You'll find out how important something is when you no longer have it. No one complains about getting good data, but as soon as they lose the information someone – usually you will hear about it. The same applies to controls. The difference between pressing a button remotely to perform some task vs driving out to a remote site and doing the task manually gets more expensive and irritating the longer you go without the function and the more times you have to repeat the task. Think about that when you are looking at cutting initial design costs for a system. Saving a few bucks on a underpowered and under designed system early in the game can bite you down the road. Build it right the first time and save yourself headaches and costs associated with operational failures. This may require hardening a system against environmental effects, designing for future workload requirements, and proper design testing to avoid any pitfalls you may not initially be aware of. Look at identifying single points of failure. Redundancy after a single point of failure won't improve anything if the failure point breaks in front of your redundant mechanism. Have a decent back up plan and proper maintenance schedule. Eliminate any bottle necks that will negatively impact operation and response times. Most of what you run into in the SCADA world depends on a combination of hardware and software. You should already be in the habit of bench testing off line anything you plan on putting into active service . You don't want to find out you made a mistake after the ship has set sail. Shoot for perfection each step of the way. Remember that a chain is only as strong as it's weakest link. Avoid building weak links, and never settle for “good enough.” It doesn't matter whether I am testing a simple Bell 202 point to point data link, a frequency hopping network, frame relay mesh, or an industrial Ethernet. Each system is important, and each design has it's own strengths and weaknesses that must be handled properly. You need to have a clear picture of what you are setting out to accomplish. Unfocused concepts, or designing on the fly may work in some limited capacity such as emergency situations, but nothing beats having a well thought out plan you can prove by testing.
