Sharing data in the chemical process industry

Citation

(1998), "Sharing data in the chemical process industry", Assembly Automation, Vol. 18 No. 1. https://doi.org/10.1108/aa.1998.03318aaf.002

Sharing data in the chemical process industry

Sharing data in the chemical process industry

Unlike car and aerospace manufacturers, the chemical process industries (CPI) have yet to capitalize on the power of computer networks and their inherent ability to share data. Programmes require customized interfaces to exchange data. This forces widespread manual translation of data, introducing errors that propagate with every reuse, increasing costs. The price of such inefficiency is high. Overall, up to 10 per cent of a process plant's lifetime costs and 40 per cent of its engineering effort, could be cut by more rigorous data management.

Help is finally on the way, in the form of a series of data interchange rules being developed by the International Standard Organization (ISO, Geneva) called the Standard for Exchange of Product Model Data (STEP). Evolving as part of the ISO 10303 series, the first of three CPI-specific application protocols (APs) will be published as a draft international standard (DIS).

The APs within STEP will allow a sector within the CPI to define the data model used to exchange information between computer applications. With plant equipment, for example, such information will include part number, version, security classification, tolerance specifications, and shape. Once entered in STEP format, the design specifications, physical equipment details, and operating and maintenance data can become part of a plant database. For most users, the goal is to establish a data warehouse allowing data integration and sharing, and lifetime data storage.

Using an application programme interface (API), software can then access and update the data warehouse throughout the plant's lifetime. STEP and APs are analogous to translators that allow people who speak different languages to communicate, says Hans Teijgler, engineering manager at Floor Daniel BV, Haarlem, The Netherlands.

Once STEP is completed, the process industries will finally realize the benefits of electronic data management that have become standard in the car and aerospace industries. For the CPI, relevant parts of the ISO 10303 series are: AP 221, covering schematics and process plant design; AP 227 for 3-D CAD; and AP 231, for conceptual design, emphasizing process simulation. AP 227, which was developed primarily by PlantStep Inc. (Wilmington, Delaware) is furthest advanced. A Draft International Standard (DIS) is now expected. Once the DIS is issued, software vendors will be able to provide product interfaces.

For engineering contractors and process plant operators, the vision is an electronic handover of a plant, where the current paper-based design data, equipment data sheets and performance specifications are passed in database form to the client. Now, only parts of projects, such as equipment data sheets, are handed over electronically. Within three years, Teijgler expects most major CPI firms to specify STEP-based databases for alphanumeric data. Transfer of full 3-D models and databases is still ten years away, he says. To accelerate the development of STEP-compliant data warehouses and commercial software, a group of major operating companies in Europe, including ICI and BP in the UK and Framatome in France, has won funding for a two-year project from the European Union. The project, Pippin, was scheduled for completion in December, 1997.

The Pippin data warehouse will permit file exchange, online sharing of data via an API and data exchange with external systems via an API. Data stored in the warehouse may come from different sources and be based on different data models. To allow this, both end-user data and the data model information will be stored in neutral format in the warehouse. On completion, the project will deliver the tools to import and export data.

STEP's first large-scale industrial application will be the Eastern Trough Area Project (ETAP), an offshore oil-field in the North Sea, which was to be commissioned in 1997 by a seven-company consortium headed by BP Int. Ltd (London), Brown & Root (B&R, Houston). The main engineering contractor on the project, uses 3-D CAD packages from Cad Centre and Intergraph; a drafting package from Matra Datavision, I + C = S's Speid and Rapid for schematics, process and instrumentation diagrams (P&IDs) and flow-sheets; and the company's Isogen and FICS packages for piping design and spool production.

By allowing design data on 100,000 tagged items from P&IDs to be automatically transferred to the project's maintenance database, STEP will cut database creation costs from about $160,000 to less than $16,000, BP claims. Over the project's 25-year lifetime, BP expects STEP to save $256 million, about 10 per cent of capital costs.

For ETAP, Brown & Root is using equipment and activity class libraries developed by the Norwegian STEP group, POSC/Caesar. Over 1,200 new classes had to be defined for the project.

Automating these class definitions is one of the goals of a parallel effort taking place in the USA, the Common Industry Material Identification Standards (CIMIS) data centre, which opened recently in Houston under the auspices of the American Petroleum Institute. CIMIS is targeting commodity materials such as pipes, fittings and fasteners, that are widely used in CPI and oil-well structures. A knowledge base and expert system have been developed to generate unique descriptions in a common format that designers, fabricators and plant owners can use. Cimis is monitoring the POSC/Caesar effort, and will strive to develop a compatible system as that standard takes shape.

Within engineering companies, STEP will enable concurrent engineering to take place globally and between different contractors and suppliers. The starting point for a project is the production of P&IDs that describe the plant in schematic terms. These can be produced using any one of a number of standard CAD programmes. The drawings are then modified to allow data to be extracted from the diagrams for use by the database and as the basis for concurrent engineering. The data model used in Engines formed part of the basis for work by SPI/NL and PiSTEP, the developers of AP 221.