Explainable AI (XAI) is an approach that leverages artificial intelligence to support human decision-making. Recently, governments of several countries including Korea are attempting objective evidence-based analyses of R&D investments with returns by analyzing quantitative data. Over the past decade, governments have invested in relevant researches, allowing government officials to gain insights to help them evaluate past performances and discuss future policy directions. Compared to the size that has not been used yet, the utilization of the text information (accumulated in national DBs) so far is low level. The current study utilizes a text mining strategy for monitoring innovations along with a case study of smart-farms in the Honam region.
The paint removal of fighter jets is just as important as the painting, because perfect paint removal ensures the quality of the exterior painting on the aircraft. However, the current conditions for paint removal work of the ROKAF’s are poor. It is identified that the painting process currently implemented by the ROKAF is not only exposed to harmful compounds such as harmful dust and hexavalent chromium, but also consumes a lot of water. Thus, the introduction of advanced facility is considered. This study compares the fighter jets painting removal process currently applied by the Korean Air Force with the improved laser coating removal process of the US Air Force, and conducts an incremental analysis to perform economic analysis for the introduction of advanced facility. Four scenarios were envisioned on the premise of an increase in the number of fighters in the future, incremental analysis shows that laser coating removal method is advantageous in all scenarios. In addition, it is recommended that paint removal cycle keeps the current 12-year and the outsourcing amount to civilian depot is reduced.
To mitigate the environmental impacts of the energy sector, the government of South Korea has made a continuous effort to facilitate the development and commercialization of renewable energy. As a result, the efficiency of renewable energy plants is not a consideration in the potential site selection process. To contribute to the overall sustainability of this increasingly important sector, this study utilizes the Black-Scholes model to evaluate the economic value of potential sites for off-site wind farms, while analyzing the environmental mitigation of these potential sites in terms of carbon emission reduction. In order to incorporate the importance of flexibility and uncertainty factors in the evaluation process, this study has developed a site evaluation model focused on system dynamics and real option approaches that compares the expected revenue and expected cost during the life cycle of off-site wind farm sites. Using sensitivity analysis, this study further investigates two uncertainty factors (namely, investment cost and wind energy production) on the economic value and carbon emission reduction of potential wind farm locations.
As the environmental impacts of fossil fuel energy sources increase, the South Korean government has tried to change non-environmental- friendly enery sources to environmental-friendly energy sources in order to mitigate environmental effects, which lead to global warming and air pollution. With both a limited budget and limited time, it is essential to accurately evaluate the economic and environmental effects of renewable energy projects for the efficient and effective operation of renewable energy plants. Although the traditional economic evaluation methods are not ideal for evaluating the economic impacts of renewable energy projects, they can still be used for this purpose. Renewable energy projects involve many risks due to various uncertainties. For this reason, this study utilizes a real option method, the Geske compound model, to evaluate the renewable energy projects on Jeju Island in terms of economic and environmental values. This study has developed an economic evaluation model based on the Geske compound model to investigate the influences of flexibility and uncertainty factors on the evaluation process. This study further conducts a sensitivity analysis to examine how two uncertainty factors (namely, investment cost and wind energy production) influence the economic and environmental value of renewable energy projects.
North Korea continues to upgrade and display its long-range rocket launchers to emphasize its military strength. Recently Republic of Korea kicked off the development of anti-artillery interception system similar to Israel’s “Iron Dome”, designed to protect against North Korea’s arsenal of long-range rockets. The system may not work smoothly without the function assigning interceptors to incoming various-caliber artillery rockets. We view the assignment task as a dynamic weapon target assignment (DWTA) problem. DWTA is a multistage decision process in which decision in a stage affects decision processes and its results in the subsequent stages. We represent the DWTA problem as a Markov decision process (MDP). Distance from Seoul to North Korea’s multiple rocket launchers positioned near the border, limits the processing time of the model solver within only a few second. It is impossible to compute the exact optimal solution within the allowed time interval due to the curse of dimensionality inherently in MDP model of practical DWTA problem. We apply two reinforcement-based algorithms to get the approximate solution of the MDP model within the time limit. To check the quality of the approximate solution, we adopt Shoot-Shoot-Look(SSL) policy as a baseline. Simulation results showed that both algorithms provide better solution than the solution from the baseline strategy.
In a group-testing method, instead of testing a sample, for example, blood individually, a batch of samples are pooled and tested simultaneously. If the pooled test is positive (or defective), each sample is tested individually. However, if negative (or good), the test is terminated at one pooled test because all samples in the batch are negative. This paper considers a queueing system with a two-stage group-testing policy. Samples arrive at the system according to a Poisson process. The system has a single server which starts a two-stage group test in a batch whenever the number of samples in the system reaches exactly a predetermined size. In the first stage, samples are pooled and tested simultaneously. If the pooled test is negative, the test is terminated. However, if positive, the samples are divided into two equally sized subgroups and each subgroup is applied to a group test in the second stage, respectively. The server performs pooled tests and individual tests sequentially. The testing time of a sample and a batch follow general distributions, respectively. In this paper, we derive the steady-state probability generating function of the system size at an arbitrary time, applying a bulk queuing model. In addition, we present queuing performance metrics such as the offered load, output rate, allowable input rate, and mean waiting time. In numerical examples with various prevalence rates, we show that the second-stage group-testing system can be more efficient than a one-stage group-testing system or an individual-testing system in terms of the allowable input rates and the waiting time. The two-stage group-testing system considered in this paper is very simple, so it is expected to be applicable in the field of COVID-19.
The function of coolant in machining is to reduce the frictional force in the contact area in between the tool and the material, and to increase the precision by cooling the work-piece and the tool, to make the machining surface uniform, and to extend the tool life. However, cutting oil is harmful to the human body because it uses chlorine-based extreme pressure additives to cause environmental pollutants. In this study, the effect of cutting temperature and surface roughness of titanium alloy for medical purpose (Ti-6Al-7Nb) in eco-friendly ADL slot shape machining was investigated using the response surface analysis method. As the design of the experiment, three levels of cutting speed, feed rate, and depth of cut were designed and the experiment was conducted using the central composite planning method. The regression expressions of cutting temperature and surface roughness were respectively obtained as quadratic functions to obtain the minimum value and optimal cutting conditions. The values from this formula and the experimental values were compared. As a result, this study makes and establishes the basis to prevent environmental pollution caused by the use of coolant and to replace it with ADL (Aerosol Dry Lubricant) machining that uses a very small amount of vegetable oil with high pressure.
Recently, elevator inspection and self-examination were strengthened through the revision of the Elevator Safety Management Act, but there have been no significant reduction in serious accidents and major failures. Therefore, the government intends to lay the foundation for reflecting the safety quality rating system, which adjusts the elevator inspection cycle, as a policy to induce safety management of preemptive and active management entities. This study systematically reviewed and classified the safety quality rating system for elevator inspection cycle adjustment in previous studies, collected expert opinions, and reconstructed the key items into realistic evaluation items, and evaluated and scored the relative importance of each factor through the AHP technique.
Various non-face-to-face services are being activated due to the influence of the Corona 19 virus around the world. However, unlike the rapid development of delivery services, social awareness of delivery services is causing many problems. Therefore, in this study, we analyze the quality attributes of delivery services from the consumer's point of view, and based on the results, we try to derive a direction for service improvement. In this study, quality factors were established through interviews and surveys with actual consumers, and quality attributes were classified through the Kano model and Timko’s customer satisfaction coefficient. “Attractive” is (‘Ease of ordering, Accurate delivery to the designated place’), “One Dimensional” is (‘Variety of payment methods, Accurate delivery on time, Accurate delivery of ordered food, Degree of non-deformation of packaging conditions, etc., Convenience of use time’), “Must be” is (‘Kindness of the delivery person’), “Reverse” is (‘provision of services, service response to order discrepancies’). This study has academic significance in that it compensated for the disadvantage of not being able to interpret the mathematical meaning of the Kano model with Teamco’s customer satisfaction coefficient. It also has practical implications in that it provides an indirect clue to future improvement directions.
Recently, many studies have been conducted to improve quality by applying machine learning models to semiconductor manufacturing process data. However, in the semiconductor manufacturing process, the ratio of good products is much higher than that of defective products, so the problem of data imbalance is serious in terms of machine learning. In addition, since the number of features of data used in machine learning is very large, it is very important to perform machine learning by extracting only important features from among them to increase accuracy and utilization. This study proposes an anomaly detection methodology that can learn excellently despite data imbalance and high-dimensional characteristics of semiconductor process data. The anomaly detection methodology applies the LIME algorithm after applying the SMOTE method and the RFECV method. The proposed methodology analyzes the classification result of the anomaly classification model, detects the cause of the anomaly, and derives a semiconductor process requiring action. The proposed methodology confirmed applicability and feasibility through application of cases.
Today, as technology advances and market competition for products intensifies, the product design to improve customer satisfaction by accurately identifying customer needs is emerging as a very important issue for company. Accordingly, the customer-oriented or customer-centered design that maximizes customer satisfaction by grasping and analyzing customer requirements is in the spotlight as an important design theory. In this study, the customer-oriented design is defined as finding the optimal value of design variable with the maximum overall customer satisfaction while minimizing the difference in individual customer satisfaction responded to various customers from multiple product quality characteristics from the perspective of robust design. Therefore, this study presents a new method for modeling the customer preference structure as the different sets of desirability functions for multiple quality characteristics and proposes a new customer-oriented design approach by applying the desirability functions to Taguchi’s robust design process to deal with multi-characteristic design problem. Finally, the proposed method is illustrated with the Kansei engineering design problem of wine glass.
Most of real-world decision-making processes are used to optimize problems with many objectives of conflicting. Since the betterment of some objectives requires the sacrifice of other objectives, different objectives may not be optimized simultaneously. Consequently, Pareto solution can be considered as candidates of a solution with respect to a multi-objective optimization (MOP). Such problem involves two main procedures: finding Pareto solutions and choosing one solution among them. So-called multi-objective genetic algorithms have been proved to be effective for finding many Pareto solutions. In this study, we suggest a fitness evaluation method based on the achievement level up to the target value to improve the solution search performance by the multi-objective genetic algorithm. Using numerical examples and benchmark problems, we compare the proposed method, which considers the achievement level, with conventional Pareto ranking methods. Based on the comparison, it is verified that the proposed method can generate a highly convergent and diverse solution set. Most of the existing multi-objective genetic algorithms mainly focus on finding solutions, however the ultimate aim of MOP is not to find the entire set of Pareto solutions, but to choose one solution among many obtained solutions. We further propose an interactive decision-making process based on a visualized trade-off analysis that incorporates the satisfaction of the decision maker. The findings of the study will serve as a reference to build a multi-objective decision-making support system.
This study attempts a comparison between AHP(Analytic Hierarchy Process) in which the importance weight is structured by individual subjective values and regression model with importance weight based on statistical theory in determining the importance weight of casual model. The casual model is designed by for students’ satisfaction with university, and SERVQUAL modeling methodology is applied to derive factors affecting students’ satisfaction with university. By comparison of importance weights for regression model and AHP, the following characteristics are observed. 1) the lower the degree of satisfaction of the factor, the higher the importance weight of AHP, 2) the importance weight of AHP has tendency to decrease as the standard deviation(or p-value) increases. degree of decreases. the second sampling is conducted to double-check the above observations. This study empirically checks that the importance weight of AHP has a relationship with the mean and standard deviation(or p-value) of independence variables, but can not reveal how exactly the relationship is. Further research is needed to clarify the relationship with long-term perspective.
Many manufacturers applying third party logistics (3PLs) have some challenges to increase their logistics efficiency. This study introduces an effort to estimate the weight of the delivery trucks provided by 3PL providers, which allows the manufacturer to package and load products in trailers in advance to reduce delivery time. The accuracy of the weigh estimation is more important due to the total weight regulation. This study uses not only the data from the company but also many general prediction variables such as weather, oil prices and population of destinations. In addition, operational statistics variables are developed to indicate the availabilities of the trucks in a specific weight category for each 3PL provider. The prediction model using XGBoost regressor and permutation feature importance method provides highly acceptable performance with MAPE of 2.785% and shows the effectiveness of the developed operational statistics variables.
Transportation in urban area has been getting hard to fulfill the demand on time. There are various uncertainties and obstacles related with road conditions, traffic congestions, and accidents to interrupt the on-time deliveries. With this situation, the last mile logistics has been a keen issue for researchers and practitioners to find the best strategy of the problem. A way to resolve the problem is to use parcel lockers. Parcel locker is a storage that customers can pick up their products. Transportation vehicles deliver the products to parcel lockers instead of all customer sites. Using the parcel lockers, the total delivery costs can be reduced. However, the inconvenience of customer has to increase. Thus, we have to optimal solution to balance between the total delivery costs and customers' inconvenience. This paper formulates a mathematical model to find the optimal solution for the vehicle routing problem and the location problem of parcel lockers. Experimental results provide the viability to find optimal strategy for the routing problem as well as the location problem.
This study attempts to analyze the economic impact of the service robot industry using Input-Output analysis, which is conducted based on Demand-driven model, the Leontief price model, the Backward and Forward Linkage Effects, and the Exogenous Methods. In a Demand-driven model analysis, we can conclude that the service robot industry contains characteristics of both the manufacturing industry and the service industry, which causes a positive impact on the overall industry by compensating for the weaknesses of the two industries. The Leontief price analysis indicates when wages in the service robot industry increase, prices related to robot manufacturing also increase. Also, when profits in the service robot industry increase, prices related to service provision increase, too. The Backward and Forward Linkage Effects analysis shows that the service robot industry is highly sensitive to the current economic condition and has a great influence on the service industry. The service robot industry can highlight the aspect of service characteristics when the manufacturing industry is in recession and vice versa. In addition, the service robot industry can be regarded as a value-adding and domestic economy promoting industry which utilizes knowledge of information and communication technologies. It is important to foster the service robot industry in South Korea, which is in economic recession to provide an opportunity to stimulate the growth of both service and robot industries.
As the 4th industrial revolution emerges, the implementation of smart factories are essential in the manufacturing industry. However, 80% of small and medium-sized enterprises that have introduced smart factories remain at the basic level. In addition, in root industries such as injection molding, PLC and HMI software are used to implement functions that simply show operation data aggregated by facilities in real time. This has limitations for managers to make decisions related to product production other than viewing data. This study presents a method for upgrading the level of smart factories to suit the reality of small and medium-sized enterprises. By monitoring the data collected from the facility, it is possible to determine whether there is an abnormal situation by proposing an appropriate algorithm for meaningful decision-making, and an alarm sounds when the process is out of control. In this study, the function of HMI has been expanded to check the failure frequency rate, facility time operation rate, average time between failures, and average time between failures based on facility operation signals. For the injection molding industry, an HMI prototype including the extended function proposed in this study was implemented. This is expected to provide a foundation for SMEs that do not have sufficient IT capabilities to advance to the middle level of smart factories without making large investments.
In the case of a die-casting process, defects that are difficult to confirm by visual inspection, such as shrinkage bubbles, may occur due to an error in maintaining a vacuum state. Since these casting defects are discovered during post-processing operations such as heat treatment or finishing work, they cannot be taken in advance at the casting time, which can cause a large number of defects. In this study, we propose an approach that can predict the occurrence of casting defects by defect type using machine learning technology based on casting parameter data collected from equipment in the die casting process in real time. Die-casting parameter data can basically be collected through the casting equipment controller. In order to perform classification analysis for predicting defects by defect type, labeling of casting parameters must be performed. In this study, first, the defective data set is separated by performing the primary clustering based on the total defect rate obtained during the post-processing. Second, the secondary cluster analysis is performed using the defect rate by type for the separated defect data set, and the labeling task is performed by defect type using the cluster analysis result. Finally, a classification learning model is created by collecting the entire labeled data set, and a real-time monitoring system for defect prediction using LabView and Python was implemented. When a defect is predicted, notification is performed so that the operator can cope with it, such as displaying on the monitoring screen and alarm notification.
Investors must adopt profitable investment opportunities to maximize their wealth. Almost all investment, finance, engineering economics textbooks explain that net present value (NPV) measures the profitability (or value) of investment opportunities in absolute size, and internal rate of return (IRR) measures the profitability of investment opportunities in relative proportions. However, NPV is a measure of the relative size of the return on investment opportunity to do-nothing alternative. Moreover, IRR can occur in multiple investment opportunities and may not exist. To make matters worse, IRR and NPV also have conflicting problems in accept-or-reject decisions. In this study, the reason why NPV and IRR cannot accurately measure the profitability of investment opportunities is identified, and fundamental characteristics that investment opportunity profitability measures should have are presented.
Companies are making design changes by improving product quality and function to succeed while meeting customer requirements continuously. Design changes are changing the product BOM's amount, item, specification, and shape while causing a change in the product's structure. At this time, the problem of inventory exhaustion of parts before design change is a big topic. If the inventory exhaustion fails, the pieces before the design change become unused and are discarded, resulting in a decrease in asset value, and the quality cost of the design change affects the company's profits. Therefore, it is necessary to decide to minimize quality costs while minimizing waste inventory costs at the time of application of design changes. According to the analysis, priorities should be prioritized according to urgency because the quantity of items before the design change affects the applied lead time.