Why Does Manufacturing Data Standardization Matter?

"Every data project in manufacturing ultimately splits into two camps — those built on standards, and those that failed without them." — Gartner, Manufacturing Data Fabric 2025

In September 2025, the EU Data Act formally entered into force. From 2027, a Digital Product Passport (DPP) will be mandatory for every battery and electronic product sold in the EU market. By requiring full lifecycle data — from raw materials to final recycling — to be shared in a standard format, this regulation has turned manufacturing data standardization from a "recommendation" into a "condition for survival."

The cost of missing standardization on the factory floor has been known for a long time. A 2024 McKinsey analysis found that the total amount global manufacturers spend on data integration and standardization averages 38% of their entire IT budget. Even so, most organizations have treated standardization as a narrow "IT project," and the result has been a cycle of solving the same problems again every year.

The standardization landscape of 2025, however, is fundamentally different from before. With ISA-95 embodied in the OPC UA Companion Specification, Catena-X built on the IDS standard, and the DPP written into EU regulation, the equation "standardization = business execution capability" now holds. This article maps that terrain and answers why manufacturing data standardization is now the most urgent strategic priority.

Standardization is an asset whose cost stays hidden — its existence becomes visible only when it is missing.

1. The Real Cost of Missing Standardization

Ironically, the need for manufacturing data standardization shows up most clearly in the cost of its absence. Scattered data incurs integration costs every time a new system is introduced — but that is only the beginning.

According to a 2024 analysis by the McKinsey Global Institute, 38% of global manufacturing groups' IT budgets is consumed by data integration, cleansing, and mapping. Much of this cost is recurring. The same data mapping work is repeated every time a new line is introduced, a new KPI is defined, or a new AI model is trained. Bain's 2025 report names this "data debt," pointing out that its interest rate is higher than the cost ratio of IT projects.

Even more serious is the opportunity cost. IDC's 2025 survey concluded that "the biggest cause of delayed AI adoption is the absence of standardization." More than 80% of AI teams' time goes into data mapping and cleansing, leaving less than 20% for actual modeling and validation. In other words, the absence of standardization slows AI adoption by a factor of five.

Direct cost — an average lead time of 6-14 months per integration project (Mastek 2025)
Recurring cost — the same mapping work repeated whenever a new system or line is added
AI delay cost — 80%+ of time consumed by data cleansing (IDC 2025)
Regulatory compliance cost — fines of up to 2% of revenue for failing to comply with the EU Data Act, DPP, and CBAM
Governance failure cost — hard-to-quantify long-term losses such as failed audits, recalls, and brand damage

2. ISA-95 — The Basic Grammar of Manufacturing Information Layers

The starting point of manufacturing data standardization is ISA-95 (IEC/ISO 62264). Developed since the late 1990s, this international standard defines the flow of information between Manufacturing Operations Management (MOM) and Enterprise Resource Planning (ERP). Today, most industrial data architectures are based on ISA-95, whether explicitly or implicitly.

The real power of ISA-95 lies in its hierarchical information model. It represents assets and processes in a five-level hierarchy — Enterprise → Site → Area → Work Center → Work Unit — and defines five core entities at each level: assets, processes, products, personnel, and materials. The moment systems share this grammar, they can talk to each other in "the same language."

But ISA-95 is a grammar, not a dictionary. To answer a question like "what is the vibration RMS of machine 4 on line 1," a concrete data model must sit on top of the ISA-95 grammar. Filling that gap is the job of the OPC UA Companion Specification, covered in the next section.

Enterprise/Site/Area/Work Center/Work Unit — the five-level asset hierarchy
Personnel/Equipment/Material/Physical Asset/Process Segment — the five core entity models
Operations Definition/Schedule/Performance/Capability — the four categories of operations information
The IEC 62264 international standard — an ISO/IEC certified architecture
The foundation of Companion Specs — the common reference point for OPC UA, MTConnect, and Umati

ISA-95 is the grammar of manufacturing data — each industry's dialect is layered on top of it.

3. OPC UA + Companion Specifications — The Standard for Industry-Specific Information Models

The OPC Foundation operates the Companion Specification system, which defines "industry-specific dialects" on top of the ISA-95 grammar. As of 2025, more than 20 Companion Specifications have been published, covering robots, injection molding machines, machine tools, vision systems, and even battery cell manufacturing.

The strength of OPC UA Companion Specifications is "reuse." A 2024 analysis by Hicron Software reported a case in the packaging industry, which adopted both OMAC PackML and the ISA-95 Companion Spec, where the time to integrate new equipment fell from an average of four months to six days. Because the data definitions already exist as standards, the "mapping" work effectively disappears.

In December 2025, the OPC Foundation released an improved version of "OPC 10030 UA Companion Specification for ISA-95 Common Object Model." Notably, it newly includes a standardization methodology for heterogeneous process chains such as battery cell manufacturing. The 2025 MDPI paper "Developing a Concept for an OPC UA Standard to Improve Interoperability in Battery Cell Production" covers in detail how this approach reduces complexity and duplication through "reuse and extension of existing Companion Specs."

Companion Specs across 20+ industries — robots, CNC, vision, energy, batteries, and more
The reuse principle — new industries are designed by extending existing specs
Auto-Discovery + metadata — the foundation of plug-and-play integration
OPC 10030 ISA-95 Companion Spec — the standard bridge for MOM-ERP integration
VDMA umati — the flagship success story built by the machine tool community

4. STEP AP242, QIF, and JT — Standards for Product and Quality Data

If ISA-95 and OPC UA are the standards for operational data, STEP AP242, QIF, and JT are the standards for product, design, and quality data. The triad of ISO 10303-242 (STEP AP242 "Managed model based 3D engineering"), ISO 14306 (JT), and ISO 23952 (QIF) forms the standards foundation for realizing the model-based enterprise.

The core of STEP AP242 is "semantic embedding of PMI (Product Manufacturing Information)." Manufacturing instructions such as GD&T (geometric dimensioning and tolerancing), surface roughness, and heat-treatment specifications are embedded in the 3D model itself in machine-readable form, so downstream systems can use them without human intervention. Capvidia's 2024 QIF Definitive Guide reported that this combination of standards cuts CMM (coordinate measuring machine) programming time by an average of 70%.

QIF (Quality Information Framework) goes further by standardizing the entire lifecycle of quality data. Measurement plans, measurement results, statistical analyses, and audit reports are all expressed in a single XML schema, connecting design, manufacturing, quality, and audit into a single data flow. STEP AP242 and QIF share each other's reference models, securing natural interoperability.

5. VDMA umati — A Flagship Success Story of Standardization

umati (Universal Machine Tool Interface), led by the German Mechanical Engineering Industry Association (VDMA) and the German Machine Tool Builders' Association (VDW), is the definitive example of what industrial data standardization delivers "when it works properly." Launched in 2019, the initiative standardized an OPC UA-based Companion Specification for machine tools, and as of 2025 more than 300 machine tool manufacturers support it.

Three factors explain umati's success. (1) Community-driven — actual users (machine tool operators), manufacturers, and system integrators developed the specification together, centered on VDMA and VDW. (2) Incremental scope expansion — it first standardized only a handful of core data points (operating status, position, alarms) and broadened from there. (3) Testing and certification — a process for certifying standards compliance was in place from the beginning.

The result is clear. A umati-compliant machine tool can be integrated with any MES or MOM system within a day. This pattern is being transplanted directly into the battery industry's "Battery Passport OPC UA Companion Spec" and the semiconductor industry's SEMI standards.

umati — how an industrial standard built by a community becomes a commercial success.

6. Catena-X and IDS — Data Standards That Cross Company Boundaries

Where the standards discussed so far (ISA-95, OPC UA, STEP, QIF, umati) standardize data within an organization, Catena-X and International Data Spaces (IDS) standardize data across company boundaries. Launched in 2021 to address the automotive industry's circular economy, supply chain transparency, and ESG regulatory needs, Catena-X had by 2025 become Europe's largest industrial data space with more than 250 participating companies.

IDS (International Data Spaces), the foundation of Catena-X, is an architecture specification that enables standardized data sharing while guaranteeing "data sovereignty." Data publishers attach usage policies to the data they share, and consumers can access the data only by automatically complying with those conditions. IDS's Dataspace Protocol (DSP) is a framework standardized by the IDSA — the interoperability layer shared by all data space implementations.

In August 2025, Catena-X and the OPC Foundation announced a strategic collaboration. At its core is the integration of data standards to support the Digital Product Passport (DPP) that the EU will make mandatory in 2027. This completes a single standards stack: OPC UA inside the company, Catena-X/IDS between companies. The successful interoperability PoC in March 2025 with the Ouranos data space of Japan's Information-technology Promotion Agency (IPA) shows that this standard is expanding toward global interoperability.

7. The EU Data Act and Digital Product Passport — Standardization Becomes Regulation

With the EU Data Act entering into force in September 2025, data standardization became a matter of regulatory compliance. The Data Act grants users of IoT devices, vehicles, and machine tools "the right to access the data they generate," and manufacturers must provide it in a standard format. This single provision forces the standardization of data interfaces across every industry.

The Digital Product Passport (DPP) is its extension. Mandatory from 2027 for batteries, electronics, construction materials, and other products in the EU market, the DPP requires full lifecycle data — from raw materials to recycling — to be shared in a standard schema. Catena-X was selected as the reference data space to support it, and OPC UA Companion Specs were designated as the standard inside the factory.

Add CBAM (Carbon Border Adjustment Mechanism), which takes effect from 2026, and the need for standardization is no longer up for negotiation. CBAM requires "carbon emissions data" to be reported in a standard format for steel, cement, aluminum, and other goods exported to the EU. According to IDC's 2025 survey, 62% of Korean manufacturers said their "preparation for CBAM data reporting is insufficient."

8. A Practical Path to Adopting Standardization

The need for standardization is clear, but where to start differs by organization. The following is a practical path distilled from what successful cases as of 2025 have in common.

Step 1: Data Definition Audit — Quantify how many different names the same concept carries within your organization. Bain calls this a "data debt audit" and reports that in most manufacturers, 100-300 concepts each carry 3-5 different names.

Step 2: Rebuild the asset hierarchy on ISA-95 — Map the Enterprise/Site/Area/Work Center/Work Unit hierarchy onto your organization's assets exactly as the standard defines it. This alone becomes the backbone for every standard adopted afterward.

Step 3: Adopt OPC UA + industry-specific Companion Specs — Require standards compliance for new assets first, and map existing assets through gateways.

Step 4: Prepare for data space connectivity — Introduce an IDS Connector for connecting to Catena-X or a regional data space.

Step 5: Regulatory mapping — Map DPP, CBAM, and Data Act requirements onto your current data model and fill the gaps.

Case 1 — Standardizing Heterogeneous Processes in Battery Cell Manufacturing (MDPI 2025)

"Developing a Concept for an OPC UA Standard to Improve Interoperability in Battery Cell Production," published in the July 2025 issue of the MDPI Technologies journal, covers in detail a standardization methodology for heterogeneous process chains such as battery cell manufacturing.

Battery cell manufacturing strings together processes that are physically and chemically completely different — coating, calendering, slitting, cell assembly, formation, and aging. Each process has used equipment from different vendors with different data representations, making the battery industry one of the hardest domains to standardize. The authors propose a methodology of "reuse and extension of existing OPC UA Companion Specifications" and report an experiment that represented six vendors' different coating machines in a single unified data model.

The result: the time to integrate a new vendor's equipment fell from an average of three months to two weeks, and a data pipeline satisfying DPP requirements (traceability at the individual battery cell level) came together naturally. This case shows the classic pattern of "the industry that suffered most from missing standardization gaining the most from standardization."

Case 2 — The Catena-X Automotive Data Space and the EU DPP

Catena-X began in 2021 with six founding partners — BMW, Mercedes-Benz, Volkswagen, SAP, Siemens, and ZF — and by 2025 has grown into Europe's largest industrial data space with more than 250 participating companies. It builds data flows on the IDS standard across the automotive industry's entire chain: OEMs, tier-1 suppliers, tier-2 suppliers, and raw material providers.

In April 2025, Catena-X released v1.1 of its onboarding guide for large enterprises. The document defines in detail the practical process by which a newly joining company deploys an IDS Connector and maps its own data to the Catena-X standard data models. Following the August 2025 collaboration announcement between Catena-X and the OPC Foundation, the integrated data flow of "OPC UA inside the company - Catena-X between companies" was standardized.

The most notable change is Digital Product Passport readiness. From 2027, the DPP becomes mandatory for all industrial and EV batteries sold in the EU, and Catena-X is the only completed reference architecture that supports it. The successful interoperability PoC between Catena-X and Japan IPA's Ouranos data space (March 2025) shows that this standard will become the foundation of global regulatory compliance.

How PlantPulse Approaches This

KOPENS PlantPulse covers every layer of industrial data standardization in a single platform. Data collected over 200+ industrial protocols is automatically normalized internally into an ISA-95 asset hierarchy model and automatically mapped to OPC UA Companion Specifications (ISA-95, PackML, umati, and others). If new equipment is standards-compliant, integration costs are close to zero; if it is not, the gateway layer converts it into the standard model.

For inter-company data sharing, PlantPulse supports IDS Connector integration. When publishing PlantPulse data to — or consuming data from — a data space such as Catena-X, the required usage policy handling, lineage tracking, and audit logging are handled automatically. Standard data schemas for Digital Product Passport requirements (individual battery cell history, CBAM carbon emissions, and so on) are provided as reference templates.

What sets PlantPulse apart is that governance is layered on top of standardization. Rather than merely storing and publishing standard data, the lineage of "who used which standard data, when, and for what purpose" is recorded from day one. Compliance with regulations such as the EU Data Act, DPP, and CBAM becomes the system's default behavior rather than a separate project.

Frequently Asked Questions (FAQ)

Q1. Is adopting ISA-95 alone enough? ISA-95 is a grammar, not a dictionary. ISA-95 alone cannot express concrete data such as "the vibration RMS of machine 4 on line 1." It becomes complete only when combined with OPC UA Companion Specifications (industry-specific dictionaries).

Q2. We already have SCADA and MES — do we have to rip everything out for standardization? No. Most organizations put in a "standardization gateway" layer that maps existing systems' data into the standard model. The incremental approach — requiring standards compliance from new assets first — works well in practice.

Q3. Is Catena-X only for the automotive industry? Catena-X started in automotive, but its underlying technology, IDS, is industry-agnostic. Data spaces for the battery, energy, chemical, and construction industries are expanding on top of IDS.

Q4. When should we start preparing for the Digital Product Passport (DPP)? Given that the mandate for batteries and electronics arrives in 2027, 2025-2026 is the preparation window. Because the DPP requires full lifecycle data from raw materials to recycling, preparing data standardization with supply chain partners takes at least 12-18 months.

Q5. Where do Korean manufacturers stand on standardization? According to IDC's 2025 survey, about 45% of large Korean enterprises have adopted ISA-95-based asset models, but only 18% have extended to OPC UA Companion Specs. Preparation for CBAM and DPP is still at an early stage.

Closing — Standardization Is Not a Project You Can Postpone

Manufacturing data standardization is an old conversation, but since 2025 it has become an entirely different problem. First, the EU Data Act has made standardization a legal obligation; second, Catena-X, the OPC Foundation, and the IDSA provide complete reference architectures; and third, AI adoption is suffering unrecoverable delays caused by the absence of standardization.

Standardization is not a project — it is a shift in organizational architecture. Set the grammar with ISA-95, standardize the dialects with OPC UA Companion Specs, unify product and quality data with STEP AP242 and QIF, and build cross-company data flows with Catena-X and IDS. What is new in 2025 is that these layers must be addressed simultaneously.

Manufacturing competitiveness over the next five years will be decided not by "how much data you have" but by "how much standardized data you have." The volume of data is already sufficient. What is missing is a common language that makes that data intelligible to other systems — and to AI. Standardization is that language.

Related resources: OPC Foundation "OPC 10030 UA Companion Specification for ISA-95 Common Object Model", MDPI Technologies 2025 "OPC UA Standard for Battery Cell Production Interoperability", Catena-X and OPC Foundation joint white paper, August 2025, "Joint Integration Architecture Approach", IDSA "Dataspace Protocol Standard", EU Data Act 2023/2854 (in force September 2025), Capvidia 2024 "QIF Definitive Guide", Hicron Software 2024 "ISA-95 & OMAC PackML Companion Specs", McKinsey Global Institute 2024 Manufacturing Data Fabric Report, IDC Korea 2025 CBAM Readiness Survey, Bain & Company 2025 "Data Debt in Manufacturing"