A Bit of History
In An Introduction to ISO 15926 we follow the development of project information exchange from the advent of Computer Aided Drafting (CAD) to the present. We look at some of the detours that were taken along the way and show how the idea of a reference data library (RDL), Part 4, and a separate data model, Part 2, came about. We do not intend to make readers into amateur historians, but a brief look at where ISO 15926 comes from will put the standard into a better perspective
For a summary here we will divide the time line into four periods.
If Only We Could Exchange CAD Drawings
Soon after the advent of CAD in engineering offices in the late 1970s, engineers started wondering why they couldn't just open a drawing authored with one program with their own system. With a large end user (the US Department of Defense) pounding on the table, the industry put their heads together and within a few years the Initial Graphics Exchange System (IGES) was created.
Although all reports about IGES are not uniformly complimentary, within a certain scope it performed as designed, allowing users of different systems to exchange drawings. As well, an organization could archive its drawings with IGES and open them years later with a different system.
A major limitation to simple CAD exchange surfaced in the manufacturing industry. For an automated manufacturing system, there is often more to drawings than the graphical information. Some drawings are essentially a graphic user interface to a computer controlled manufacturing process. For instance, what appears to be a circular element on a drawing could well drive a machine that selects the correct drill bit and drills the hole. In such a drawing the graphical elements are only a small portion of the value, yet when the drawing is exchanged using IGES, only the graphical elements remain.
If Only We All Used the Same Data Model
The limitations of IGES (and other CAD exchange systems around the world) prompted the development of the Standard for Exchange of Product Information (STEP), which was adopted by the International Organization for Standardization (ISO) in the late 1980s as ISO 10303. Using ISO 10303, engineers could exchange full product models (including the materials, finish, fabrication details, and, if it were important, packaging) without loosing any information.
This was accomplished by creating a comprehensive data model, which included all of the details of a product. Software vendors created export and import functions which converted their proprietary data models into the STEP format. Once a design had been converted into the intermediate, or "neutral" format, any other STEP-compliant software could read it.
Under the STEP protocol, every industry was to create its own Application Protocol (AP). Today AP 203 and AP214 are widely used in the aerospace and automotive industries, and indeed in any industry that manufactures parts.
With the success of the STEP approach to manufacturing, the process industry caught on and attempted to create APs for process plants. AP-221 was created for schematic, or 2D, plant information, and AP-227 was created for spatial, or 3D, plant information. Both were tested with pilot projects, which reported great success. In practical use, however, the STEP approach proved to be unwieldy. Managing a comprehensive data model over the relatively short-lifetime of a manufactured part worked well, but over the several-decade life of a typical process plant, doing so was impractical.
If Only We All Used the Same Reference Data
By the early 1990s the thinking had changed from the large, comprehensive data model of STEP into a simpler, generic data model with a large part of the intelligence pushed into a Reference Data Library (RDL). With a common set of definitions the data model would not be as cumbersome.
This approach was used on several world-scale projects in the mid to late 1990s. From this, the data model became ISO 15926 Part 2, and the Reference Data Library became ISO 15926 Part 4.
The RDL of Part 4 matured fairly quickly, becoming the basis of what we might call a small "Plant Model Exchange" industry. For a few years this was the only real-world use of ISO 15926, so this approach became synonymous with ISO 15926.
If Only We Could Embed Meaning Into Our Information Exchanges
Meanwhile, research continued to developed practical means of using the generic data model of Part 2. What we wanted is to be able to do is embed the meaning of our data within the data itself. If the data were self-describing we would not have to spend such a large effort to exchange the data. The idea of Templates, analogous to phrase books for a foreign language, was embodied in ISO 15926 Part 7 to enable a wider use of Part 2.
Today parts of ISO 15926 are in use in many production environments around the world, and practical use of Part 2 has been demonstrated, as we will see shortly.