Dry goods-case sharing of decision-making intelligent system for iron and steel enterprises

Iron and steel industry is one of the important raw material industries in China, which urgently needs

Dry goods-case sharing of decision-making intelligent system for iron and steel enterprises

Iron and steel industry is one of the important raw material industries in China, which urgently needs to be upgraded from "big" to "strong". In the last article, we shared the main difficulties and pain points of digital transformation of traditional enterprises from the perspective of Olin technology delivery team. In this issue, we will share the case of digital transformation of the steel industry.

Large-scale iron and steel group is a large-scale iron and steel joint enterprise with production capacity 1000 million yuan and tax payment1000 billion yuan. As a leading domestic iron and steel enterprise, it is ready to further implement the conversion requirements of old and new kinetic energy, and plans to build an advanced iron and steel production base through reduction and replacement. After years of informatization construction, the enterprise has vertically established the automation and informatization system of L 1~L4, and horizontally realized the end-to-end informatization coverage of procurement, inventory, production, sales, logistics and finance.

But the information system of this iron and steel group is also facing new challenges:

1. The data self-recovery rate is low. There are many problems in informatization, such as manual input adjustment, data inconsistency caused by multiple inputs, timely coordination and dislocation of information, etc.

2. The information closed loop is not formed. There are some problems in data analysis, such as the core data management can not be upgraded independently, the data access performance is risky, the data usage mode is single, the closed loop of information processing in the system is not fully formed, and the data analysis ability is relatively weak.

3. Excellent experience and knowledge are not solidified. A large number of business data analysis relies on manual completion and personal experience judgment, which can not be analyzed and fed back in real time and synchronously with the business. Historical data assets are used relatively little, and historical data has not yet formed the intelligent analysis support of enterprise management;

4. There are risks in traditional buildings. This system extends the traditional IOE information technology, and there are technical support risks.

All the above problems have hindered the intelligent transformation process of this steel enterprise.

Supported by the products of Olympic Mathematical Platform and Event Network, Aoshu Technology opens up the whole business process data of procurement, production, inventory, sales, orders and marketing, analyzes the data from the overall operation of the enterprise, and helps customers realize the optimization of ore cost balance, the optimization of enterprise quantitative decision-making, and the intelligence of enterprise business analysis through the quantitative decision-making system, and constructs a complete intelligent decision-making auxiliary analysis system.

Internal data, external supply chain data, industrial cycle data, macroeconomic data, competitive environment data, industrial big data, etc. Comprehensive construction, according to the time dimension to form a three-dimensional multidimensional data model. According to the data model, quantitative analysis and insight based on big data are given, which are pushed to PC and mobile phone in the form of events and risks, providing direct suggestions for leadership decision-making.

Through artificial intelligence technology, the digital models of operating experience and key business nodes of various departments are established, and at the same time, different models are connected into multi-dimensional business models of the whole enterprise through knowledge maps, so that the data of various functional departments can play a role in the overall perspective of the enterprise and form global optimization; Through iterative model training, quantitative analysis and optimal scheme are provided to assist decision-making, and lean management is formed.

1. Enterprise Management

The data, process, information system and business activities of Iron and Steel Group are deeply analyzed and sorted out, and implemented according to the survey results and the available scope of information data of Iron and Steel Group.

2. Master data management, establishing a unified master data asset management platform.

The master data asset management platform contains a set of specifications and technologies for generating and maintaining master data. The complete platform includes metadata management, information system integration, data governance, data analysis, data exchange and other functions.

Examples include:

Feasibility implementation scheme of master data system of iron and steel group (including data acquisition, data quality analysis, data source analysis, data resource investigation, management granularity, etc.). ) I tidied it up.

Master data management system implementation (basic environment deployment, prototype iteration and preview, master data collection, data cleaning, conversion, data mapping, master data quality management implementation, system performance tuning, etc. ).

3. Data Lake, establishing a unified data integration platform.

Examples include:

Combined with the feasible implementation scheme of information system data lake construction in Iron and Steel Group (including data infrastructure, data access scope, model and data integration standard, etc.). ), reasonably plan the granularity of data storage, build a hierarchical structure of dimensions, and form a unified data center. Through the extraction, conversion, cleaning and loading functions of multilevel ETL, the organic combination of various data sources is realized, which ensures the quality of data sources and the integrity and consistency of information.

The realization of data lake management system (through the configuration management of the pre-data extraction function of each information system, the management data of each professional system can be obtained and integrated in time).

4. Make comprehensive operational decisions and establish a unified display platform.

Through the intuitive display of key indicators, operators can obtain the company's business information completely, timely, globally and efficiently, so as to achieve the purpose of transparent business information.

Examples include:

Combing the comprehensive business decision-making and business experience solidification of the Iron and Steel Group (including index system, existing business processes, ERP system docking, CRM system docking, cost analysis and cost calculation, etc.). ), and reasonably define the functional division of business system and business analysis system.

Implementation of comprehensive operation decision system (thematic data induction and processing, algorithm model design and development, deep learning and model optimization of historical data, customization of management console, customization of information system interaction and docking, etc.). ).

5. Purchasing inventory optimization, establishing purchasing inventory optimization assistant decision-making service and its application.

Considering the importance of procurement and inventory optimization of raw materials and auxiliary materials and the urgent need for improvement, the corresponding auxiliary modules will be planned and implemented separately.

Through comprehensive modeling, analysis and deep learning optimization on the data of iron and steel group's pre-ironmaking production data, pre-ironmaking equipment maintenance data, purchasing data, inventory data, purchasing price index, logistics data, ore blending scheme, pre-ironmaking quality data, cost calculation model, production plan data, production performance data and finished products, the dynamic intelligent recommendation aided procurement optimization scheme, dynamic optimal inventory scheme and accounts payable structure optimization scheme are formed.

Through the construction of the above five dimensions, the management decision-making ability of iron and steel group has been successfully improved: managers can keep abreast of the company's operation at any time, and provide basis for the daily data analysis and decision-making of senior leaders and business department analysts of iron and steel group. At the same time, it can reduce the operation difficulty of users, reduce the training cost of users, and provide fast and rich statistical analysis data and decision support for the management of the company, so that it can focus more on business optimization and management, thus further improving the business operation efficiency and decision-making ability of the iron and steel group company.

In addition to improving the management decision-making ability of iron and steel group, the following value benefits have been realized:

? Analysis on the transparency of company management and the combination of industry and finance. Compare whether the annual target and historical data maintain sustained and stable development from multiple angles.

? Auxiliary support for the company's business decision. Benefit prediction, simulation and auxiliary optimization suggestions, market fluctuation impact analysis, etc.

? Coordination of contract order life cycle. Taking the whole life cycle of contract orders as the main line, production, supply and marketing cooperate to find problems.

? Production, supply and marketing decision support. In the case of reasonable capital occupation, the supply and production are stable, and the dynamic balance between production and sales under market changes is supported.

As a large-scale complex process industry, it is difficult to obtain the whole process production data in iron and steel enterprises, and most of them are opaque "black boxes". Enterprise-level digital twinning based on Olin technology event network technology makes the whole process of procurement, production and sales of iron and steel enterprises transparent. Through the simulation of enterprise digital twins, the optimization strategy is obtained by using artificial intelligence model, and the relevant instructions are fed back to the production and operation departments for execution, forming a closed loop of enterprise overall intelligent optimization. By providing digital decision-making and operational support capabilities for the iron and steel group, the soft power of the enterprise and the competitiveness of the whole industry have been effectively improved, and the benchmark for digital transformation of the iron and steel industry has been created.