Font size:
Small
Medium
Large

Common Information Platform connects information from production to consumption

“Common Information Platform” (abbreviated as CIP) established by Taiwan Agricultural Research Institute (abbreviated as TARI) can introduce the “Digital Twin” technology into the vegetable and fruit greenhouse by collecting big data. This technology can improve the efficiency of greenhouse management and provide decision-making suggestions for the environmental control system. In addition, through the information integration of production and sales, this platform assists “Campus Food Ingredients Registration Platform” of Ministry of Education to obtain 3 Labels and 1 QR Code ingredient traceability data. These data can help managers to quickly trace the flow of ingredients and shorten the review time. The labels mentioned previous include CAS Taiwan Excellent Agricultural Products Label, Taiwan Organic Agricultural Products Label, Taiwan Agricultural and Food Traceability Label and Taiwan Agricultural Production Traceability QR Code (abbreviated as 3 Labels and 1 QR Code).

Due to the impact of extreme weather, such as heavy rain, drought, high or low temperature and other extreme weather, crops have suffered heavy losses, resulting in unstable food supply and fluctuations in vegetable prices. In addition, the aging population of farmers and the low willingness of young people to work in agriculture have made the shortage of agricultural workers increasingly serious. In the past few years, there have been many black-hearted food news, which has reduced the public's trust in the quality of products on the market and gradually paid more attention to food safety issues. If the information related to the production process is transparent, it can improve consumers' trust in the safety of agricultural products. In order to improve Taiwan's agricultural environment, it is imperative to transform from traditional agriculture to smart agriculture.

Taiwan Agricultural Research Institute (abbreviated as TARI) has established a “Common Information Platform” (abbreviated as CIP) in 2017. In addition to collecting agricultural, water, and livestock production and traceability data, this platform uses “Digital Twin” technology to analyze greenhouse environmental control data for providing managers past experience, real-time data, and how to control system feedback. This information enables managers to make appropriate judgments to achieve the effect of human-machine collaboration and decision-making optimization. Agricultural enterprises can not only save manpower, but also reduce energy consumption, achieve the purpose of energy saving, and establish innovative agricultural production models. This technology won the “R&D 100 Awards” in 2019, known as the Oscars of the technology industry.

The concept of “Digital Twin” was first proposed by Michael Grieves, a professor at the Florida Institute of Technology in the United States in 2002. It was initially used in industrial manufacturing. It means that each system consists of two modes, an original physical mode and another virtual mode with entity information. “Digital Twin” technology was jointly developed by TARI together with Institute for information industry (III) and was initially applied to agriculture. In order to solve the problems of experience inheritance and shortage of workers in agricultural technology, “Digital Twin” technology can introduce the Internet of Things and information and communication technology to assist in environmental control management and analysis, and upgrade the original environmental monitoring and control equipment management to an intelligent level.

The first agricultural “Digital Twin” technology demonstration farm is located in Yumei Biotechnology Co., Ltd., Dadu District, Taichung City. This technology upgrades greenhouse equipment and makes management intelligent, which can reduce product supply imbalances caused by human management and judgment. First, the greenhouse must be equipped with environmental sensing devices (temperature, humidity, luminosity, soil moisture) and actuation equipment (skylights, hoists, shade nets, fans, and drip-irrigation systems). These environmental control data provide “Digital Twin” technology for model building. The “Digital Twin” model is divided into two parts: “Greenhouse Doctor” and “Greenhouse Coach”. The Greenhouse Doctor (Figure 1) can diagnose environmental control data, including alerting equipment data abnormalities and providing operational behavior analysis. Its concept is similar to traditional Chinese medicine. “Observe, Listen, Ask, Analyze”. The greenhouse coach (Figure 2) models the administrator's operation behavior with artificial intelligence (AI), and provides decision-making suggestions to assist the administrator in efficient management. “Digital Twin” technology has been successfully applied in tomato greenhouse. Through intelligent environmental control management and managers’ feedback, managers can significantly reduce the number of controller adjustments. The introduction of this technology can effectively improve management, and each managers can increase the greenhouse management area by about 3 times.

In 2021, we have applied “Digital Twin” technology to other fields, the mushroom industry. Mushroom Bio-Tech Co., Ltd., located in Chunghwa, has introduced the “Digital Twin” technology (Figure 3) to cultivate organic mushrooms. By collecting temperature, humidity, various gas concentration data, and analyzing the operating habits of managers, the “Digital Twin” model of mushrooms is completed. This mode allows managers to understand whether the device is operating abnormally and provide suggestions on how to adjust the operation to avoid wasting resources.

Fig. 3. Description of the Demonstration Base for Smart Cultivation of MushroomsFig. 3. Description of the Demonstration Base for Smart Cultivation of Mushrooms

“Digital Twin” technology not only helps managers improve management efficiency, but also helps new farmers learn to manage greenhouses. TARI will use this technology to train young farmers this year (in 2022). This technology can combine artificial intelligence (AI) and human intelligence (HI) to collect data and analyze and model using intelligent technologies such as the Internet of Things and ICT. After the model is established, it can simulate the planting habits of professionals and provide environmental control management suggestions for accelerating new farmers to handle greenhouses. Through the continuous development of “Digital Twin” technology, the establishment of a greenhouse demonstration farm for vegetable and fruit production is a concrete demonstration of smart agriculture, so that smart agricultural applications can be implemented.

Ministry of Agriculture (abbreviated as MOA) started to promote the smart agriculture project in 2017. CIP uses Open API technology for data exchange and integrates the upstream and downstream supply demand of consumers' menus and pesticide inspection data. All these information can be verified through “Campus Food Ingredients Registration Platform” to create a new application of food safety traceability. In addition, CIP connects agricultural weather data, 3-Labels-and-1-QR-Code data, pesticide and fertilizer data, etc. so that users can quickly and conveniently interface with relevant data only through this platform for effectively shortening the time to communicate with each agency.

In 2016, the Ministry of Education and Ministry of Agriculture cooperated to promote the policy of preferentially using locally produced traceable ingredients for school lunches. This policy encourages school lunch to give priority to agricultural, fishery, and livestock ingredients of 3-Labels-and-1-QR-Code. The use of certification marks can track school lunch ingredients to ensure food safety. This transparent information can reduce parents' concerns about food safety. In addition, it can also establish a low-carbon footprint of local ingredients for eating habits. In 2017, CIP integrated the 3-Labels-and-1-QR-Code traceability data, and provided more than 3,000 primary and secondary school lunch secretaries to use it to facilitate the input of 3-Labels-and-1-QR-Code traceability data on “Campus Food Ingredients Registration Platform”.

In order to understand the traceable ingredients used by schools and their corresponding suppliers, we obtained the ingredients usage information (including primary and secondary school lunch ingredients and supplier information) from the “Campus Food Ingredients Registration Platform” on the government data open platform, and then combined them with “3 Labels and 1 QR Code” traceability data for big data analysis. Sankey diagram is a specific type of flow diagram. Since the starting flow and ending flow are the same, the sum of the widths of all main branches is equal to the sum of the widths of all branches, which is conducive to showing the correlation between dimensions and suitable for expressing the development of clusters. Therefore, the flow of ingredients is more clearly presented through the Sankey Diagram. This platform provides 9 data dimensions Sankey diagram analysis of the frequency of use of “3 Labels and 1 QR Code” traceability data. Users can set up to 6 dimensions and date ranges to find out the flow data of ingredients. After the graph is displayed, the user can change the order and type of the data dimensions according to their needs, so as to present a different order and flow of the data.

Through data collection and Sankey diagram analysis results (Figure 4), this data is automatically compared with the pesticide residue inspection data to find out whether unqualified ingredients have been used, as a reference for supplier evaluation. Furthermore, the control of food safety risks must be prevented before they occur. Among the group meal manufacturers or food ingredients suppliers that supply more than 3,000 schools across the country, how to effectively carry out the auditing and inspection of food ingredients is an important issue. Through Sankey diagram analysis, this platform can provide a list of the main suppliers of food ingredients to the Agriculture and Food Agency, which can be listed as key objects for inspection, and actively provide manufacturers with good management guidance to avoid food safety incidents.

Fig. 4. Sankey diagram of ingredients flow starting from schoolFig. 4. Sankey diagram of ingredients flow starting from school

In the past, the content of the label could only be viewed by the label number. Now that the menus in the campus lunch are integrated with the corresponding data of the ingredients, the label of the ingredients used in all menus can be directly queried through the menu name. In addition, you can quickly search the ingredients directly corresponding to the dish names planned every day, quickly obtain the label information of all ingredients, and increase the efficiency of ingredient verification. It can also assist relevant inspectors to analyze the frequency of use of traceability in “3 Labels and 1 QR Code”. By selecting relevant parameters with a visual and concise operation interface, analysis results can be obtained quickly, reducing the original inspection time from about a week to one minute. In the past, Excel was used to summarize and present, and only monthly overall analysis data could be provided. Now through the Sankey diagram visual analysis module, inspectors can perform multi-dimensional analysis functions according to the inspection needs, including different dates, months, labels, regions and schools.

The digital transformation of agricultural production technology has been regarded as the key to Taiwan's agricultural upgrading in the future. At the same time, it is also one of the keys to industrial competitiveness to drive information and communication technology application services. Experts in three fields - agricultural experts, information experts and agricultural producers - successfully applied the “Digital Twin” technology of manufacturing to the production of greenhouse crops through information management and application of artificial intelligence technology. It is an important milestone to extract the wisdom and experience of “experts” and pass it on. According to the needs of the agricultural industry, the information experts of Institute for information industry standardize and commonize the knowledge and information and build it on the “Common Information Platform” to lead agricultural production to a new digital production method.

It is expected that there will be a balance between supply and demand between production and consumption in the future. Data integration starts from the menu to form a complete traceability function from production to sales. The data from production can include label information and production management information in the future, and the consumer can collect lunch supply information from the consumer. After integrating the information of the two parties, data exchange is carried out to form a demand-oriented integrated analysis service, and finally a cross-platform application from production to consumption can be formed. The cross-ministerial and cross-platform integration of “Campus Food Ingredients Registration Platform” can advance towards the goals of trust traceability, risk assessment, and local consumption, establish a complete food supply traceability chain system, and then extend to related food supply fields.

TOP