Cost (LCC) for civil infrastructures, such as pavements, bridges, and dams, has been emphasized. However there are few cost models for road tunnel especially for maintenance phase. The road network is composed of highways, bridges, and road tunnels. Thus it is as important as for road tunnels to keep safe for traffic. The maintenance strategies for road tunnels can be achieved based on the minimization of LCC in maintenance phase. For this purpose, in this paper, cost model and cost classification for road tunnel in maintenance phase are suggested.

• Proceedings of the KSR Conference
• /
• 2004.10a
• /
• pp.1765-1769
• /
• 2004

The number of parameters and the number of samples are the key point of trans-log cost function which studied recently. Most of models show that the number of parameters is more than the number of samples. Therefore, these studies gave unreliability of the estimation results. First, we surveyed theoretical cases and researches for the formulation of cost function of railroad industry. Second, we will suggest trans-log cost function by analyzing cost data of KNR. And, the cumulative data will be needed for the confidence of estimation results.

• Journal of Korean Society of Water and Wastewater
• /
• v.31 no.1
• /
• pp.51-62
• /
• 2017

For the asset management of a water pipe network, it would be necessary to understand the extent of the maintenance cost required for the water pipe network for the future. This study would develop a method to draw the optimum cost required for the maintenance of the water pipe network in waterworks facilities to maintain the aim revenue water ratio and to achieve the target revenue water ratio, considering the water service providers' waterworks condition and revenue water ratio comprehensively. This study conducted a survey with 96 water service providers as of the early 2015 and developed models to estimate the optimum maintenance cost of the water pipe network, considering the characteristics of the water service providers. Since the correlation coefficient of all the developed models was higher than 0.95, it turned out that it had significant reliability, which was statistically significant. As a result of applying the developed models to the actual water service providers, it was drawn that increasing revenue water ratio to more than a certain level can reduce the maintenance cost of the water pipe network by a great deal. In other words, it is judged that it would be the most efficient to secure the reliability of waterworks management by increasing the short-term revenue water ratio to more than a certain level and gradually increase the revenue water ratio from the long-term perspective. It is expected that the proposed methodology proposed in this study and the results of the study will be used as a basic research for planning the maintenance of water pipe network or establishing a plan for waterworks facilities asset management.