Zhang Hancheng: Thoughts on the construction and application scenarios of customs big data platform

Text / Zhang Hancheng

The 13th Five-Year Plan of the Customs clearly stated that big data technology should be used to enhance the level of customs management intelligence. In order to empower the customs with data science, the improvement of the level of intelligence must be inseparable from the construction of the platform, the driving of customs business by various algorithm models of big data, and trade intelligence (deep mining of customs data) Therefore, the author shares the research on the platform construction of customs big data and the research of big data application scenarios, and throws a brick.

Let's take a look at the five main contents of big data mentioned in the 13th Five-Year Plan of the Customs :

1) Establish an overall coordination mechanism for the application of customs big data, implement lead management responsibilities, and form a work pattern with clear responsibilities, sharing and sharing, and collaborative promotion.

2) Combining the "three mutual" customs clearance and "single window" construction, in accordance with the national "Internet government service " technical system construction standard, various external data interfaces are reserved to promote technology integration, business integration, data integration, and docking with the central government website The general portal and platforms of various departments in various regions serve the construction of a nationally integrated national big data center.

3) Adapting to the characteristics of massive, high-growth and diversified information sources, based on data such as customs declarations and manifests, a new processing mode is adopted to extensively collect and analyze information and intelligence related to commodities, industries, and enterprises for risk analysis Provide a unified " data pool " for prevention and control . Implement the leading strategy of information intelligence, build a modern customs intelligence system of " extensive collection, professional research and judgment, scientific disposal, sharing and interconnection", give full play to the leading role of information intelligence in customs supervision and smuggling, and achieve intelligent perception of high-risk targets Intelligent identification and precise locking.

4) Intensively transform the customs integrated business management platform, establish a digital closed management mechanism for operation monitoring, problem diagnosis, execution feedback, and comprehensive display, strengthen data quality control, and play its role as an information portal for customs business management.

5) In the direction of intelligence and automation, improve the customs law enforcement information system, establish a predictive corporate credit evaluation model and risk analysis model, realize computer automatic collection and formation of control parameters, and realize automatic adjustment of parameters and corporate credit by disposal results. Improve the pertinence and effectiveness of customs supervision.

It is not difficult to see from the planning that the construction of customs big data must first clarify the team, secondly, promote the three integrations mentioned in the big data action plan, and provide an open platform for customs data to connect to a local government terminal, and thirdly, customs declaration Bills, manifests and other data should be widely documented by intelligence, provide a data pool for risk analysis, realize intelligent perception, identification, and fourth, improve the original business platform of the customs, conduct business data and data business-oriented closed-loop management, Fifth, we must establish an intelligent automation direction;

In simple terms, it is to establish a multi-party data sharing and interconnection customs big data platform, rebuild a new business system on it, realize the interconnection of business systems, and use big data to drive customs, supervision, taxation, private investigation and compilation Customs statistics are processed intelligently and automatically.

Business dataization, data businessization, and business intelligence are the three stages of data-driven business that everyone basically recognizes. Naturally, the construction of big data platforms by the customs is also this routine. Data collection, data management, data modeling, data analysis, Data visualization (mobile, data application portal, large-screen supervision), so the big data system platform for customs construction will also go through three stages and five steps, summed up is the establishment of five major systems and 18 applications.

Because the construction of the platform will be different because of the specific data volume, data processing requirements, the relationship between data and business, data real-time, data security, data governance, management, etc., each system market has mature products, It can be improved according to the specific business. Therefore, this process is long and constantly improving and adjusting according to the needs. Specifically, you can make some adjustments to the technical architecture in the following figure:

Regarding big data analysis of customs management, the main scenarios of business applications mainly include the following two categories:

The first category: the application of internal data management and data decision-making:

1. After the customs business is digitized, the comprehensive management capabilities of the customs are improved, which will play a vital role in the core functions of China's customs: supervision, taxation, private investigation and compilation of customs statistics. Supervision and utilization of data will be more refined. Taxation itself is also dependent on data. Tax evasion and branch control are inseparable from the data. Audit smuggling can be discovered through big data and related analysis. The compilation of customs statistics is an important basis for the country to develop trade.

2. After the customs business is digitized, it can provide trade intelligence information for import and export trade. ;

The analyzed and processed information will enable the foreign trade executives to clearly understand the status of international trade transactions, transportation, and the status, performance, capabilities, and motivations of competitors in the products or industries they care about.

3. Customs business data mining can help correctly identify high-quality buyers and establish direct cooperation;

4. Customs data mining can help to insight into product changes, master the most cutting-edge market information, and monitor the dynamic development of the industry;

Provide a basis for the country to formulate product support policies;

5. Customs data can provide trade intelligence products and guide enterprises to adjust their trade strategies;

The second category: the application of customs services to the outside world mainly includes

Companies looking for sellers: This is mainly because customs data bills of lading are transaction vouchers for buyers and sellers, and are updated monthly, not only to ensure the authenticity, accuracy and timeliness of buyer information, but also all currently active buyers; The inquiry and analysis of its transaction records can clearly know the buyer's purchased products and strength, so as to find the most suitable buyer.

2 ) Use customs data analysis to help companies find the relationship between the number of products, delivery time, and replenishment time, and find out the buyer's procurement rules. Recommend products for overseas sellers and increase the success rate of signing orders. It is also possible to know from the big data which suppliers of the same type of product the customer purchases from. Through comparative analysis of competitors and themselves, it can discover other needs of customers, find product gaps, and adjust production.

3 ) Use customs data analysis to understand competitors ’transaction records and tracking, to be able to grasp their buyers’ data and procurement laws, and to judge competitors ’operating conditions. Once a competitor goes bankrupt, it can respond to its customers immediately Get the most benefit.

It can be seen that the macro and micro data analysis of customs data benefits the country and the enterprise.

In addition, the author collected some application cases of big data we can see on the Internet in customs, and share as follows:

1) Xiamen Customs

Through big data analysis , he grasped a major clue that a trading company in Xiamen was suspected of smuggling imported wine under-reported prices. The company was suspected of smuggling imported wine and more than 600 containers, and the value of the case was verified to be more than 200 million yuan.

2 ) Shanghai Customs locks on smuggling of "foreign garbage"

The Shanghai Customs found through big data analysis that a trading company in Shanghai, etc. , has been suspected of smuggling imported more than 28,000 tons of waste plastics under the false name since October last year .

3 ) Qingdao Customs

By sharing and using 6.65 million external data such as industry and commerce, social security, taxation, foreign exchange, etc., an intelligent supervision data pool has been established for more than 70,000 enterprises under the jurisdiction of Qingdao Customs . There are 145 intelligent rules in the system , covering customs, supervision, processing trade and other customs business areas, to achieve intelligent "portrait" of the enterprise and precise targeting. According to the system "portrait", customs staff implements differentiated treatments such as auditing and risk early warning to enterprises, and implements the working ideas of strong supervision, excellent service, risk prevention, and plugging loopholes.

4 ) Capital Customs

The capital customs workers scan the barcode through the machine. If everything is normal, then pass a machine that can see through the six-dimensional three-dimensional "perspective". The parcel "journey" can be completed in less than 30 seconds. If suddenly, an "abnormal" package slides from the fork to the re-inspection channel on the first floor of the simple board room next to the sorting line. Scan the bar code again, in addition to the package name, receiving address, etc., the customs officer's computer also has more information such as the declared value and the photo of the inspection. Wang Zhaoxiang carefully compared the shopping tickets and issued the "release" instruction. The "abnormal" package is re-entered on the road, and it is automatically sorted to the corresponding compartment. Next, it will be sent to the customer. The seemingly simple process operation relies on a big data analysis system as a "super brain" . "In a few scans, the system can analyze massive amounts of cross-border commodity data and set the Screening for risk factors, and integrating customs supervision data at various sites through cloud servers to enhance the comprehensive supervision and risk prevention and control capabilities of customs cross-border e-commerce. "

5 ) Hangzhou Customs

We have reached a consensus with Alibaba's e-commerce platform to jointly crack down on infringements in the cross-border field of the Internet, and cooperate in the areas of clue exchange, big data analysis, and collaborative tracking. At the same time, Hangzhou Customs has strengthened law enforcement cooperation with the public security organ's economic investigation department, and conducted in-depth exchanges around issues such as improving the notification mechanism of intellectual property case clues, strengthening the intelligence management of key cases, and focusing on the use of e-commerce platforms to export infringing goods. Deepen the connection between customs administrative law enforcement and public security criminal law enforcement.

6 ) Shantou Customs

According to Shantou Customs, at the beginning of this year, the customs found through data analysis that since 2013 , a Shantou trading company imported “DORCO” brand blades in Shantou Guan District. The declared price was around RMB 600 per box ( 10,000 pieces) . The market price of a double-sided security blade is between 1800 and 1900 yuan per box, and the price difference between the two is quite different. Compared with the import price of this type of blade in other customs areas, it is preliminarily concluded that the company has a high and low concealed price. In response, the Shantou Customs Anti-smuggling Department, on the one hand, increased its efforts to investigate the external market, and on the other hand, paid close attention to the customs clearance of the suspected company. Here you can use data to analyze the comparison of similar products to find the problem.

However, from the current point of view, it is not difficult to find that many of our big data applications are currently used for internal inspection and crackdown on smuggling, and the analysis of intelligence big data in the face of international trade is a blind spot. From the media, it is very There are few articles of intelligence analysis. But the importance of trade intelligence, the author checked the Baidu Encyclopedia.

In Baidu Encyclopedia, this describes intelligence analysis. In the West, trade intelligence has become the latest multi-purpose weapon in the strategic arsenal of the enterprise market. According to statistics, the top 100 companies in the Global 500 of the US magazine "Happiness" and the companies in the United States have their own specialized departments responsible for collecting competitive intelligence, which penetrates into every branch of the company. At the end of the last century, the total output value of the global trade intelligence industry had reached 70 billion US dollars. At present, a large number of enterprises will inevitably participate in the " big global market " competition, and the accelerated pace of trade transparency will directly bring Chinese foreign trade enterprises to face more and more serious information asymmetry. Taking North American countries as an example, a large number of companies can obtain the international market conditions of any product in many countries around the world within about 5 to 60 seconds; in 1 minute to 3 minutes, you can query the United States, Detailed transaction information of the import and export trade commodities, specifications, etc. of Mexico, Canada, and Southeast Asian countries and all parts of the world; it can take 3 minutes to 5 minutes to find out the production and operation data of 10,000 key enterprises at home and abroad for the current or past years; From minutes to 10 minutes, you can call up various laws, decrees and parliamentary records made by governments ... " This set of data, which is not new in developed countries, is still both desirable and exciting for many Chinese companies. Shameful.

Compared with this, China's application of trade intelligence, especially in the field of international trade, is still a blank. Some of the large number of Chinese enterprises, to collect information through a variety of ways, a variety of ways, but the effect is minimal because of its weak business intelligence competitive nature of our industry is currently the basis of intelligence, "inherent" lack of development and competition The " acquired " dysfunction of intelligence .

Final summary: Customs big data development has encountered major opportunities. Customs practitioners need to establish big data thinking, learn the changes brought by big data technology, and use big data analysis to improve business capabilities. At the same time, the big data platform and application scenarios of customs can consider the following points:

First, taking into account the cross-industry data fusion, especially how to integrate tax and financial data necessary to establish the concept of data of the lake, to promote the integration of data 3, 5 across the building (technology integration, business integration, data integration, to achieve cross (Level, cross-region, cross-system, cross-department, cross-business collaborative management and services);

Second, considering the productization of the scene, it is convenient and easy to use, and it is easy to make clear the value of the data ( trade inspection, document inspection, smuggling inspection ) ;

Third, data releases value in the form of intelligence business models to provide decision support for trading companies;

Fourth, data service companies allow data to run more, and companies run less.

The data contains business laws, discovering laws, and doing things according to the laws, which can save time and improve efficiency. The construction of customs big data can help us grasp the trade market trends in a timely, comprehensive and objective manner, analyze the business situation in overseas markets, and provide decision support for import and export trade.

Welcome to communicate with peers! Zhang Hancheng (WeChat: waynezhanghc )

Zhang Hancheng is an Oracle data expert. His research areas mainly include: Introduction to Big Data Fundamentals, Application Practices of Big Data in Enterprises and Governments, Business Models for Data-Driven Business Transformation, Medical Big Data Operating System, Tax and Big Data, Customs Big Data, Operator Big data construction plan, tourism big data platform construction plan, data asset management, big data industry ecological analysis, data transaction market, blockchain, artificial intelligence and other new technologies are the value and empowerment solutions for traditional enterprises.