ABSTRACT
Advancements in automation and distributed control systems make possible the collection of large quantities of data. But without the corresponding adequate tools, it is not possible to interpret the data. Every modern industrial site believes that this data bank is a gold mine of information if only the important and relevant information could be extracted painlessly and quickly. Timely interpretation of data would improve quality and safety, reduce waste, and improve business profits. This interpretation is possible except for the following dilemmas: undetected sensor failures, uncalibrated and misplaced sensors, lack of integrity of the data historian and of data compression techniques used to store the data, and transcription errors. It is no wonder that data analysis methods in the face of these serious problems may appear to be inadequate. Meanwhile, the data bank continues to grow without appearing to garner any useful information. Without accurate and timely measurements, feedback control of the process to some specified objective is very difficult, if not impossible. For example, in the chemical industry, composition measurements are usually not made online; rather they are sampled and analyzed off-line. The delay between the samples and the results are usually on the order of hours. Thus, timely information about the purity of the composition is unavailable to take remedial control action.
