주문생산 기업을 위한 기계학습 기반 총생산시간 예측 기법

박도명(Do-Myung Park),최형림(HyungRim Choi),박병권(Byung-Kwon Park) 한국지능정보시스템학회 2021 지능정보연구 Vol.27 No.1

Due to the development of the fourth industrial revolution technology, efforts are being made to improve areas that humans cannot handle by utilizing artificial intelligence techniques such as machine learning. Although on-demand production companies also want to reduce corporate risks such as delays in delivery by predicting total production time for orders, they are having difficulty predicting this because the total production time is all different for each order. The Theory of Constraints (TOC) theory was developed to find the least efficient areas to increase order throughput and reduce order total cost, but failed to provide a forecast of total production time. Order production varies from order to order due to various customer needs, so the total production time of individual orders can be measured postmortem, but it is difficult to predict in advance. The total measured production time of existing orders is also different, which has limitations that cannot be used as standard time. As a result, experienced managers rely on persimmons rather than on the use of the system, while inexperienced managers use simple management indicators (e.g., 60 days total production time for raw materials, 90 days total production time for steel plates, etc.). Too fast work instructions based on imperfections or indicators cause congestion, which leads to productivity degradation, and too late leads to increased production costs or failure to meet delivery dates due to emergency processing. Failure to meet the deadline will result in compensation for delayed compensation or adversely affect business and collection sectors. In this study, to address these problems, an entity that operates an order production system seeks to find a machine learning model that estimates the total production time of new orders. It uses orders, production, and process performance for materials used for machine learning. We compared and analyzed OLS, GLM Gamma, Extra Trees, and Random Forest algorithms as the best algorithms for estimating total production time and present the results.

디스플레이산업에서 수주생산방식의 개선 및효율화 제고 방안

조명호(Myong Ho Cho),조진형(Jin Hyung Cho) 한국산업경영시스템학회 2016 한국산업경영시스템학회지 Vol.39 No.4

MTO (Make to Order) is a manufacturing process in which manufacturing starts only after a customer’s order is received. Manufacturing after receiving customer’s orders means to start a pull-type supply chain operation because manufacturing is performed when demand is confirmed, i.e. being pulled by demand (The opposite business model is to manufacture products for stock MTS (Make to Stock), which is push-type production). There are also BTO (Build to Order) and ATO (Assemble To Order) in which assembly starts according to demand. Lean manufacturing by MTO is very efficient system. Nevertheless, the process industry, generally, which has a high fixed cost burden due to large-scale investment is suitable for mass production of small pieces or ‘mass customization’ defined recently. The process industry produces large quantities at one time because of the lack of manufacturing flexibility due to long time for model change or job change, and high loss during line-down (shutdown). As a result, it has a lot of inventory and costs are increased. In order to reduce the cost due to the characteristics of the process industry, which has a high fixed cost per hour, it operates a stock production system in which it is made and sold regardless of the order of the customer. Therefore, in a business environment where the external environment changes greatly, the inventory is not sold and it becomes obsolete. As a result, the company’s costs increase, profits fall, and it make more difficult to survive in the competition. Based on the customer’s order, we have built a new method for order system to meet the characteristics of the process industry by producing it as a high-profitable model. The design elements are designed by deriving the functions to satisfy the Y by collecting the internal and external VOC (voice of customer), and the design elements are verified through the conversion function. And the Y is satisfied through the pilot test verified and supplemented. By operating this make to order system, we have reduced bad inventories, lowered costs, and improved lead time in terms of delivery competitiveness. Make to order system in the process industry is effective for the display glass industry, for example, B and C groups which are non-flagship models, have confirmed that the line is down when there is no order, and A group which is flagship model, have confirmed stock production when there is no order.

공간 정리가 소비자의 신제품 수용 의도에 미치는 영향: 제품의 새로움의 조절 효과

Xiaobin Zhou,주재우 한국유통과학회 2019 유통과학연구 Vol.17 No.2

Purpose - People have long thought that disorderly space would influence their lives negatively. However, disorderly space may have positive impact such as increasing sales for new products for store managers. Borrowing the prior findings that disorderly space enhances creativity, we developed two hypotheses regarding space order, product newness, and new product adoption. We hypothesize that space order negatively influences new product adoption and that the proposed negative effect of space order on new product adoption is moderated by product newness. Research design, data, and methodology - We conducted a pre-test in China using 70 undergraduate students to test whether two newly developed cosmetic products were appropriate for the experimental stimuli. We confirmed that a fragrance patch concept, which is unavailable in the market, is a RNP (Really New Product), and an edible fragrance, which is available in the market, is an INP (Incrementally New Product). Next, we conducted a main experiment with a 2 (Space order: orderly vs. disorderly) x 2 (Product newness: RNP vs. INP) between-subjects design in China using 100 undergraduate students. Half of the participants answered questions on the disorderly desk and the other half answered questions on the orderly desk. Results - We obtained evidence from the main experiment that consumers’ new product adoption is the function of space order and product newness. First, participants were more likely to adopt the RNP when they were in the disorderly space than in the orderly space. However, the effect of space order on new product adoption disappeared when the product was INP; the adoption rates for the INP did not differ between the participants who answered questions on the disorderly desk and the participants who did so on the orderly desk. Conclusions - Our findings are counter intuitive. Contrary to the conventional wisdom that neat and tidy space benefits store managers, the research demonstrated that disorderly space increases sales, providing fresh insights into store managers to manage their stores and shops to sell new products. In order to help store visitors understand the value of really new products, store managers will have to sacrifice space order to some extent.

빅데이터 분석을 통한 소량다품종 주문생산 납기 개선

이재식(Jae-Sik Yi),김지현(Ji-Hyeon Kim),홍윤식(Youn-Sik Hong) 한국산학기술학회 2023 한국산학기술학회논문지 Vol.24 No.5

제조업은 양산과 주문생산 체계의 두 유형으로 구분할 수 있다. 이들 중 주문생산은 소량 다품종 생산체계가 특징이다. 고객의 요구 기한 내에 납품하기 위해서는 수주 행위, 생산계획 및 공급사슬로부터 부품 공급까지를 광범위하게 다루는 ATP 시스템이 중요하다. 최근 ATP 시스템은 효과적인 계획 수립을 위해 빅데이터에 근거한 기계학습 모델을 이용하는 것으로 발전하였다. 소량 다품종 주문생산 유형의 대표적 사례로 국내 A 중소기업을 분석하였다. A사는 고객사로부터 수주한 다양한 튜브 제품을 생산하여 납품한다. A사의 경우, 월평균 주문 261건에 납기 미준수 116건으로 고객이 요구한 납품 기한을 넘긴 지연 납품 비율이 45%로 높게 나타났다. A사의 원시 데이터에 생산성 관련 속성이 빠져 있어, 본 논문에서는 상위 수준의 계획생산과 하위 수준의 생산기한을 도입해 A사의 납기 준수율을 높일 수 있는 개선 방안을 제시하였다. 원시 데이터를 전처리한 후에 빅데이터 분석을 통해 납기 미준수율과 납기 미준수 시 지연일수를 분석하였다. 이를 토대로, 지연 납품 문제를 해결할 수 있는 계획생산, 생산기한, 가주문 생산 및 주문변경 제한도입의 대안을 제시하였다. 끝으로, 이러한 제안이 A사의 지연 납품 문제를 해결하여 고객사의 불만을 해소할 수 있을 것으로 생각한다. Manufacturing can be divided into two types: mass production and make-to-order production. Between them, the latter can be characterized as a low-volume, high-variety (LVHV) production system. In order to deliver within the customers deadline, an available-to-promise (ATP) system that covers a wide range of orders, production planning, and parts supply chains is important. Recently evolving are machine learning models based on big data for effective production planning. As a representative case of LVHV make-to-order production, we analyze Company A, a domestic small enterprise that produces and delivers various tube products. For Company As 261 orders per month, on average, there are 116 cases of non-compliance with delivery deadlines. That means the rate of delayed deliveries was high at 45%. Since the raw data of the Company A do not contain productivity-related attributes, this paper proposes an improvement plan that can increase the compliance rate by introducing higher-level planned production and lower-level production deadlines. After pre-processing the raw data, we analyze the rate of non-compliance and the number of days past the delivery deadline. Based on this big data analysis, we propose alternatives for planned production, production deadlines, temporary order production, and limits on order changes to solve the late-delivery problem. Such proposals can solve complaints from customers by solving delayed delivery.

현행 문서제출명령 제도 보완에 관한 소고

권오상 충북대학교 법학연구소 2023 法學硏究 Vol.34 No.2

Although the current document production order system has been evaluated as a generalized obligation to produce documents, it has limitations due to the fact that it recognizes a wide range of exceptions and lacks strong and effective sanctions for non-compliance with document production orders. Article 344(2) of the Civil Procedure Act stipulates that general documents are subject to the obligation to produce, but excludes documents possessed by governmental entity. If the subject of a document production order is a third party, the penalty is a fine of not more than 5 million won, and if the subject of the order is a party, the court may recognize the other party's claim that the document is true. In order to overcome the limitations of the document production order system and improve its effectiveness, it is necessary to expand the material and personal scope of the obligation to produce, i.e., documents possessed by governmental entity should be included in the same way as general documents, and information stored in electronic form should be expanded to be submitted. Sanctions for production orders should also be strengthened. Where documents are under the control of the other party, such as in modern litigation, it is reasonable to assume that a party's failure to comply with a production order is a prima facie case. In addition, the discretionary sanction method should be changed to a mandatory sanction method, and the means of sanction should be diversified, such as increasing the penalty. In addition, it is desirable to change the procedure for appealing an order to produce documents to an appeal procedure at the appropriate level instead of an immediate appeal.

제품조달시간 제약을 가지는 주문에 대한 생산 및 배송 일정계획

하귀룡(Gui Ryong Ha),박현재(Hyun Jai Park),전재호(Jaeho Juhn),윤상흠(Sang Hum Yoon) 한국산업경영학회 2008 경영연구 Vol.23 No.1

본 연구에서는 주문생산방식(make-to-order)에서 고객들로부터 접수된 주문(order)들을 모두 처리하는데 소요되는 최종 완료시간을 최소화할 수 있는 주문간 생산 및 배송 순서를 결정하기 위한 주문일정계획모델(order scheduling model)을 고려하고 있다. 각 주문은 자체생산품과 외부조달품으로 구성되고, 자체생산품에 대해서는 독자적인 생산일정계획이 필요하지만, 외부조달품의 경우에는 외부에서 조달되며, 정해진 조달완료시점에 도착한다. 또한, 각 주문에 속한 제품들이 모두 준비된 이후에는 고객으로의 배송이 이루어지게 된다. 본 연구에서는 이러한 운영상의 구조를 생산 및 조달단계와 그 이후의 배송단계로 연결되는 2단계 모델로 구성하고 배송단계에서의 배송시간과 배송차량의 가용성에 따라 3가지 변형모델로 구분하여 각 모델에 대한 경영과학적 해법을 제공하고자 한다. 각 모델에서 해에 대한 우월성질들을 개발하고 최종 완료시간을 최소화 할 수 있는 최적 일정계획법을 제안한다. 특히, 마지막 모델에 대해서는 근사해를 제공하는 3가지 발견적 해법(heuristic)을 제안하고, 각 해법의 성능을 여러 가지 수치데이터에 대한 실험을 통해 검증하였다. This paper considers a coordinated production and delivery scheduling problem to minimize the maximum completion time of a finite number of orders received from customers. Each order consists of two types of products. One type is manufactured by an in-house facility, but the other type is outsourced from the outside facility(or, subcontractor) so that the available time of the outsourced product is given as positive lead time. After all the products belonging to each order are prepared, the order is delivered to customer. The completion time of each order is measured by the time when delivery of the order is completed. The objective of this paper is to find out optimal production and delivery schedule minimizing the maximum completion time of all orders. We suggest three different models according that whether the delivery times to each customer are different or not, and that whether the capacity of delivery vehicle is finite or not. In the problem analysis, we first characterize some solution properties and then derive simple methods to find out the optimal schedules for the model 1 and 2. For the last model 3, a branch and bound solution is suggested and efficient heuristic algorithms to find approximate solutions are also developed. The performances of the proposed heuristics are evaluated through numerical experiments.

PFDL: A Production Flow Description Language for an Order-Controlled Production

Peter Detzner,Andreas Ebner,Maximilian Horstrup,Soren Kerner 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11

In Industry 4.0, the order-controlled production (OCP) is an order-centric manufacturing approach. Among others, OCP favors a dynamic (re-)assignment of production resources that are required to execute the order. In this paper, we propose a uniform description format of production processes - the Production Flow Description Language (PFDL - a fully automated processable input for OCP environments. The PFDL enables flexible use of existing production resources by taking advantage of the increased factory floor connectivity. Compared to existing solutions, the PFDL supports concurrency, data handling between processes, synchronization, and flow control.

감사인의 감사노력은 수주산업의 이익조정을 억제하는가?

배홍기 ( Hong Gi Bae ),양동훈 ( Dong Hoon Yang ),최준혁 ( Jun Hyeok Choi ) 한국회계학회 2018 회계저널 Vol.27 No.5

수주산업의 진행기준 수익 인식은 추정과 판단이 개입됨으로써 불확실한 미래환경변화 및 경영자의 자의적 판단으로 인한 회계품질이 문제가 되고 있다. 잘못된 추정은 향후 손실 인식으로 이어져 회계절벽의 원인으로 작용하고, 수주산업의 회계분식 스캔들은 규제기관의 각종 대책과 입법으로 이어져 최근 회계의 가장 큰 이슈로 떠오르고 있다. 수주산업 경영자의 추정과 판단에 기초한 수익인식은 발주처가 인정하는 부분과 차이가 나타나게 되고 그 차이만큼 미청구공사에 누적된다. 미청구공사는 수주산업 경영자의 이익조정을 반영하게 되고 따라서 감사인의 효과적인 감사노력은 미청구공사 계정에 영향을 미칠 것이다. 본 연구에서는 수주산업에서 감사인이 경영자의 추정과 판단의 적절성을 잘 파악하여 효과적인 감사를 해왔는지를 직급별감사시간이 미청구공사 수준에 미치는 영향을 검증함으로써 확인해 보았다. 미청구공사는 산업적 특성에서 자연스럽게 발생하는 부분과 경영자의 자의적/낙관적 추정과 판단으로 인한 부분이 혼재되어 있으므로, 정상미청구공사와 비정상미청구공사를 구분할 필요성이 제기된다. 비정상미청구공사 추정에는 강호영 등(2011)의 건설업 비용 결정요인들을 활용하였다. 감사인 감사노력의 대용치로는 직급별감사시간 공시자료를 활용하였으며 하위직급 감사인의 낮은 감사효과(배홍기 등 2017)를 감안하여 파트너와 등록공인회계사의 감사시간에 논의를 한정하였다. 연구 결과는 다음과 같다. 파트너와 등록공인회계사의 감사시간은 미청구공사를 억제하는 효과가 나타나지 않았다. 전체산업, 건설조선업, 비건설조선업의 산업별 구분과 미청구공사, 비정상미청구공사, 정상미청구공사의 어떤 조합에서도 유의한 음(-)의 효과는 관찰되지 않았다. 미청구공사를 수준변수 대신 전년대비 변화율 변수로 사용해도 동일했다. 다만 극단적으로 미청구공사 수준이 높은 상황에서 감사시간은 비정상미청구공사에 대해 유의한 음(-)의 관계를 보였으며, 이 결과는 전체 표본과 건설조선업 표본에서만 나타났다. 따라서 감사인은 미청구공사 수준이 위험하게 높을 경우에만 경영자의 경영판단을 엄격히 판단하는 경향이 나타난다고 볼 수 있다. 본 연구는 수주산업의 자의적인 수익 인식이 회계스캔들로 이어지는 상황에서 감사인이 수익 인식에 대해 엄격한 감사를 하고 있지 못해왔다는 것을 확인한 연구로서, 그 원인으로 는 근본적으로 수익인식이 추정과 판단에 기초하지만 내부정보에 의거한 경영판단이라는 이유로 외부감사인의 감사에 일정한 한계가 있기 때문일 수도 있다. 이와 같은 수주산업의 근본적 한계를 극복하기 위해서는 최근 감독당국의 전반적인 제도 강화 노력에 추가하여 수주산업만의 표준감사시간을 통해 효과적인 감사인력의 집중 투입을 유도하고, 전문가 활용을 확대하며, 수주산업에 특화된 산업전문감사인을 양성하기 위해 파트너 의무교체제도에 예외를 규정하는 등 수주산업의 독특한 특성을 반영한 제도 개선이 요청된다. The production-to-order industries’ percentage-of-completion based revenue recognition relies on estimation of uncertain future events; therefore it is affected by managements' arbitrary judgments. Inaccurate estimates may lead to future losses, causing the “accounting cliffs” phenomenon and the earnings management scandals which triggered recent changes in regulations. The estimation-based revenue recognition of production-to-order industries crates recognition gab between companies and clients and the differences accumulate in the unbilled revenue account. The unbilled revenue account reflects managements’ optimistic bias in estimation, in order words the earnings management, and therefore an effective audit effort should reduce the level of the unbilled revenue. In this study, we tested whether the auditors in the production-to-order industry have properly evaluated the appropriateness of the managerial estimations and judgments, by testing the relationship between audit efforts and unbilled revenue accounts. Along with the management optimism and intention to increase reported profits, the account naturally occurs because of the contractual conditions and future economic changes. Therefore, it is critical to distinguish between normal and abnormal portions. The normal portion represents the natural accumulation of the account due to the contractual and economic situation; the abnormal part captures the management bias to report an increased profit. We used Kang et al. (2011)’s five cost determinants in the productionto- order industry and assumed that those five determinants will also determine the unbilled revenue because in the production-to-order industries revenue estimation is based on how much the company estimates its total and current costs. In the percentage-in-completion method the cost directly decides how much earnings the company will recognize for a period. As the proxy for audit efforts, we used the partner hours and registered CPA hours from 2014 to 2016. Because prior studies showed that lower-level auditors were less effective than high-level auditors in detecting earnings management, we limited our tests to the high-level auditors. Our empirical findings are as follows. We found that the partner and CPA total audit hours have no significant effects on the unbilled account. For any industry subgroups or dependent variables, we failed to discover any evidences that the audit efforts reduce the unbilled revenue account. However, we also found that only in the situation where the level of unbilled account is extremely high, audit efforts reduced the account, which means that only in an extreme case auditors do not allow the business judgement plea. Sensitivity tests using a discretionary accruals model, a 2SLS model, and the Lee (2017)’s abnormal unbilled account model confirmed our findings. Our findings suggests that under the current regulations there is certain limits for the external auditors to preventing the aggressive accounting practices in the production-to-order industries due to the systemic reasons. It is necessary to encourage more allocation of audit hours to high-level auditors who possess know-how on auditing or more use of industry professionals who understands the industries by regulating a new industry-specific standard rate system. To increase the number of industry specialized auditors who have accumulated industry knowledge at the level of company managers, it is required to allow an exception to the current mandatory partner or audit firm rotation rules.

Bidirectional effects of training order on perception and production accuracy of English L2 vowels

박현수,이주경 한국음운론학회 2022 음성·음운·형태론 연구 Vol.28 No.2

The present study investigates integrated L2 vowel training with different orders between production and perception and their effects on production and perception accuracy. In the experiment, two training groups and one control group participated. One of the training groups, composed of 10 Korean learners of English, received production training preceding perception training (Pro-Per) for the English L2 vowels /ɑ/, /ʌ/, and /ɔ/, and the other group received perception training preceding production training (Per-Pro). They participated in 9 sessions of training for 10 weeks, and both production and perception training were conducted within each session. They were evaluated for perception and production accuracy in three tests: pretest, posttest and generalization test. Results show that the combined training was found to be satisfactorily effective to improve perception ability irrespective of training order. On the other hand, Pro-Per training order was more effective to improve production accuracy than Per-Pro training. This might be attributed to the motor cortex activation occurring during perception learning, which serves as a function of confirming the earlier production learning. Production learning, which is usually delayed and requires constant exposure in comparison to perception learning, did not seem to be effectively substantiated by back-up learning through perception training.

영국 수사기관의 미결수용자 접견 및 호송에 관한 소고: 수용자 소환명령(Inmate Production Orders)을 중심으로

조윤오 동국대학교 사회과학연구원 2019 사회과학연구 Vol.26 No.1

This study aims at exploring the policy of inmate production and escort orders of UK to improve the correctional dysfunctions toward inmates who have not been convicted yet in the facilities in South Korea. Contrary to other developed countries’ system, the current investigation procedures toward inmates who have not get sentenced have different steps between the inmate investigation of police agency and that of prosecutor organization in South Korea. Inmates who are in the middle of waiting the trial are escorted and sent to prosecutors’ office mandatorily for investigation before their trials, while the inmates are interviewed by police in prison or jail facilities. As a result, criminals who have been incarcerated without conviction are more likely to lose equivalent power to prepare the upcoming trials. This study found out that “Inmate Production Order” policy of UK can have positive implications in terms of inmates’ legal rights by developing effective prison based interview processes between prosecutors and inmates who have not convicted in South Korea. This study result showed that investigation agencies are required to visit prison facilities and meet inmates in order to conduct investigation instead of requesting unnecessary inmate escort service in UK. Crime Sentence Act of 1997 allowed prison governors to decide whether or not the exceptional requests of inmates production orders can be implemented. If the prison governors consider the exceptional inmate production orders, the inmates who have not been convicted can be assigned and escorted to prosecutor office or police department for investigation. In this case, the formal and legal responsibility for inmate safety rely on the investigation agency which had requested the inmate production order to the representative of prison facilities. The further policy implication will be discussed for addressing the critical issues of inmates escort system and interview procedure with police and prosecutors in South Korea. 본 연구는 법적인 근거 없이 관행 차원에서 이루어지고 있는 우리나라 교정 당국의 검찰 수사기관에 대한 미결수용자 연출, 동행 호송 지원업무의 문제점을 검토하고, 영국에서 이 루어지고 있는 수사기관의 미결수용자 접견, 호송 실태를 살펴보는데 목적이 있다. 우리나 라는 현재 경찰 수사기관에게는 교정시설 내 접견 조사를 원칙으로 하고 있고, 검찰 수사기 관에는 검찰 시설로 직접 미결수용자를 호송시켜 사건 접견 조사를 받도록 하는 이원화된 구조를 운용하고 있다. 이에 미결수용자의 방어권 보장 및 인권 보장이 크게 우려되는 상황 인바, 최근의 문헌고찰을 통해 영국의 사례를 가지고 해외 교정에서 이루어지고 있는 수사 기관 지원 호송 업무 절차를 검토할 필요가 크다. 국내외 관련 문헌연구 고찰 결과, 수용자에 대한 수사기관의 면담은 영국의 경우 원칙적 으로 “구금시설 방문”을 기본으로 하는 것으로 나타났다. 즉, 영국과 같은 선진 교정 시스템 을 갖춘 나라에서도 미결수용자의 외부 호송(escort) 업무는 매우 예외적으로 이루어지고 있다고 볼 수 있다. 원칙적으로 교정기관 내 수사접견실로 경찰이나 검찰이 직접 방문하여 구금시설 내에서 수용자를 면담하는 것이 통상적이라고 하겠다. 예외적인 경우에만 수사기 관이 요청한 “수용자 소환명령(Production Orders of Prison Inmates)”신청을 교도소장이 허가하게 되고, 이 소환명령이 내려진 경우에만 수사기관이 미결수용자를 호송해서 교정시 설 밖으로 데리고 나갈 수 있다. 이 연구에서는 영국 교도소장의 미결수용자 소환명령을 상세히 살펴본 후, 우리나라 미결 수용자 보호 및 방어권 보장에 관한 정책 변화의 필요성 등을 검토해 보았다. 최근 우리나라 에서 심각한 문제로 대두되고 있는 교도소 과밀화 및 교정인력 부족 문제를 해결하기 위해서 영국과 같은 “미결수용자 소환명령” 절차가 어떻게 도입되어야 하는지 살펴보고, 경찰과 달 리 검찰 수사기관만이 관행적으로 사용하고 있는 현재의 미결수용자 외부 검찰청 호송 면담 방식을 어떻게 수정해야 가야 하는지를 구체적으로 탐색해 보는 기회를 가졌다고 하겠다.