Liuzhou Iron & Steel Group: How Apusic AMDM Manages Millions of Data Assets
Open Innovation, Customer Success

A group enterprise rated A+ (extremely strong) in competitiveness among Chinese steel companies for five consecutive years, with businesses spanning more than a dozen segments — steel, logistics, trading, new materials, energy and chemicals — and hundreds of affiliated companies: when the group decided to advance its digital transformation, the first stumbling block was not a lack of systems, but data. The same business partners carried different codes in different systems; the same categories of materials followed different classification standards across segments; cross-system data reconciliation became a daily drain. This was the true starting point of Liuzhou Iron & Steel Group's master data system project, and a deep implementation of the Kingdee Apusic Master Data Management platform (AMDM) in the steel industry.
A Steel Giant's Data Pain Points Go Beyond "Inaccurate Data"
The steel industry is a key battleground for digital transformation. In 2024, nine departments including the Ministry of Industry and Information Technology jointly issued the Implementation Guidelines for the Digital Transformation of the Steel Industry, calling for a basic shift from point breakthroughs and localized applications to systematic, comprehensive digital development by 2026. Behind this policy window lies a challenge common to large steel enterprises: the more systems they build, the more fragmented their data becomes.

As a Fortune China 500 company, Liuzhou Iron & Steel Group started informatization early and invested heavily, equipping every business segment with its own management systems. But problems at the master data level gradually surfaced after years of accumulation: core data on business partners, materials, organizations, and personnel was scattered across dozens of systems without unified coding rules or field definitions; data quality issues often required repeated manual verification; and the existing master data platform offered limited functionality, no longer able to support group-level, cross-segment unified management.
The direct consequences of these problems went far beyond simple "data inaccuracy." Cross-segment procurement collaboration required repeated confirmation of partner information; consolidating financial statements consumed enormous time reconciling account data; and every new business system launch first had to solve the question of "how to reconcile with the legacy systems." Fragmentation at the data layer was slowing responsiveness at the business layer.
One Liugang, One System, One Platform
Facing this situation, Liuzhou Iron & Steel Group set a clear goal: one Liugang, one system, one platform. For the platform selection, the group ultimately chose Apusic's AMDM master data management platform.

AMDM is positioned as a master data governance platform for large group enterprises, with core capabilities spanning master data model management, rule configuration, a workflow engine, system integration, and log monitoring. For Liuzhou Iron & Steel Group, the value of this capability lies in being "configurable and extensible" — master data scenarios in steel enterprises are highly complex, encompassing both general partner and material management and industry-specific needs such as shipping locations and project classifications. Standardized product features cannot be applied directly; the platform must support deep customization on top of its foundation.
After the Apusic team came on board, the first step was not installation and deployment, but spending several weeks working with Liugang's business teams to investigate the status quo segment by segment, sort out pain points, and confirm standards. The two teams adopted a strategy of "thematic breakthroughs, domain-by-domain advancement": splitting the eight major master data domains — business partners, materials, human resources, shipping locations, and others — into independent workstreams, confirming the data model, management process, and responsible owner for each, and then solidifying them through the AMDM platform.
Tackling Eight Domains One by One, from Blueprint to Launch
The project covered the group's headquarters, its listed joint-stock company, and core subsidiaries, involving eight master data domains and integration with dozens of business systems. Implementation proceeded in stages — research, blueprint design, system configuration, data cleansing, key-user training, UAT testing, and go-live — each with clearly defined acceptance checkpoints.
Business partner master data was among the hardest domains to crack. Liugang's business characteristics make its partner types extremely complex, spanning external suppliers and customers as well as internal suppliers, affiliated companies, employee suppliers, and individual partners from collaborative industrial units. Using AMDM's model configuration capabilities, the project team designed dozens of finely grained attribute fields for partner data, refining management granularity to a level sufficient to support operations across procurement, finance, logistics, and other lines.
Other domains such as human resources, shipping locations, and chart of accounts followed a similar logic: the business side and the Apusic team first jointly confirmed standards and processes, then solidified them into the system through AMDM's workflow engine and rule configuration, and finally distributed the standardized data to downstream business systems via the platform's unified integration interfaces.
Value Beyond Data Quality
The direct benefit of a master data management project is improved data quality, but its value goes far beyond that.
At the data layer: master data shifted from scattered maintenance to centralized control and unified distribution. After hundreds of thousands of master data records were standardized on the AMDM platform, they were synchronized to dozens of business systems through unified integration interfaces. A change at the data source now triggers coordinated updates across all systems, eliminating duplicate entry and inconsistent information.
At the management layer: the group established a standardized process covering the full master data lifecycle — from application and review to change and retirement — with a clearly designated owner for every step. Powered by AMDM's built-in workflow engine, data issues that once required repeated cross-departmental confirmation can now be routed and handled automatically according to preset rules, markedly improving management efficiency.
At the business layer: the unified master data system removed obstacles to cross-segment collaboration. Financial accounting and business operations are no longer "two separate skins" — from business occurrence to financial posting, from purchase ordering to supplier settlement, data flows automatically between systems, giving group-level business analysis and risk control a more reliable data foundation. More importantly, the master data platform lays the necessary groundwork for deeper digital applications ahead, whether big data analytics or intelligent decision support.
An Industry Perspective: Why Master Data?
The digital transformation of steel enterprises is at a critical stage of transition from localized applications to systematic construction. After investing heavily in MES, ERP, and supply chain systems, many enterprises have gradually realized a problem: no matter how advanced the upper-layer applications, if the underlying data speaks different languages, the results they produce may contradict one another.
By choosing to start with master data at this juncture, Liuzhou Iron & Steel Group was essentially answering a more fundamental question: what exactly does digital transformation transform? It is not simply about deploying a few more systems — it is about making data a truly circulable, trustworthy, and reusable asset.
Apusic AMDM's practice in this project also validates the value of master data governance for large manufacturing groups: for large manufacturers advancing digital transformation, a master data platform is not icing on the cake but part of the digital infrastructure. Liugang's experience offers a reference-worthy example: as an enterprise grows larger, its businesses more diverse, and its legacy systems more complex, unify the language of data first before pursuing intelligence in upper-layer applications. This step may look basic, yet it is often the most easily overlooked — and the hardest to make up for later.
Latest News

Driving Innovation and Development in China's Foundational Software Industry

Driving innovation and development in China's foundational software industry

High-Quality Development: Interim Results Review
