Semantic Interoperability Drivers
There’s growing interest in semantic interoperability in manufacturing because it:

• Improves collaboration: Manufacturing organizations are looking for innovative ways to improve their collaboration with their suppliers and business partners.
• Reduces integration costs: Manufacturing organizations are actively seeking new ways to reduce integration costs.
• Increases business agility: Today, manufacturers are under tremendous stress and have never seen the shifts that are happening. As a result, business agility continues to be on the minds of global executives as a way to stay ahead of the curve or simply survive. SOA holds the key to this business agility.

Semantic Interoperability Challenges
To achieve manufacturing interoperability and information sharing, manufacturers have to overcome the following challenges:

• Semantics: Semantics is related to the understanding and integrity of the information and requires an agreed business language. Semantics also requires collaboration across organizations (in the enterprise) and across enterprises (with the suppliers). Achieving consensus on meaning is the most difficult challenge. Agreement on semantics and syntax is difficult to achieve because of:
  
  • Perceived loss of control (resistance to change): Reluctance to give up one’s view of the world (process and data)
    
  • Lack of incentives to cooperate and collaborate: What’s in it for me?
    
  • Cost: Lack of a program budget for activities outside the program or the project.

• Standards: There are too many overlapping standards supporting manufacturing interoperability such as OAGi, ISA95, MIMOSA, OPC, WBF, ACES, PIES, and ISO 10303 (STEP). These standards cover the interfaces, messages, and documents, but don’t cover the business processes. Many of these data interchange standards are adopting XML as the basis for specifying their data content standards, which can be used to tag collections of data with labels. As part of the standardization activity, communities can agree on the names for these labels. An interoperability problem remains, however, if different people have a different understanding of the meaning of these labels. In other words, XML standardizes the syntax of data exchange, but wasn’t designed to capture the semantics of the data. This limitation isn’t an issue if used in a common context, but it becomes a problem when moving data from one context to another, for example, sending data from a manufacturing context to a financial context. Without an explicit and rigorous definition of terms, a misunderstanding is inevitable.
• Globalization: Globalization is the major trend in manufacturing today – globalization of markets and globalization of partners. The globalization of markets means that companies want to design anywhere, manufacture anywhere, and sell their products anywhere. The globalization of partners means that supply chain members are located anywhere and do business anywhere with the manufacturing. Both have led to an explosion in the amount of information sharing that must take place. It’s absolutely critical to the success of companies and their suppliers that this sharing is done correctly, efficiently, and inexpensively.
• People: Changes in technology are impacting the way in which information sharing takes place. Nevertheless, people still provide the bulk of the understanding needed to determine what the information means and most of the tacit knowledge needed to make decisions based on that understanding.

Manufacturing Semantic Interoperability Framework
To overcome some of the semantic interoperability issues, manufacturing should create a semantic interoperability framework. This framework is a strategy for achieving a common view of information. It should cover what can be shared for cost and security reasons. Every enterprise has many areas that can potentially benefit from semantic interoperable systems and more cost-effective integration. Interoperability is achieved by implementing standards, and semantics is achieved by implementing ontology.

The process of creating this framework consists of:

• Identifying the areas in manufacturing that can benefit from applying the framework.
• Identifying the interoperability standards that can be applied to these different areas.

To identify the different areas in manufacturing that can benefit from applying an interoperability framework, we should start off with the touch points between the different manufacturing organizations, their partners, and the kind of information they share such as product development, manufacturing, quality, and supply chain management. They also share and exchange information with suppliers, customers, and aftermarket services providers.

Figure 1 illustrates manufacturing entities, the type of information exchanged by them, and touch points with external entities.

Interoperability Standards
Manufacturing collaborates and interfaces with a number of external organizations to produce its products. It collaborates with suppliers, dealers, and aftermarket service providers and retailers. It shares different information with these entities such as technical (engineering), product and parts, and aftermarket such as warranty information. Usually, in some large supply chains, a dominant OEM will mandate that supply chain partners conform to a particular proprietary solution. This has been the practice, for example, in the automotive sector. The problem with this approach is that the interoperability problems are simply pushed lower down the supply chain – they’re not eliminated. First, subtier suppliers are forced to buy and maintain multiple, redundant systems if they want to do business with several major OEMs.

Published standards offer some stability in representing information and help in sharing information consistently. To identify the different interoperability standards, we need to take a close look at the product life cycle, the collaboration model, and the type of information shared at different phases of production with different entities based on the collaboration model. For example, a product goes through a design phase first. At this point, the manufacturer collaborates with its suppliers by exchanging technical information about the design and the different technical specs of the product and its parts. The standard available here for exchanging technical information is ISO 10303 (STEP).

Figure 2 illustrates a product life cycle, the collaboration entities, the information shared, and the available interoperability standards.

In general, manufacturing standards can be divided into two kinds:

1. Business-to-Business interoperability standards (B2B)
2. Plant-to-Business interoperability standards (P2B)

B2B Interoperability Standards
Below is an overview of the manufacturing standards available:

• STEP: The Standard for the Exchange of Product Model Data is a comprehensive ISO standard (ISO 10303) that describes how to represent and exchange digital product information. Nearly every major CAD/CAM system now includes a module to read and write data defined by one of the STEP Application Protocols (APs). STEP APs support roughly 40 kinds of information exchange in an effort to support the Model Data during its entire life cycle, from concept design to final disposal. In the U.S. the most commonly implemented protocol is called AP-203 (Configuration Controlled Design). It’s used to exchange data describing designs represented as solid models and assemblies of solid models. In Europe a very similar protocol called AP-214 (Core Data for Automotive Mechanical Design Processes) does the same thing.