Hyunwoo Hwangbo - Academia.edu (original) (raw)

Uploads

Papers by Hyunwoo Hwangbo

The Society of Fashion and Textile Industry, Aug 1, 2017

ICT 기술의 발전에 따라 제조 산업은 공정 상에서 생성되는 제조 데이터를 분석하여 효율을 높이고자 많은 노력을 하고 있다. 본 논문에서는 스마트 공장의 일환으로 의사결정나무... more ICT 기술의 발전에 따라 제조 산업은 공정 상에서 생성되는 제조 데이터를 분석하여 효율을 높이고자 많은 노력을 하고 있다. 본 논문에서는 스마트 공장의 일환으로 의사결정나무 알고리즘(CHAID)을 이용한 데이터 마이닝 기반 제조공정을 제안한다. 약 5개월간 수집된 실제 제조 공정의 432개 센서 데이터를 활용하여 불량률이 낮은 안정적인 공정 기간과 불량률이 높은 불안정한 공정 기간 간에 유의미한 차이를 보이는 변수를 찾아냈다. 선정된 최종 변수가 불량률 개선에 실제로 효과가 있는지를 측정하기 위해 해당 변수의 안정 값 범위를 설정하여 14일 간 공정에서 해당 센서가 안정 값의 범위를 벗어나지 않도록 공정 설정 값을 조절했고, 불량률 개선의 효과를 측정하였다. 이를 통해 제조 산업에서 생성되는 공정 센서 데이터를 활용 및 분석하여 불량률을 개선할 수 있는 실증적인 가이드라인을 제시할 수 있을 것으로 기대한다.

This study investigated the effects of influencing factors on the sales volume of apparel product... more This study investigated the effects of influencing factors on the sales volume of apparel products. Based on previous studies, weekend effect, discount rate, and meteorological factors including daily average temperature, rainfall, sea level pressure, and fine dust were selected as independent variables to calculate their effects on sales quantity of apparel products. The daily sales data during 2015 - 2016 were collected from casual brands and outdoor brands which “A” apparel manufacturing company had operated. The actual data of “A” company were analyzed using SAS® 9.4 and SAS® Enterprise Miner 14.1. The results of this study were as follows: First, the influencing factors on total sales volume of apparel products were proved to be the weekend effect, discount rate, and fine dust. Second, the analysis of influencing factors on sales volume of apparel products according to season showed: 1) In casual brands, the average temperature had a significant influence on the sales volume of...

본 연구는 가격 할인 구간별 고객 수요의 민감도, 즉 탄력성을 통해 프로모션 기간 중 어느 시기, 어떤 가격대의 상품에 얼마만큼의 할인율을 적용하는 것이 수익을 극대화하는 가... more 본 연구는 가격 할인 구간별 고객 수요의 민감도, 즉 탄력성을 통해 프로모션 기간 중 어느 시기, 어떤 가격대의 상품에 얼마만큼의 할인율을 적용하는 것이 수익을 극대화하는 가장 효율적인 방안인지를 실증적으로 분석하였다. 이를 위해 본 연구는 한국의 패션 A기업에 대한 4년간의 상품 가격 및 판매량 데이터를 바탕으로 시즌 구간을 설정하고 동일한 조건의 가격탄력성을 구한 후 그에 따른 최적 할인율 구간을 도출하였다. 본 연구의 결론으로 한국의 경우, 20% 할인율에서 그리고 시즌 초기보다는 중반기 이후에 주로 민감하게 반응하여 소위 구매 여력이 되는 소비자들이 먼저 구매를 하고, 시즌 후반기로 갈수록 가격 할인을 통해 프로모션을 진행하는 것이 효과적이라는 결론을 내릴 수 있는 반면, 중국은 할인율이 30% 이상, 그것도 시즌 초기 할인에 민감하게 반응한다는 것을 알아냈다. 이는 초기 브랜드에 민감하게 반응하는 중국의 집단주의적 성격과도 매우 관련이 있으며, 특히 단일 요소 분석에서 드러난 것과 같이 고가의 상품에 대해 민감하게 반응하는 중국인의 특성상 명품에 대한 선호와도 괘를 같이 하는 것이라 짐작할 수 있다. 본 연구는 4년간에 걸친 다국적 기업의 대용량 데이터를 바탕으로 실증분석을 시행함으로써 다양한 정책에 활용할 수 있는 실무적 기여를 한 것뿐만 아니라, 단순히 하나의 요소에 대한 가격타력성이 아닌 여러 요소를 조합하여 복합적인 수요의 가격탄력성을 제시하고 이를 통해 최적의 탄력구간을 도출하였다는 점에서 학문적인 기여 또한 상당하다고 할 수 있다.

Electronic Commerce Research

In recent years, fashion e-commerce has become more and more popular. Since there are so many fas... more In recent years, fashion e-commerce has become more and more popular. Since there are so many fashion products provided by e-commerce retailers, it is necessary to provide recommendation services to users to minimize information overload. When users look for a product on an e-commerce website, they usually click the product information sequentially. Previous recommenders, such as content-based recommenders and collaborative filtering recommenders, do not consider this important behavioral characteristic. To take advantage of this important characteristic, this study proposes sequence-aware recommenders for fashion product recommendation using a gated recurrent unit (GRU) algorithm. We conducted an experiment using a dataset collected from an e-commerce website of a Korean fashion company. Experimental results show that sequence aware recommenders outperform non-sequence aware recommender, and multiple sequence-based recommenders outperform a single sequence-based recommender because...

Global E-Business Association, Jun 1, 2017

Global E-Business Association, Aug 1, 2017

The Korea Journal of BigData, 2018

The e-Business Studies, 2017

International Area Studies Review, 2017

In this paper, we report an industry-academia collaborative study on the distribution method of f... more In this paper, we report an industry-academia collaborative study on the distribution method of fashion products using an artificial intelligence (AI) technique combined with an optimization method. To meet the current fashion trend of short product lifetimes and an increasing variety of styles, the company produces limited volumes of a large variety of styles. However, due to the limited volume of each style, some styles may not be distributed to some off-line stores. As a result, this high-variety, low-volume strategy presents another challenge to distribution managers. We collaborated with KOLON F/C, one of the largest fashion business units in South Korea, to develop models and an algorithm to optimally distribute the products to the stores based on the visual images of the products. The team developed a deep learning model that effectively represents the styles of clothes based on their visual image. Moreover, the team created an optimization model that effectively determines t...

The e-Business Studies, 2017

The e-Business Studies, 2017

The e-Business Studies, 2020

We study how the nature of a hybrid system (perfect fluid, solid or a mixture of them) could be r... more We study how the nature of a hybrid system (perfect fluid, solid or a mixture of them) could be related to the induction of general relativistic surface degrees of freedom on phase-splitting surfaces upon perturbation of its phases. We work in the scope of phase conversions in the vicinity of sharp phase transition surfaces whose timescales are either much smaller (rapid conversions) or larger (slow conversions) than the ones of the perturbations (ω −1 , where ω is a characteristic frequency of oscillation of the star). In this first approach, perturbations are assumed to be purely radial. We show that surface degrees of freedom could emerge when either the core or the crust of a hybrid star is solid and phase conversions close to a phase-splitting surface are rapid. We also show how this would change the usual stability rule for solid hybrid stars, namely ∂M 0 /∂ρ c ≥ 0, where M 0 is the total mass to the background hybrid star and ρ c its central density. Further consequences of our analysis for asteroseismology are also briefly discussed.

International Journal of Distributed Sensor Networks, 2018

Connected cars, which are vehicles connected to wireless networks through the convergence of auto... more Connected cars, which are vehicles connected to wireless networks through the convergence of automotive and information technologies, have become an important topic of academic and industrial research on automobiles. In this research, we conducted a field experiment to understand vehicle maintenance mechanisms of a connected car platform. Specifically, we investigated the feasibility of prognostics and health management under different driving circumstances, with varying vehicle models, vehicle conditions, drivers’ propensity for speeding, and road conditions. We collected sensor data through a two-stage model of vehicle communication using an on-board diagnostics scanner and data transmission using wireless communication. We found that device defects can be predicted based on driving situations such as the driving mode, mechanical characteristics, and a driver’s speeding propensity.

Sustainability, 2019

Retailers need accurate movement pattern analysis of human-tracking data to maximize the space pe... more Retailers need accurate movement pattern analysis of human-tracking data to maximize the space performance of their stores and to improve the sustainability of their business. However, researchers struggle to precisely measure customers’ movement patterns and their relationships with sales. In this research, we adopt indoor positioning technology, including wireless sensor devices and fingerprinting techniques, to track customers’ movement patterns in a fashion retail store over four months. Specifically, we conducted three field experiments in three different timeframes. In each experiment, we rearranged one element of the visual merchandising display (VMD) to track and compare customer movement patterns before and after the rearrangement. For the analysis, we connected customers’ discrete location data to identify meaningful patterns in customers’ movements. We also used customers’ location and time information to match identified movement pattern data with sales data. After class...

The Society of Fashion and Textile Industry, Aug 1, 2017

ICT 기술의 발전에 따라 제조 산업은 공정 상에서 생성되는 제조 데이터를 분석하여 효율을 높이고자 많은 노력을 하고 있다. 본 논문에서는 스마트 공장의 일환으로 의사결정나무... more ICT 기술의 발전에 따라 제조 산업은 공정 상에서 생성되는 제조 데이터를 분석하여 효율을 높이고자 많은 노력을 하고 있다. 본 논문에서는 스마트 공장의 일환으로 의사결정나무 알고리즘(CHAID)을 이용한 데이터 마이닝 기반 제조공정을 제안한다. 약 5개월간 수집된 실제 제조 공정의 432개 센서 데이터를 활용하여 불량률이 낮은 안정적인 공정 기간과 불량률이 높은 불안정한 공정 기간 간에 유의미한 차이를 보이는 변수를 찾아냈다. 선정된 최종 변수가 불량률 개선에 실제로 효과가 있는지를 측정하기 위해 해당 변수의 안정 값 범위를 설정하여 14일 간 공정에서 해당 센서가 안정 값의 범위를 벗어나지 않도록 공정 설정 값을 조절했고, 불량률 개선의 효과를 측정하였다. 이를 통해 제조 산업에서 생성되는 공정 센서 데이터를 활용 및 분석하여 불량률을 개선할 수 있는 실증적인 가이드라인을 제시할 수 있을 것으로 기대한다.

This study investigated the effects of influencing factors on the sales volume of apparel product... more This study investigated the effects of influencing factors on the sales volume of apparel products. Based on previous studies, weekend effect, discount rate, and meteorological factors including daily average temperature, rainfall, sea level pressure, and fine dust were selected as independent variables to calculate their effects on sales quantity of apparel products. The daily sales data during 2015 - 2016 were collected from casual brands and outdoor brands which “A” apparel manufacturing company had operated. The actual data of “A” company were analyzed using SAS® 9.4 and SAS® Enterprise Miner 14.1. The results of this study were as follows: First, the influencing factors on total sales volume of apparel products were proved to be the weekend effect, discount rate, and fine dust. Second, the analysis of influencing factors on sales volume of apparel products according to season showed: 1) In casual brands, the average temperature had a significant influence on the sales volume of...

본 연구는 가격 할인 구간별 고객 수요의 민감도, 즉 탄력성을 통해 프로모션 기간 중 어느 시기, 어떤 가격대의 상품에 얼마만큼의 할인율을 적용하는 것이 수익을 극대화하는 가... more 본 연구는 가격 할인 구간별 고객 수요의 민감도, 즉 탄력성을 통해 프로모션 기간 중 어느 시기, 어떤 가격대의 상품에 얼마만큼의 할인율을 적용하는 것이 수익을 극대화하는 가장 효율적인 방안인지를 실증적으로 분석하였다. 이를 위해 본 연구는 한국의 패션 A기업에 대한 4년간의 상품 가격 및 판매량 데이터를 바탕으로 시즌 구간을 설정하고 동일한 조건의 가격탄력성을 구한 후 그에 따른 최적 할인율 구간을 도출하였다. 본 연구의 결론으로 한국의 경우, 20% 할인율에서 그리고 시즌 초기보다는 중반기 이후에 주로 민감하게 반응하여 소위 구매 여력이 되는 소비자들이 먼저 구매를 하고, 시즌 후반기로 갈수록 가격 할인을 통해 프로모션을 진행하는 것이 효과적이라는 결론을 내릴 수 있는 반면, 중국은 할인율이 30% 이상, 그것도 시즌 초기 할인에 민감하게 반응한다는 것을 알아냈다. 이는 초기 브랜드에 민감하게 반응하는 중국의 집단주의적 성격과도 매우 관련이 있으며, 특히 단일 요소 분석에서 드러난 것과 같이 고가의 상품에 대해 민감하게 반응하는 중국인의 특성상 명품에 대한 선호와도 괘를 같이 하는 것이라 짐작할 수 있다. 본 연구는 4년간에 걸친 다국적 기업의 대용량 데이터를 바탕으로 실증분석을 시행함으로써 다양한 정책에 활용할 수 있는 실무적 기여를 한 것뿐만 아니라, 단순히 하나의 요소에 대한 가격타력성이 아닌 여러 요소를 조합하여 복합적인 수요의 가격탄력성을 제시하고 이를 통해 최적의 탄력구간을 도출하였다는 점에서 학문적인 기여 또한 상당하다고 할 수 있다.

Electronic Commerce Research

In recent years, fashion e-commerce has become more and more popular. Since there are so many fas... more In recent years, fashion e-commerce has become more and more popular. Since there are so many fashion products provided by e-commerce retailers, it is necessary to provide recommendation services to users to minimize information overload. When users look for a product on an e-commerce website, they usually click the product information sequentially. Previous recommenders, such as content-based recommenders and collaborative filtering recommenders, do not consider this important behavioral characteristic. To take advantage of this important characteristic, this study proposes sequence-aware recommenders for fashion product recommendation using a gated recurrent unit (GRU) algorithm. We conducted an experiment using a dataset collected from an e-commerce website of a Korean fashion company. Experimental results show that sequence aware recommenders outperform non-sequence aware recommender, and multiple sequence-based recommenders outperform a single sequence-based recommender because...

Global E-Business Association, Jun 1, 2017

Global E-Business Association, Aug 1, 2017

The Korea Journal of BigData, 2018

The e-Business Studies, 2017

International Area Studies Review, 2017

In this paper, we report an industry-academia collaborative study on the distribution method of f... more In this paper, we report an industry-academia collaborative study on the distribution method of fashion products using an artificial intelligence (AI) technique combined with an optimization method. To meet the current fashion trend of short product lifetimes and an increasing variety of styles, the company produces limited volumes of a large variety of styles. However, due to the limited volume of each style, some styles may not be distributed to some off-line stores. As a result, this high-variety, low-volume strategy presents another challenge to distribution managers. We collaborated with KOLON F/C, one of the largest fashion business units in South Korea, to develop models and an algorithm to optimally distribute the products to the stores based on the visual images of the products. The team developed a deep learning model that effectively represents the styles of clothes based on their visual image. Moreover, the team created an optimization model that effectively determines t...

The e-Business Studies, 2017

The e-Business Studies, 2017

The e-Business Studies, 2020

We study how the nature of a hybrid system (perfect fluid, solid or a mixture of them) could be r... more We study how the nature of a hybrid system (perfect fluid, solid or a mixture of them) could be related to the induction of general relativistic surface degrees of freedom on phase-splitting surfaces upon perturbation of its phases. We work in the scope of phase conversions in the vicinity of sharp phase transition surfaces whose timescales are either much smaller (rapid conversions) or larger (slow conversions) than the ones of the perturbations (ω −1 , where ω is a characteristic frequency of oscillation of the star). In this first approach, perturbations are assumed to be purely radial. We show that surface degrees of freedom could emerge when either the core or the crust of a hybrid star is solid and phase conversions close to a phase-splitting surface are rapid. We also show how this would change the usual stability rule for solid hybrid stars, namely ∂M 0 /∂ρ c ≥ 0, where M 0 is the total mass to the background hybrid star and ρ c its central density. Further consequences of our analysis for asteroseismology are also briefly discussed.

International Journal of Distributed Sensor Networks, 2018

Connected cars, which are vehicles connected to wireless networks through the convergence of auto... more Connected cars, which are vehicles connected to wireless networks through the convergence of automotive and information technologies, have become an important topic of academic and industrial research on automobiles. In this research, we conducted a field experiment to understand vehicle maintenance mechanisms of a connected car platform. Specifically, we investigated the feasibility of prognostics and health management under different driving circumstances, with varying vehicle models, vehicle conditions, drivers’ propensity for speeding, and road conditions. We collected sensor data through a two-stage model of vehicle communication using an on-board diagnostics scanner and data transmission using wireless communication. We found that device defects can be predicted based on driving situations such as the driving mode, mechanical characteristics, and a driver’s speeding propensity.

Sustainability, 2019

Retailers need accurate movement pattern analysis of human-tracking data to maximize the space pe... more Retailers need accurate movement pattern analysis of human-tracking data to maximize the space performance of their stores and to improve the sustainability of their business. However, researchers struggle to precisely measure customers’ movement patterns and their relationships with sales. In this research, we adopt indoor positioning technology, including wireless sensor devices and fingerprinting techniques, to track customers’ movement patterns in a fashion retail store over four months. Specifically, we conducted three field experiments in three different timeframes. In each experiment, we rearranged one element of the visual merchandising display (VMD) to track and compare customer movement patterns before and after the rearrangement. For the analysis, we connected customers’ discrete location data to identify meaningful patterns in customers’ movements. We also used customers’ location and time information to match identified movement pattern data with sales data. After class...