"INTRODUCTION The term process control implies that there is a process for which there is a desired behavior and that there is some controlling function that acts to elicit that desired behavior. This broad concept can embrace everything from societal processes governed by some regulatory control authority to automated manufacturing processes. In practically all cases@ however@ a common thread is that some measure of the actual process behavior is compared with the desired process behavior. This feedback action then generates a control policy that acts to minimize or eliminate the deviation between desired and actual behavior. We are concerned in this book with a particular segment of automated process control??that which is applied to chemical@ refining@ pulp and paper@ power generation@ and similar types of processes. Even within this limited scope of applications@ we will limit the discussion primarily to processes that are operated continuously for long periods of time and within a narrow region of the operating variables. In other words@ we exclude such important operating modes as batch processing@ start-ups@ and grade changes. Many of the control techniques to be presented here@ however@ can be adapted to these other modes of operation. For the processes we focus on in this book@ the process's behavior is often characterized by measured values of such process variables as temperatures@ flow rates@ pressures@ and the like. The desired behavior@ then@ is stated to be the set points of those process variables. Until fairly recent times@ most applications of industrial process control used simple feedback controllers that regulated the flows@ temperatures@ and pressures. These controllers required a form of adjustment called tuning to match their controlling action to the unique requirements of individual processes. Occasionally@ more advanced forms of control@ such as ratio and cascade@ could be found; even more rarely one might find a feedforward control loop. As long as most of the control systems were implemented with analog hardware@ applications were limited to simple regulatory control. This was due to the cost of additional components@ the additional interconnections more advanced control required@ the burden of maintenance@ and the vulnerability to failure of many devices in the control loop. With the advent of digital control systems@ however@ more sophisticated loops became feasible. Advanced regulatory control??which includes the previously mentioned ratio@ cascade@ and feedforward control as well as additional forms such as constraint (selector) control and decoupling??could readily be implemented simply by configuring software function blocks. With this additional capability@ however@ a need developed for a systematic approach toward using it. This is called control strategy design. In order to design a technically successful and economically viable control strategy@ the control system engineer must be well grounded in the techniques of feedback control as well as the tools of advanced regulatory control. The requisite knowledge includes both how to implement and how to tune. Even before that@ however@ the control system engineer must be adept at recognizing when to use (and conversely@ when not to use) certain control methods as well as in projecting the expected benefits. Using advanced regulatory control provides many benefits. One of the most important is simply closer control of the process. It will become very clear later in this book that with basic regulatory (i.e.@ feedback) control@ there must be a deviation from set point before control action can occur. We will call this the ""feedback penalty."" The objective of advanced regulatory control is for the control action to be taken by incurring only a minimal feedback penalty. The reduction in feedback penalty may be stated in a variety of ways@ such as a reduction of the maximum deviation from set point@ as a reduction of the standard deviation@ or simply as a reduction in the amount of off-spec product produced. This reduction in feedback penalty can provide several forms of economic benefit@ such as improvement in product quality@ energy savings@ increased throughput@ or longer equipment life."