Beyond the use of commercially available automation systems, the unique mix of products, equipment, software and the way they are adapted to the specific requirements becomes proprietary "embedded" knowledge.

Monitoring and control systems operate through the correlation, aggregation, display and adjustment of diverse information: set-points and adjustments, desired outputs, review of real-time and historical data, monitoring of failures, minimizing down-time and the like.

System complexity tends to increase in the continuous drive for better efficiency and productivity, as automation systems pursue increased effectiveness through monitoring and control of increasing numbers of input/output (I/O) points. The operator interfaces become increasingly complex, so that more and more operator expertise and training and are demanded.

Where does one find enough good operators to train, and how long does it take to train them? Improved effectiveness comes not from training the operator to use increasingly complex systems, but from developing systems that adapt effectively to maximize throughput with a minimum of operator involvement.

For example, operators can acknowledge and attempt to correct just a few alarms. But, under really adverse conditions, human intervention is ineffective, and the control system must be self-correcting and self-optimizing - which means that the system must adapt heuristically to reduce, not increase, the need for operators.

Beyond just displaying measurements, trends and alarms, effective systems demand diagnostics. Traditionally, this has been failure analysis - after a problem has occurred. But improved effectiveness comes from diagnostics that is preventive (to avoid failure) and predictive (warnings about future failures).

Optimizing controls supplement conventional distributed control systems (DCS) and programmable logic controllers (PLCs) to maximize profitability of plant or factory operations. There are two optimization approaches: historical review to determine what mix of variables and control strategies resulted in peak performance; and, extrapolating, calculating and predicting the optimum mix of variables and controls that will provide further improvements.

Consider this: What if the entire plant or factory must be duplicated in a completely different location? The conventional approach would be to try replicating the entire plant - which is virtually impossible. That involves not just the purchase and installation of the same control systems and software, but including the accumulated variations and software patches and modifications, control strategies, diagnostics and more. And there may be different requirements in the new location that demand variations to optimize for local materials and changed local needs. The system must adapt, and the adaptations must be tracked; it's like keeping track of a living, changing entity.

The control system genome

Recording, or to follow the analogy, "sequencing," the control system genome involves tracking many different instrumentation, measurement and control systems purchased from several different suppliers - large automation majors as well as smaller independent equipment suppliers and systems integrators. For this reason, system genome tracking is something that cannot be provided by just one of the primary suppliers (who typically can only optimize for their own equipment). It must be provided by an independent company that specializes in tracking the performance and adaptations of multi-vendor systems, and is dedicated to process optimization through genome tracking. PAS is a pioneer in this fast-developing market.