간헐적 수요예측을 위한 이항가중 지수평활 방법

하정훈 한국산업경영시스템학회 2018 한국산업경영시스템학회지 Vol.41 No.1

Intermittent demand is a demand with a pattern in which zero demands occur frequently and non-zero demands occur sporadically. This type of demand mainly appears in spare parts with very low demand. Croston’s method, which is an initiative intermittent demand forecasting method, estimates the average demand by separately estimating the size of non-zero demands and the interval between non-zero demands. Such smoothing type of forecasting methods can be suitable for mid-term or long-term demand forecasting because those provides the same demand forecasts during the forecasting horizon. However, the smoothing type of forecasting methods aims at short-term forecasting, so the estimated average forecast is a factor to decrease accuracy. In this paper, we propose a forecasting method to improve short-term accuracy by improving Croston’s method for intermittent demand forecasting. The proposed forecasting method estimates both the non-zero demand size and the zero demands’ interval separately, as in Croston’s method, but the forecast at a future period adjusted by binomial weight according to occurrence probability. This serves to improve the accuracy of short-term forecasts. In this paper, we first prove the unbiasedness of the proposed method as an important attribute in forecasting. The performance of the proposed method is compared with those of five existing forecasting methods via eight evaluation criteria. The simulation results show that the proposed forecasting method is superior to other methods in terms of all evaluation criteria in short-term forecasting regardless of average size and dispersion parameter of demands. However, the larger the average demand size and dispersion are, that is, the closer to continuous demand, the less the performance gap with other forecasting methods.

A Binomial Weighted Exponential Smoothing for Intermittent Demand Forecasting

Chunghun Ha(하정훈) 한국산업경영시스템학회 2018 한국산업경영시스템학회지 Vol.41 No.1

Intermittent demand is a demand with a pattern in which zero demands occur frequently and non-zero demands occur sporadically. This type of demand mainly appears in spare parts with very low demand. Croston’s method, which is an initiative intermittent demand forecasting method, estimates the average demand by separately estimating the size of non-zero demands and the interval between non-zero demands. Such smoothing type of forecasting methods can be suitable for mid-term or long-term demand forecast-ing because those provides the same demand forecasts during the forecasting horizon. However, the smoothing type of forecasting methods aims at short-term forecasting, so the estimated average forecast is a factor to decrease accuracy. In this paper, we propose a forecasting method to improve short-term accuracy by improving Croston’s method for intermittent demand forecasting. The proposed forecasting method estimates both the non-zero demand size and the zero demands’ interval separately, as in Croston’s method, but the forecast at a future period adjusted by binomial weight according to occurrence probability. This serves to improve the accuracy of short-term forecasts. In this paper, we first prove the unbiasedness of the proposed method as an important attribute in forecasting. The performance of the proposed method is compared with those of five existing forecasting methods via eight evaluation criteria. The simulation results show that the proposed forecasting method is superior to other methods in terms of all evaluation criteria in short-term forecasting regardless of average size and dispersion parameter of demands. However, the larger the average demand size and dispersion are, that is, the closer to continuous demand, the less the performance gap with other forecasting methods.

머신러닝과 시계열 기법 기반의 초단기 시간단위 수요예측방법론 개발 연구

민경창,하헌구 한국로지스틱스학회 2022 로지스틱스연구 Vol.30 No.3

Demand forecasting is an important field and it is safe to say that forecasting is a key component of economic activity. An accurate forecasting is the key to determining the competitiveness of all economic players. Forecasting an uncertain future is a difficult task and radical change in the external environment are adding to the difficulty of forecasting. Amid the increasing demand for accurate demand forecasting, the emergence of Big data, AI, ML, and DL following the development of computing power is becoming a major turning point in the demand forecasting field as well. In addition to the traditional forecasting methodologies, the use of dataming techniques is also rapidly increasing. And various efforts have been continued to improve the forecasting accuracy. In this paper, a hybrid forecasting methodology which is combined time series model and data mining technique and a multistage methodology are presented for short-term forecasting. Specifically, we developed a hybrid forecasting model that combines SARIMA(Seasonal Autoregressive Integrated Moving Average) and Random Forest, and a multistage methodology that utilizing the forecasting result of the upper-category as a variable in the forecasting process of the sub-category. In order to verify the methodologies presented in this paper, we use the rental data of ‘Seoul bike’(shared bicycle in Seoul) as verification data. As a result of the forecasting ‘Seoul bike’ demand for the next 7 days(every 3 hours) of rental point clusters, the average forecasting accuracy was 81.5%. It is high accuracy level considering that the forecasting unit was 3hours, forecasting horizon was next 56 steps, and the average accuracy by Random forest was 65%. In addition, it was confirmed that high accuracy was maintained steadily regardless of the time difference from the forecasting point unlike the characteristics of general demand forecasting, And the high accuracy level was confirmed as a forecasting model not only a 3 hours forecasting, but also daily(90.1%) and weekly(91.7%) forecasting. The research shows the forecasting methodologies of this paper is worth to use as a short-term forecasting model. And we confirmed that the methodologies are very useful to forecasting daily and weekly demand as well. It is expected that the methodologies proposed in this paper will be widely used as an accurate forecasting model in more diverse fields.

장래교통수요예측을 고려한 도로 유지관리 방안

김정민,최승현,도명식,한대석 한국도로학회 2016 한국도로학회논문집 Vol.18 No.3

PURPOSES : This study aims to examine the differences between the existing traffic demand forecasting method and the traffic demand forecasting method considering future regional development plans and new road construction and expansion plans using a four-step traffic demand forecast for a more objective and sophisticated national highway maintenance. This study ultimately aims to present future pavement deterioration and budget forecasting planning based on the examination. METHODS: This study used the latest data offered by the Korea Transport Data Base (KTDB) as the basic data for demand forecast. The analysis scope was set using the Daejeon Metropolitan City’s O/D and network data. This study used a traffic demand program called TransCad, and performed a traffic assignment by vehicle type through the application of a user equilibrium-based multi-class assignment technique. This study forecasted future traffic demand by verifying whether or not a realistic traffic pattern was expressed similarly by undertaking a calibration process. This study performed a life cycle cost analysis based on traffic using the forecasted future demand or existing past pattern, or by assuming the constant traffic demand. The maintenance criteria were decided according to equivalent single axle loads (ESAL). The maintenance period in the concerned section was calculated in this study. This study also computed the maintenance costs using a construction method by applying the maintenance criteria considering the ESAL. The road user costs were calculated by using the user cost calculation logic applied to the Korean Pavement Management System, which is the existing study outcome. RESULTS : This study ascertained that the increase and decrease of traffic occurred in the concerned section according to the future development plans. Furthermore, there were differences from demand forecasting that did not consider the development plans. Realistic and accurate demand forecasting supported an optimized decision making that efficiently assigns maintenance costs, and can be used as very important basic information for maintenance decision making. CONCLUSIONS : Therefore, decision making for a more efficient and sophisticated road management than the method assuming future traffic can be expected to be the same as the existing pattern or steady traffic demand. The reflection of a reliable forecasting of the future traffic demand to life cycle cost analysis (LCCA) can be a very vital factor because many studies are generally performed without considering the future traffic demand or with an analysis through setting a scenario upon LCCA within a pavement management system.

우리나라 기업들의 수요예측 실태에 관한 연구 -2011년과 2020년 설문조사 비교·분석-

김종배 ( Jong Bae Kim ),김유선 ( You Sun Kim ),박민영 ( Min Young Park ) 한국유통물류정책학회 2021 유통물류연구 Vol.8 No.4

최근 세계 경제의 불확실성이 고조되면서 기업들의 장단기 경영계획 수립 시 애로가 가중되고 있다. 기업 경영에서 공급망관리(SCM)의 모든 계획은 자사 제품이나 서비스에 대한 장래 수요예측에 기반을 둔다. 시장의 실제 수요보다 과다한 수요예측은 불필요한 재고 발생에 따른 비용을 유발하고, 반대로 시장의 실제 수요보다 낮은 수요예측은 결품 발생에 따른 고객서비스 수준의 저하로 나타난다. 따라서 정확한 수요예측은 기업의 경영효율을 높이는 데 중요한 역할을 담당한다. 그러나 우리나라에서는 기업들의 수요예측 실태에 대한 조사 자료가 거의 없는 실정이다. 본 연구는 우리나라 기업들의 수요예측 실태에 대한 조사를 하고, 2011년 조사한 결과와 비교분석을 통해 국내 기업들의 수요예측에 대한 인식과 수준의 변화 트랜드를 파악해 보고, 외국 기업과의 사례 비교를 통해 국내 기업들의 수요예측에 대한 인식과 수준에 대해 살펴보고자 한다. 조사대상을 제조, 유통, 물류 등 기업유형별로, 그리고 대기업과 중소기업 등 기업 규모별로 분류하고, 수요예측 적용 여부에 대해 설문을 하였다. 조사결과, 우리나라 기업들은 약 56% 정도가 수요예측을 실시하고 있으며, 물류기업이 유통 및 제조업보다 수요예측 중요성에 대한 인식이 상대적으로 낮은 것으로 조사되었다. 또한, 중소기업보다 대기업이 수요예측에 대한 인식이 높고, 예측 빈도는 월 단위에서 주 단위로 짧아지는 추세이며, 수요예측 방법으로는 정량적 기반에 정성적으로 수정하는 방법을 가장 많이 사용하고 있는 것으로 조사되었다. 예측정확도 측정 방법으로는 평균오차가 주로 사용되며, 수요예측을 하는 기업의 35%만이 예측정확도를 기업의 핵심성과지표(KPI)에 반영하여 관리하고 있는 것으로 조사되었다. Uncertainty in the global economy has increased recently, and difficulties in long-term and short-term business planning of companies is being weighted. All supply chain management (SCM) plans in enterprise management are based on forecasts of future demand for their products and services. Forecasts that are more than the actual demand in the market induce costs associated with unnecessary improvements, while forecasts that are lower than the actual demand in the market reduce customer service levels due to shortages. Therefore, accurate demand forecasting plays an important role in improving the management efficiency of a company. However, in South Korea, the reality is that there are few survey materials on the actual state of demand forecasts by companies. In this study, we conducted a survey on the actual situation of demand forecasts of Korean companies, and through comparative analysis with the 2011 survey, identify trends in the perception and level of demand forecasting of domestic companies. By comparing cases with foreign companies, we will examine the perception and level of demand forecasting of domestic companies. The subjects of the survey were classified by company type such as manufacturing, distribution, and logistics, and by company size such as large companies and small and medium-sized enterprises, a survey was conducted on demand forecasting. As a result of the survey, it was found that about 56% of Korean companies carry out demand forecasting, and logistics companies are relatively less aware of the importance of demand forecasting than distribution and manufacturing industries. In addition, large companies are more aware of demand forecasts than small and medium-sized enterprises, and the forecast frequency is shortening from monthly to weekly basis. As a method of measuring forecast accuracy ME (Mean Error) are mainly used, and it was found that only 35% of companies that forcast demand refleced the forecasting accuracy in the company’s key performance indicator (KPI) and managed them.

Forecasting Logistics Demand Using Unbiased GM (1,1) Model Optimized by AIWPSO Algorithm

Li-Yan Geng,Zhan-Fu Zhang 보안공학연구지원센터 2016 International Journal of Hybrid Information Techno Vol.9 No.10

Accurate forecast of logistics demand can provide scientific guidance for logistics planning and decision making. With the complexity and uncertainty characteristics in logistics demand, the forecasting of logistics demand shows comprehensive and complex. The forecasting precision of the traditional forecasting methods often are not satisfying. It is necessary to look for novel forecasting methods to enhance the forecasting precision of logistics demand. Integrating the unbiased GM (1,1) model (UGM (1,1)) into the adaptive inertia weight particle swarm optimization (AIWPSO) algorithm, this paper developed a novel model for forecasting logistics demand, called AIWPSO-UGM (1,1) model, in which the UGM (1,1) model was used to forecast logistics demand and the AIWPSO algorithm was adopted to optimize the grey parameters needed in UGM (1,1) model. Two examples were selected to prove the out-of-sample performance of the AIWPSO-UGM (1,1) model in forecasting logistics demand. The results imply that the proposed AIWPSO-UGM (1,1) model performs better in logistics demand forecasting compared to the GM (1,1) model optimized by AIWPSO algorithm (AIWPSO-GM (1,1)), UGM (1,1), and GM (1,1) models.

Forecasting Steel Consumption Based on the Industrial Configuration: Case Study on the Korean Steel Industry

이재영,송상화,허성오 한국유통경영학회 2009 유통경영학회지 Vol.12 No.4

The steel industry is a key national strategic industry that supplies materials to various other industries including machinery, shipbuilding, electronics, and construction it greatly affects the overall industry. Since the steel industry has been continuously advancing to newly developing markets such as India and China, it has been placed under an extremelycompetitive system not only in the world steel market but also in the domestic market. Due to the characteristics of the processing industry, the steel industry requires large-scale investmentsat opportune times to acquire global competitiveness. Therefore, accurate forecasting of domestic steel demand greatly affects success or failure in competition. In the past, steel demand was predicted based on the intensity of use and growth curve using the macroeconomic index. Although somewhat meaningful in forecasting the long-term directionality in steel demand changes, the growth curve requires serious efforts with respect to accurate demand forecasts. In particular, applying a uniform growth curve without explicitly considering the structural differences in the steel consuming-industries between countries may affect the accuracy of demand forecasts. In this study, a demand forecast methodology reflecting the structureof the steel consuming industry is presented; its accuracy is determined by analyzing it in comparison with the actual domestic steel consumption. The results of this analysis shows an extremely high accuracy when the model reflecting the structure of the consuming industry is used. The methodology presented in this study is expected to be applicable in forecasting not only domestic steel demand but also the steel demand in India, Vietnam, and China as Korea’s new steel markets. The steel industry is a key national strategic industry that supplies materials to various other industries including machinery, shipbuilding, electronics, and construction it greatly affects the overall industry. Since the steel industry has been continuously advancing to newly developing markets such as India and China, it has been placed under an extremelycompetitive system not only in the world steel market but also in the domestic market. Due to the characteristics of the processing industry, the steel industry requires large-scale investmentsat opportune times to acquire global competitiveness. Therefore, accurate forecasting of domestic steel demand greatly affects success or failure in competition. In the past, steel demand was predicted based on the intensity of use and growth curve using the macroeconomic index. Although somewhat meaningful in forecasting the long-term directionality in steel demand changes, the growth curve requires serious efforts with respect to accurate demand forecasts. In particular, applying a uniform growth curve without explicitly considering the structural differences in the steel consuming-industries between countries may affect the accuracy of demand forecasts. In this study, a demand forecast methodology reflecting the structureof the steel consuming industry is presented; its accuracy is determined by analyzing it in comparison with the actual domestic steel consumption. The results of this analysis shows an extremely high accuracy when the model reflecting the structure of the consuming industry is used. The methodology presented in this study is expected to be applicable in forecasting not only domestic steel demand but also the steel demand in India, Vietnam, and China as Korea’s new steel markets.

수요 특성이 계층적 수요예측법의 퍼포먼스에 미치는 영향:해군 수리부속 사례 연구

문성민 한국경영과학회 2012 經營 科學 Vol.29 No.1

The demand for naval spare parts is intermittent and erratic. This feature, referred to as non-normal demand, makes forecasting difficult. Hierarchical forecasting using an aggregated time series can be more reliable to predict non-normal demand than direct forecasting. In practice the performance of hierarchical forecasting is not always superior to direct forecasting. The relative performance of the alternative forecasting methods depends on the demand features. This paper analyses the influence of the demand features on the performance of the alternative forecasting methods that use hierarchical and direct forecasting. Among various demand features variability, kurtosis, skewness and equipment groups are shown to significantly influence on the performance of the alternative forecasting methods.

장래교통수요예측을 고려한 도로 유지관리 방안

김정민,최승현,도명식,한대석 한국도로학회 2016 한국도로학회논문집 Vol.18 No.3

PURPOSES : This study aims to examine the differences between the existing traffic demand forecasting method and the traffic demand forecasting method considering future regional development plans and new road construction and expansion plans using a four-step traffic demand forecast for a more objective and sophisticated national highway maintenance. This study ultimately aims to present future pavement deterioration and budget forecasting planning based on the examination. METHODS: This study used the latest data offered by the Korea Transport Data Base (KTDB) as the basic data for demand forecast. The analysis scope was set using the Daejeon Metropolitan City’s O/D and network data. This study used a traffic demand program called TransCad, and performed a traffic assignment by vehicle type through the application of a user equilibrium-based multi-class assignment technique. This study forecasted future traffic demand by verifying whether or not a realistic traffic pattern was expressed similarly by undertaking a calibration process. This study performed a life cycle cost analysis based on traffic using the forecasted future demand or existing past pattern, or by assuming the constant traffic demand. The maintenance criteria were decided according to equivalent single axle loads (ESAL). The maintenance period in the concerned section was calculated in this study. This study also computed the maintenance costs using a construction method by applying the maintenance criteria considering the ESAL. The road user costs were calculated by using the user cost calculation logic applied to the Korean Pavement Management System, which is the existing study outcome. RESULTS : This study ascertained that the increase and decrease of traffic occurred in the concerned section according to the future development plans. Furthermore, there were differences from demand forecasting that did not consider the development plans. Realistic and accurate demand forecasting supported an optimized decision making that efficiently assigns maintenance costs, and can be used as very important basic information for maintenance decision making. CONCLUSIONS : Therefore, decision making for a more efficient and sophisticated road management than the method assuming future traffic can be expected to be the same as the existing pattern or steady traffic demand. The reflection of a reliable forecasting of the future traffic demand to life cycle cost analysis (LCCA) can be a very vital factor because many studies are generally performed without considering the future traffic demand or with an analysis through setting a scenario upon LCCA within a pavement management system.

Development of an urban energy demand forecasting system to support environmentally friendly urban planning

Yeo, I.A.,Yoon, S.H.,Yee, J.J. Applied Science Publishers 2013 APPLIED ENERGY Vol.110 No.-

This study proposes a new urban energy demand forecasting system that includes the following improvements: (a) a facility planning information database (DB), (b) an energy and planning statistics DB, and (c) an enhancement of the accuracy of the energy calculation method. Each of these improved aspects is involved in energy demand forecasting for urban planning. The results from this study are as follows. (1).An Environment and energy Geographical Information System Database (E-GIS DB), which provides the mesh unit facility information, was utilized to allow for the forecasting and control of urban energy demands for each unit space. (2).An energy consumption unit figure was connected with an energy simulation to diversify the level of the urban energy consumption sector and the primary energy into hourly information. This figure allows for more accurate demand forecasting. (3).Urban facilities were categorized according to energy use characteristics and were modeled to allow for energy demand forecasts. (4).The energy demand was considered in an urban climate during summer with the characteristics of the heating methods that are suitable for domestic circumstances. Thus, a separate algorithm was suggested for a cooling period and a heating/intermission period to enhance the accuracy of the demand forecasts. (5).The performance of this energy demand forecasting system was validated, such that excessively high or low calculated values can be modified from the current method in a 'planned city' while the urban energy demand can be forecasted relatively correct and in detail with differences of a factor of 0.20-0.44 for the cooling period in the 'existing city'. (6).The proposed urban energy demand forecasting system was constructed as EnerISS Solver which is an automated module. This module drastically reduced time consuming for predicting energy demand and urban climate to respond for the urban energy planning subjects' needs immediately.