AI Pork attack: what is "AI Pork"?

From its establishment in 1991 to the present, Jilin Jingqishen Organic Agriculture Co., Ltd. (hereinafter referred to as "Jingqishen"), a private pig enterprise, has raised Shanhei pigs for 28 years, and its products are provided to Walmart Sam Store, Carrefour, Aeon, China Resources Ole, JD.com 7Fresh and so on. After 40-year-old CEO Sun Yanchunzi inherited his father's career, he had a new plan in mind: apart from actively introducing mechanized equipment to improve efficiency, could more intelligent means be introduced to raise pigs and improve the management process?

For example, one of the difficult problems of raising pigs is how to count the specific number of thousands of piglets and how to recognize each pig. Furthermore, from the perspective of the management process of the pig farm, if we can use the power of machines to remind employees of "when and where", can it greatly improve the work efficiency?

"these technological advances are logically simple, but traditional companies like ours can't do this kind of research and development. Therefore, we urgently have this need, and we need some high-tech enterprises and enterprises with the ability to integrate resources to do this. " Sun Yanchun said.

1. Pig face recognition is coming.

On June 13, at the Fusong Jingqi breeding Base in Baishan, Jilin Province, Li Jialong, deputy general manager of JD.com, said that the initial research and development of pig face recognition technology was not part of the current "intelligent solution for animal husbandry." it's for the needs of JD.com 's insurance team.

"it's a long story. We started to do pig face recognition because JD.com insured pigs. For example, I raised 100 pigs and insured 10 of them, but no matter which pig died, I could apply for reimbursement because the insurance company did not know which 10 of the 100 pigs I had insured, so there was this problem. If more than 90 other pigs die, they can be reimbursed. JD.com insurance team is very eager to recognize these 10 pigs through a means, so we began to study pig face recognition technology in 2017. " "since September of that year, we often went to pig farms to collect data and learned about the current situation of animal husbandry in China, so we began to do JD.com farming," Li said. "

China is a big country in the breeding industry, in which the number of live pigs ranks first in the world, and the pork output accounts for about half of the world, but in terms of labor productivity, feed conversion rate, sow productivity level, production cost and so on, there is still a certain gap from the international advanced level.

This has become the background of the establishment of JD.com 's farming and animal husbandry team and the entry of a number of Internet companies into agriculture: in the world's largest pig farming and consumer market, intelligent means are used to transform and upgrade, in order to reduce some of the costs of pig farming and improve production efficiency.

In March 2018, JD.com 's farming and animal husbandry team began to carry out experiments. In December 2018, JD.com officially invested energetically and acquired a 9.43% stake. At present, Jingqishen has 1600 core groups and 8200 basic sows. At present, there are 7 large-scale pig breeding farms and 275 cooperative households in Jilin Province, and 200000 commercial pigs are produced annually, belonging to medium-sized pig-raising enterprises.

How does 2.AI raise pigs?

In February 2019, after the Spring Festival, several developers from JD.com 's agricultural and animal husbandry team settled in Jingjing's pig farm in Fusong, Baishan City, Jilin Province. According to JD.com 's farming and animal husbandry team, JD.com 's agricultural and animal husbandry intelligent farming solution based on AI, IoT and SaaS technology has been deployed in more than 100mountain black pig houses in two breeding parks in Baishan, Jilin Province, including Shennong brain, Shennong Internet of things equipment, Shennong system related systems and hardware equipment.

On June 13, the author saw in the exuberant Shanhei pig house that the pig house has been loaded with breeding inspection robots, feeding robots, agricultural-level cameras, telescopic semi-limited pig pens and other facilities. Through the above equipment, the application of image recognition and artificial intelligence technology is playing an important role in the field of agriculture.

The inspection robot is installed on the roof of the pig house and "walks" back and forth along the track of the inspection car, using the camera to inspect the movement in the shed.

According to JD.com 's agriculture and animal husbandry team, take the pig quantity statistics as an example, the general breeding enterprises will carry on the quantity count at the end of the month, the traditional naked eye pig count efficiency is very low, and the accuracy is not high, often needs 3 to 4 days to complete the quantity count work, as for the pig weighing, the traditional breeding only carries on the weight when going out of the fence, previously can only estimate based on experience.

The inspection robot uses a special agricultural camera to "glance at" the pigsty to know the number of pigs and weigh each pig, which is supported by the AI algorithm.

The camera in front of the feeding fence is part of an intelligent feeding robot that can "recognize the pig's face": each feeding area contains only one pig, and the image can be captured by the camera as the pig enters.

Feeding in the traditional mode, feeding pigs rely on manual feeding of 10 to 15 pigs, each pig can only rely on strength to eat, as a result, the same column of pigs are often different in weight.

Through the "pig face recognition" technology, the "identity information" of each pig can be identified, its weight and growth cycle information can be obtained, more accurate automatic feeding can be realized, and the weight difference of pigs in the same column can be reduced as much as possible.

It is worth mentioning that there are different voices about the practicability of "pig face recognition" technology, and there are other ways to solve the problem of identity recognition and quantity calculation on the market, including putting electronic intelligent "earrings" on live pigs, or tattooing coding on the pig's back and importing it into the system through the camera.

However, Li Jialong believes that the installation process of the above practice will still incur a lot of costs and have a certain impact on the pig itself, and "pig face recognition" can more effectively avoid the above situation. Li Jialong said that when the "pig face recognition" technology is specifically used, it needs to be localized for about 30 to 50 days.

In addition to the loading of the above equipment, the core of JD.com 's agricultural and animal husbandry technology solution is the "Shennong brain" developed by artificial intelligence technology. To put it simply, "Shennong brain" preserves professional pig-raising knowledge and is able to make independent judgments in various situations.

For example, when a pig suddenly barks weakly in the pig house and coughs twice from time to time, Shennong's brain first detects the abnormality and asks the camera to observe the pig's body temperature. call the past body temperature, eating records and exercise volume of the pig in the database to compare, if the barking, body temperature and movement are abnormal, then Shennong's brain can make a preliminary judgment that the pig is likely to be ill, and then Shennong's brain will alert the breeder.

3. Pay for AI's pig farming.

However, loading these cameras and robotic equipment is still a big investment for farmers. The breeding capacity of Jingshen is medium scale, which can achieve a certain scale effect in terms of cost saving, but for more small and individual farmers, it is necessary to calculate the input-output ratio before using this set of intelligent equipment system.

Jingqishen team said: "at present, after we put into the AI breeding system, taking Jingqishen Beishan Pig Farm and Xingshen Pig Farm as an example, it is expected that we will increase the average cost of each pig by 92.50 yuan (full investment, according to the annual production of 50, 000 pigs for 8 years, a total of 400000 pigs), but can save 30 million yuan per head in feed; labor cost input can save 8 yuan per pig. In terms of improving the survival rate, the average saving of 15 yuan per head, the accurate control of the body weight of the hurdle can save 50.8 yuan, and the total saving is 91 yuan. Judging from the direct production, we basically tied. "

However, as the duration of use increases, the average cost may fall further.

In terms of feed cost, the Jingqi team said in the written material, "through the previous test of the experimental group for nearly a year, we found that it can increase the feed conversion efficiency of pigs by at least 10%." This will be a huge benefit for us to produce 200000 mountain black pigs a year, and it is initially estimated that it can reach the annual festival level of about 12 million yuan. "

More importantly, intelligent technology has also brought about changes in management ideas and processes. "the key point is to reduce employees' blind homework and do less useless work. At the same time, we should also analyze and do a good job in the management of employees' work performance through the systematic management of the work process and results."

Sun Yanchun said, for example, that in the past, many rules and regulations were designated in farms, but their implementation depended on employees' sense of responsibility, so it was difficult to measure employees' performance, but now relying on smart devices and SaaS enterprise operating systems can achieve more refinement at the management level. "for example, there is a place in the pig farm where there are a lot of dung spots. A robot inspects it and sends a message to employees to clean it up. The first 10 minutes does not prompt the employees themselves, the second 10 minutes does not prompt the supervisor, and the third 10 minutes does not report to the factory director. After the orientation and compliance have been changed, the work quality and efficiency of employees can really do things around the core values of the enterprise. "

Sun Yanchun said that through intelligent management of equipment and information tips, a farm director can manage more pigs, "like our farm." In the past, there used to be a field director, two deputy field directors, and seven or eight technicians, such as the childbirth stage, the nursing stage, and the conservation stage. Now, after we have optimized it, there is only one chief director and one deputy director. With three or four more technicians, you can do all this work. "

Jingqi also said, "in addition to the most direct comparison of input and output mentioned above, the environmental protection value, breeding value and management value brought to us by the (intelligent breeding) system, as well as the basic depreciation of the system's software and hardware, are values that do not enter the scope of comparison. On the whole, this is a very cost-effective project. "

One of the areas with great potential is that the collection and monitoring of aquaculture information by intelligent systems may be able to pry the financial industry to increase support for agriculture.

4. JD.com 's solution for Intelligent farming of Agriculture and Animal Husbandry

The Jingqi team affirmed the role of the above-mentioned "Shennong brain" at the practical level: "when farmers observe the health of pigs, in addition to the limitation of technical level and means, what is more important is that they cannot spend a lot of energy and manpower to observe the health of each pig, which leads to the embarrassment of a large disease, that is, the end of the epidemic. With the intelligent system, the system will monitor the pig's behavior, momentum, body temperature, food intake, and even voiceprint information one by one, and identify abnormal individuals through calculation, and mark them to guide staff tracking and treatment. "

For the environment of the pig house, Shennong's brain will also carry out unified management and intelligent analysis of the collected pig farm environmental parameters, and then adjust them through intelligent equipment such as intelligent fans, intelligent aerators, intelligent humidity regulators, and so on. to ensure that the temperature, humidity and air of the pig farm are maintained in the best condition suitable for the healthy growth of pigs.

The above systems and equipment actually informationize the growth of each pig, according to JD.com 's agricultural and animal husbandry plan, from pig birth, growth, quarantine, slaughtering, processing, transportation to sales, all trying to be traceable.

"Jingqi Sheng will simultaneously transmit JD.com and big data, who is intelligently bred, to the Jilin Provincial Animal Husbandry Administration through the Internet, and integrate biosafety data and production traceability data into the epidemic prevention management platform of (Jilin) provincial bureau and the product traceability platform of 'Gong e Gong'." The spirit team said.

Not only daily breeding and management, in Sun Yanchun's plan, in breeding, the use of intelligent means or greatly improve the efficiency of breeding.

5. Visualize the information of the farm

Sun Yanchun mentioned that in the past, banks were quite cautious about agricultural enterprise loans. "We have to take a lot of collateral, which is very painful." The pig house of our enterprise is not recognized by others, nor do the pigs, so we take our hard assets, the real estate, land certificate and equity of the factory to mortgage. "

"it's all right this time. The (inspection robot) running above the farm runs around, and how many pigs there are can be counted very clearly at any time and anywhere. The bank can see it very clearly, and now they are relieved." Sun Yanchun said, "in the future, we will take the initiative to access the competent government departments and the financial institutions and insurance companies that provide us with credit services, so that the industry competent departments and business service departments can sit in the office as if they were in the front line. Know the stock condition and health status of the pig farm, so that everyone can carry out the management and service work at ease.

According to JD.com 's introduction of agriculture and animal husbandry, more and more aquaculture enterprises have begun to come into contact with smart solutions from Internet companies. Not only raising pigs, Li Jialong said that JD.com 's intelligent farming system for agriculture and animal husbandry will also play a role in raising cattle, chickens, aquatic products, planting, and so on.

"AI fresh butcher shop" comes out

On the eve of June 18 this year, the topic of "AI fresh meat shop" was listed on Weibo, causing a heated discussion among netizens. This pork shop named "AI fresh Meat Shop" is located in the JD.com Seven fresh supermarket store in Beijing Dazu Square. The storefront is neat and clean and full of technology. It is very eye-catching and looks very high-end. The "AI fresh butcher shop" jointly launched online and offline by JD.com Agriculture and Animal Husbandry under JD.com Mathematical Science and Jilin Jingjingshen Organic Agriculture Co., Ltd. is due to the fact that the pork sold by the fresh butcher shop is pork produced by means of artificial intelligence technology.

Artificial intelligence pig raising should be paid attention to.

In 1956, "artificial intelligence" was proposed. In 1997, Deep Blue beat chess master Kasparov. In 2017, "artificial intelligence" was selected as the "Top Ten buzzwords of the year in Chinese Media". The application fields include computer science, finance and trade, medicine and health, industrial transportation, remote communication, online services, legal advice, scientific discovery, music games, farming and so on. In 2019, the second session of the 13th National people's Congress included legislative projects closely related to artificial intelligence in the legislative plan.

In February 2018, Ali and Huaxi Hope Special Drive and Sichuan Dekang agriculture and animal husbandry cooperation, trying to complete the development and training of intelligent pig farms through the joint efforts of all parties and information sharing, to promote the intelligent development of domestic pig industry; in June, Ali Yun launched "ET Agricultural brain", which is applied to pig farming, apple and melon planting, with functions such as digital file generation, full-life cycle management, intelligent agricultural analysis, and full-chain traceability.

In August 2018, JD.com Yun, Nong Xin Interconnection and Tianzhao Pig Industry established a Smart Pig raising Alliance; in November, JD.com set up an agricultural and animal husbandry company to enter the pig industry and plans to develop and launch intelligent breeding solutions in cooperation with China Agricultural University and the Chinese Academy of Agricultural Sciences. The main goal is to reduce labor costs of large and medium-sized pig farms by more than 30%, save feed by 8% and 10%, and shorten the listing time by about one week.

At present, "African classical swine fever" is raging all over the world, and pork has been playing a prominent role in the meat consumption of Chinese residents. People are full of expectation and attention to artificial intelligence pig raising.

According to analysis, the mature artificial intelligence pig raising model can greatly save labor, reduce feed input and improve growth and reproductive performance by adopting advanced information, communication and Internet of things technology to realize highly automatic control of the whole production process of pig farms. Improve production efficiency, achieve real-time disease monitoring and effective control, make the whole process of pork production traceable and ensure food quality and safety. Compared with traditional farming, there are obvious advantages.

The pig industry itself is an industry with large investment and long capital recovery cycle, while the artificial intelligence hardware equipment is expensive and the investment is huge, and its stability and matching have not been well integrated and verified. the normal operation and maintenance of facilities and equipment requires high knowledge of professional and technical personnel, and the existing equipment and systems have not yet formed a mature model and reliable ability to accurately guide the whole process of pig farms of different sizes.

Science and technology is the primary productive force, artificial intelligence is advancing by leaps and bounds, we should be full of expectations and positive treatment of artificial intelligence pig farming. Combined with the current situation, for the development of artificial intelligence pig farming, the government should reasonably arrange special funds to encourage scientific and technological innovation, achievement recognition, technological integration, demonstration and promotion, and formulate policies and measures conducive to scientific and technological innovation. More guide social resources investment, and encourage the formation of a centralized firepower joint research, centralized breakthrough system and mechanism, to avoid the relevant subjects fighting their own battles and wasting social resources. Encourage powerful industry leaders such as Ali and JD.com to join forces with scientific research institutes to act as explorers, innovators, practitioners and promoters of artificial intelligence pig farming. For many small and medium-sized pig farms, it is better to pay timely attention to the scientific and technological progress in related fields and the innovation of facilities and equipment, and to share the achievements of scientific and technological development with an open mind. For any emerging technology, we should not only pay attention to the technical feasibility, but also consider the economic feasibility and optimization.