최소 단어 이상 선택하여야 합니다.
최대 10 단어까지만 선택 가능합니다.
다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
NTIS 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
DataON 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Edison 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Kafe 바로가기주관연구기관 | 서울대학교 Seoul National University |
---|---|
연구책임자 | 권오상 |
참여연구자 | 박윤선 , 반경훈 , 이승호 , 이재요 , 이혜원 , 조현경 , 김현석 , 권오성 , 김세림 , 김기련 , 민지원 , 이원석 , 최하연 |
보고서유형 | 최종보고서 |
발행국가 | 대한민국 |
언어 | 한국어 |
발행년월 | 2021-02 |
과제시작연도 | 2020 |
주관부처 | 농촌진흥청 Rural Development Administration(RDA) |
등록번호 | TRKO202100010086 |
과제고유번호 | 1395064254 |
사업명 | 첨단기술융복합차세대스마트팜기술개발(R&D) |
DB 구축일자 | 2021-10-09 |
키워드 | 스마트팜.소비트렌드.빅데이터.농가 최적화.구조변화.Smart Farm.Big data.Farm optimization model.Consumption trend.Structural change. |
- 최근 스마트팜 증가에 따라 활용 가능한 생산, 소비 부문의 빅데이터를 통해 농가의사결정에 활용하고자 함
- 이에 생산과 소비를 고려한 스마트팜 최적화 및 시장모형 구축, 의사결정 예측 및 수익극대화 모형 개발, 과채류 소비 트렌드 분석 연구를 진행함
- 스마트팜 시설채소 주요품목인 토마토, 딸기, 파프리카 농가 단위 최적화 모형 및 시장모형 구축
- MDCEV, AIDS, QU-AIDS 모형을 이용한 과채류 수요 함수 추정
- 경영여건 변화에 따른 스마트팜 농가 의사결정 예측 및 수익극대화 모형 개발
- 최근 스마트팜 증가에 따라 활용 가능한 생산, 소비 부문의 빅데이터를 통해 농가의사결정에 활용하고자 함
- 이에 생산과 소비를 고려한 스마트팜 최적화 및 시장모형 구축, 의사결정 예측 및 수익극대화 모형 개발, 과채류 소비 트렌드 분석 연구를 진행함
- 스마트팜 시설채소 주요품목인 토마토, 딸기, 파프리카 농가 단위 최적화 모형 및 시장모형 구축
- MDCEV, AIDS, QU-AIDS 모형을 이용한 과채류 수요 함수 추정
- 경영여건 변화에 따른 스마트팜 농가 의사결정 예측 및 수익극대화 모형 개발
- 주요 과채류 소비 트렌드 변화 분석
- 빅데이터(비정형 데이터)를 이용한 주요 시설채소 소비 트렌드 분석
(출처 : 요약서 3p)
□ Purpose&Contents
Domestic agriculture has faced various difficulties: a decrease in the population of rural areas, an aging population, intensification of the opening market due to the expansion of FTA, stagnation in the income of farm households, and increasing uncertainty due to climate chang
□ Purpose&Contents
Domestic agriculture has faced various difficulties: a decrease in the population of rural areas, an aging population, intensification of the opening market due to the expansion of FTA, stagnation in the income of farm households, and increasing uncertainty due to climate change. Thus, smart farms' expectations are growing recently as a power source for increasing farm income and developing high-tech agricultural industries. However, the research developing optimized smart farm business models that reflect the market and environmental conditions is insufficient.
In particular, managing and collecting big data of cultivation and breeding to help decision making of smart farms have already been implemented by rural development administration. Also, microdata about consumption and unstructured data are being collected. Such preparation works to magnify the popularization of smart farms, and their management improvement has continuously proceeded. Therefore, the need to develop a model that can reflect these accessible data in smart farm operations has increased. Besides, since the investment and operation of smart farms that do not take into account market conditions or future price conditions can lead to deterioration in the management balance due to oversupply, the means considering macro variables outside of farms such as market and environmental variables in farmers' decision-making along with research on smart farm technology needs to be developed.
On that ground, this research comprehensively analyzes the consumption pattern of greenhouse vegetables in smart farms, prediction of supply according to the spread of smart farms, demand stabilization plans, and develops a smart farm profitability optimization model which simultaneously considers such production and consumption.
Figuring out consumers' behavior and consumption pattern of greenhouse vegetables from smart farms through the structured data of consumption, price and markets and unstructured data in Internet and SNS, and accurately grasping the impact on demand could contribute to the profit optimization supporting the smart farms' production control plan.
In addition to the micro decision-making model that passively reflects exogenous market information from individual farmers' point of view, this study builds a relatively macro decision-making model that reflects inverse effects of decision-making on farms changing supply and demand conditions market conditions such as price.
Also, it is necessary to analyze the impact of changes in planted area by the changes in investment of smart farms on crop prices to maintain stable income of smart farming farmers. In addition, even if the spread of smart farms does not decrease the market price of the crops, the investment costs of smart farming system would be a big burden for farmers. Hence it is necessary to analyze the profitability of adopting the smart farming system. If adopting smart farming system has economic feasibility, a potential adoption rate of smart farming system under various scenarios should be derived to decide more effective government policy in the near future. Finally, based on the results from the above research subjects, this study establishes an optimal strategy to maximize the income of smart farms.
□ Results
The ultimate business performance of smart farms comes from gaining stable and high profits, for which the farm's yield is an immediate and the most influential factor. Thus, a yield prediction model including environmental variables and farm-specific variables is established. Also, we conduct a survival analysis to estimate the first harvest time according to the initial environment setting. We develop the smart farm optimization model for tomato, strawberry, and paprika. The smart farms optimization model is consist of a objective function which maximize profit of smart farms, relation expressions for main variables, and constraint equations for environmental variables. We analyze the effect of price change and outdoor temperature and heating cost change to producer decision of smart farms using optimization model. The results show that smart farms producer can actively response to change of production environment.
We also develop a smart farm market model to analyze the effect of gross supply change by the smart farms production. The market model has a objective function which maximizes the sum of producer surplus and consumer surplus created in each item market. We analyze the effect of reducing production cost of smart farms and increasing smart farms share in facility vegetables production using the market model. The results show that the smart farm producers should prepare strategies for the change of market and price in long term. If the smart farms share of an item is already high, increasing the smart farms share of this item should make the share of ordinary facilities decrease. It is needed to consider the order of priority by items in supporting smart farms installation.
In the second assignment, the price flexibility coefficients of planted area for strawberry and tomato were estimated to determine whether strawberries and tomatoes were suitable for the smart farm diffusion policy. The price flexibility coefficients for strawberry and tomato was ,respectively, 1.865, 0.354. Despite the increase in production due to the increase in planted area, the prices of tomatoes and strawberries have been analyzed to increase, which means that two crops are suitable for smart farm diffusion policies. Since tomato and strawberry were analyzed as suitable crops for smart farm diffusion policies, we conduct the economic evaluation of a smart farm investment and potential conversion rate analysis. This study analyzed the economic evaluation with the net present value (NPV) method and estimated the adoption potential of the smart farm with the trade-off analysis, minimum data (TOA-MD) model. The results were as follows: The analysis of the net present value shows that the smart farm investment for the two crops are economically feasible. Finally, we have derived the income changes of smart farms by changing harvest starting point based on the estimated changes of market prices and production. The results show that smart farms of strawberries and tomatoes could earn the highest income if they delay the harvest about two weeks than start harvest as usual.
In the third assignment, we analyzed the change in consumption of smart farm crops using QUAIDS model of consumer panel data, and identified supplements or substitutes. Secondly, unstructured data of news containing information on weather factors, food-related diseases or adverse events, health and nutrition information were collected, web scraping, and preprocessing. and big data analysis was conducted to analyze trends related to consumption by vegetable grown in facilites.
Third, artificial neural network analysis and panel analysis were conducted using consumer purchase data of consumer panel data to compare consumer purchase prediction. Finally, we analyzed short-run and long-run between news articles, consumer purchases, and wholesale prices. So, it was confirmed that news article data could be helpful in making smart farm decisions.
In the last assignment, we analyzed consumption trend and domestic and foreign environment factors and developed the sustainable growth strategy about the main fruit vegetables. Second, we wrote the paper about the strategies for the development of the fruit vegetables industry using unstructured big data analysis. Thrid, our study is to look the change of issues and consumer’s perception about the main fruit vegetables using the unstructured big data analysis.
□ Expected Contribution
The smart farm-related farmhouse unit optimization model and a market model constructed in this research can be used to help understand the smart farm production sector, provide implications for the future expansion of smart farms and the impact on the agricultural industry from its operation, and also useful information for establishing agrarian policy.
In addition, a decision-making process that maximizes profit can be derived by using a consumption-related model to predict consumption patterns or market price, combining the information with smart farmproduction management technology.
However, the farmhouse unit optimization model was constructed based on smart farm production data at a minimal level compared to the collected data's size. This is because only the representative characteristics of smart farmhouse type information were used, in which the number of samples was large and complete data was well collected at the initial stage of data collection for smart farm production. Further research using newly acquired production data is needed as the smart farm production data are being investigated continuously, and the targets of the survey are extending. Even for a smart farm producing the same items, productivity may appear differently depending on various characteristics such as the type of cultivation facility, variety, production area, and cultivation extent. A more detailed model configuration taking this into account can be established in the future.
The research covering the farmhouse unit optimization model for tomato was published as a thesis, and the research handling the strawberry or paprika farmhouse unit optimization model or the market model is summarized to prepare for publication.
(출처 : SUMMARY 8p)
과제명(ProjectTitle) : | - |
---|---|
연구책임자(Manager) : | - |
과제기간(DetailSeriesProject) : | - |
총연구비 (DetailSeriesProject) : | - |
키워드(keyword) : | - |
과제수행기간(LeadAgency) : | - |
연구목표(Goal) : | - |
연구내용(Abstract) : | - |
기대효과(Effect) : | - |
Copyright KISTI. All Rights Reserved.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.