This article was printed in CONTROL's March 2009 edition.
By Dan Hebert, Senior Technical Editor
Implementing integration between plant control systems and higher level computing platforms such as ERP systems is more profitable now than ever thanks to the increased cost volatility of most plants’ two main inputs―energy and raw materials―and more price volatility for plant product outputs.
At the same time, integration has never been technically easier because of the spread of standards that define terms for data exchange among disparate computing systems. Because these technical integration hurdles are coming down, implementation costs are dropping rapidly.
Increased profitability coupled with lower costs should result in a flurry of control/ERP system integration projects, but this is not happening for a number of reasons. Chief among them are non-technical barriers to integration that remain the same or higher than ever―security concerns, reduced plant technical staff and, especially, turf wars between process automation professionals and their IT counterparts.
Price Volatility Increases
Process plants turn raw material and energy inputs into finished product outputs. Some plants were designed to minimize raw material use; other plants are better when it comes to using energy efficiently. Some plants were designed to produce very few products with a high degree of efficiency. Other plants excel at producing a wide range of products, albeit at lower efficiency.
For a single plant, profits are maximized by producing highest margin products. To do this, one needs to know the cost of raw material inputs, the cost of energy inputs and the selling price for plant outputs. For firms that have multiple plants all capable of producing the same range of products, optimization for maximum profitability becomes more complex, but the problem is essentially the same. One must still produce the highest margin products using the least possible amount of raw materials and energy.
Basic plant input/output equations haven’t changed much in decades, but volatility of input costs and output prices has increased dramatically in the past year. For example, the price of oil peaked last summer at about $145 per barrel. Per barrel prices then dropped rapidly to about $35 per barrel by December of last year. Other commodity prices also experienced unprecedented rapid changes over that same time period, and there is no reason to expect volatility to decrease.
The story is much the same for energy prices. Electricity and natural gas prices move up and down rapidly, and tariffs from utilities to process plants increasingly reflect market reality. Many utilities used to change tariffs annually, but industry leaders now use locational marginal prices (LMP).
“With LMP, the utility fixes the price every five minutes, depending upon grid congestion,” says Paul Kurchina, the director of industry consultant KurMeta Inc. “LMP requires new billing objects in the ERP system, more intense monitoring in facility control systems and closer communications between control and ERP systems.”
For many process plants, volatility is also increasing in terms of prices paid by their customers for their outputs. If your process facility is a power plant, it’s easy to see how LMP constantly changes prices for plant outputs. A refinery doesn’t see price changes for its outputs every five minutes, but prices can change substantially over the course of a few days.
Some process plants see much less volatility in prices for their finished products. Food, beverage, pharmaceutical and consumer packaged goods firms are good examples. But while these plants see slower changes in output prices, they still often experience rapid changes in costs for raw material and energy inputs.
How do these factors increase the profitability of control system/ERP integration?
In the old days of stable prices, it was a lot easier to optimize plant production. One just had to enter monthly or even annual cost data for inputs and make similar data entries for prices of outputs. Plant production schedules could then be optimized based on these data.
But in today’s era of volatile prices, monthly data entry is not good enough. Instead, data must be changed dynamically and in real time, and this means that electronic links are needed between control and ERP systems.
The bad news is that your plant and your company cannot remain competitive without real-time data exchange between your control system and your ERP system. The good news is that this exchange is technically easier than ever to establish and maintain.
Standards Ease Integration
The bad old days of custom coding for communication between control systems and ERP systems are largely over. A number of industry-wide data exchange standards have arisen over the last few years, and just about every vendor of note is complying with one or more of them.
For end users, these standards save time and money. “We use OPC to communicate between our Oracle database ERP and our DeltaV control system,” says Dan Cox, director of engineering for AOC Resins, Collierville, Tenn.
“We use an OPC tool developed by Matrikon called Generic Database Access (GDA). The GDA tool turns user-defined Oracle variables into OPC parameters. We wrote Oracle stored procedures that complete a function as needed. For example, via an OPC mirror, we insert the amount of a material used in a certain batch to an Oracle table that stores batch material usage for costing,” explains Cox.
If OPC were not an established standard, the alternative for AOC Resins would have been to develop custom .Net programming. “This custom work would have been difficult to scope, and the development would have been much more costly,” Cox believes.