Monitoring process performance is a critical requirement in any manufacturing process as producing quality products within specification reproducibly is a prerequisite of an economically viable process. Without effective monitoring and control strategy, as key requisite, a capable manufacturing process could not be successful.

Monitoring is essential for various aspects of the control strategy - the quality of raw materials is usually tested on intake, process equipment often has to be rigorously qualified (e.g., in the highly regulated pharmaceutical or food industries), environment is controlled by implementing manufacturing-area classification where relevant, waste is treated prior to release and the quality of the final product is tested before release. Initiatives, such as quality by design (QbD) and a supporting enabling technology of process analytical technology (PAT) championed by the US Food and Drug Administration (FDA) in the pharmaceutical industry, aim to shift the focus for manufacturing from end-product quality testing to building the quality in the process.

Tightened Quality Control

Such a shift in emphasis would not be possible without reliable and effective monitoring. Indeed PAT has been defined as ''a system for designing, analyzing, and controlling manufacturing through timely measurements (that is, during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product quality.'' Traditional process control strategies based upon information from laboratory assays and supervisory computer systems (SCADA) are routinely used to regulate process operation and correct for disturbances resulting from raw material variations through to production plant variations. If PAT can provide additional information on disturbances and deviations, giving greater plant insight, then the effects of disturbances can be reduced and quality control tightened. However, greater benefits are to be gained by the systematic use of PAT tools in process development to increase fundamental understanding and more robust definition of the design and control space of the process operation.

Food Safety

An analogy in the food industry in terms of the importance of effective monitoring procedures can be seen in the hazard analysis critical control point (HACCP) food safety standard, which is now widely incorporated into national food safety legislation of many countries. The seven basic principles of HACCP implementation consist of:

1.       Conduct hazard analysis, considering all ingredients, processing steps, handling procedures, and other activities involved in a foodstuff's production

2.       Identify critical control points (CCPs)

3.       Define critical limits for ensuring the control of each CCP

4.       Establish monitoring procedures to determine if critical limits have been exceeded and define procedure(s) for maintaining control

5.       Define corrective actions to be taken if control is lost (i.e., monitoring indicates that critical limits have been exceeded)

6.       Establish effective documentation and record-keeping procedures for developed HACCP procedure

7.       Establish verification procedures for routinely assessing the effectiveness of the HACCP procedure, once implemented

Clearly effective monitoring is critical to ensuring product quality regardless of the type of manufacturing industry. Essential components of effective monitoring include representative measurement and a robust representation of the obtained information, allowing appropriate action to be taken.

Sensor Specification

A complete review of specific process instrumentation for critical parameter measurement is beyond the scope of this section, and the emphasis will be placed on the characteristics of measurements to be used in a critical parameter control scheme. These characteristics raise important questions that must be answered prior to sensor specification and they lead to the establishment of specific protocols that need to be followed during sensor use. Such characteristics would be equally applicable to established as well as emerging PAT measurement methodologies. The key considerations for a sensor are:

Accuracy and Resolution. A useful sensor provides measurement at an appropriate accuracy for the control task. If, for example, a temperature is to be controlled in the range of ±0.1°C then the measurement must be significantly more accurate than that. If that was not the case, the actual process may be subject to larger deviations, although it may appear that the process is controlled within this range.

Precision is the probability of obtaining the same value with repeated measurements on the same system and it is particularly important in the longer term operations. For instance, sensor drift from calibration can cause deterioration in system performance because the desired values are not achieved. Drift is often inevitable, so it is important to know the rates of likely drift so that recalibration can be performed as necessary.

Sensitivity is defined as the ratio between the sensor output change ΔS and the given change in the measured variable Δm (sensitivity S=ΔS/Δm). If the critical control parameter value changes, it is important that the sensor responds to such a change.

Reliability. Sensors provide information which is acted upon either by process operators in a ''human in the loop'' control scheme or directly by closed-loop control schemes. When operators use the information, there is some opportunity for human interpretation of the results. Failed sensors are more difficult to detect in a hardware-based closed-loop scheme. If the information is essential and a sensor fails, then implications on operation can be severe. Reliability is a function of the failure rate, of the failure type, ease of maintenance and repair, and physical robustness. Redundancy and planned maintenance programs to maintain the sensors are required to maintain reliability.

Response time is defined as the time required for a sensor output to change from its previous state to a final settled value within a tolerance band of the correct new value. The dynamic sensor characteristics are important as the sensor must respond significantly faster than the process. If a sensor has a long response time it may indicate an ''average'' value rather than the actual process value.

Practicality. The environment within a process may be particularly demanding-for instance, the sensors may be exposed to high temperatures or pressures. Whilst a sensor may in theory measure the variable of interest in ideal conditions, the range of the operational environment could render it incapable of functioning or may influence reliability.

Cost. Sophisticated instrumentation is now available for process monitoring with PAT, but the price can be high. However, the benefits gained can be significant if sensor information leads to raw material/resource savings or increases productivity. A cost benefit analysis should be performed to assess whether the instrumentation is appropriate.

A significant issue to be addressed in effective monitoring is the placement of a sensor as it influences the frequency of available measurements. Theory dictates that for a measurement to be of value it must be sampled above a certain minimum frequency. Often instruments are used on-line (say temperature or pH) or they can be multiplexed to save cost, but the frequency of information supply is limited because the instruments must serve several vessels (e.g., mass spectrometer measurements). However, it is off-line sample analysis where problems with low frequency measurement are most likely to arise.

Initiatives such as PAT lead to increased use of sophisticated sensor technology, such as near infrared spectroscopy (NIR), which requires more powerful data interpretation and monitoring tools.

Authors:

Krist V. Gernaey, Technical University of Denmark, Department of Chemical and Biochemical Engineering, Lyngby, Denmark

Jarka Glassey, Newcastle University, Faculty of Science, Agriculture and Engineering, Newcastle upon Tyne, United Kingdom

Sigurd Skogestad, Norwegian University of Science and Technology, Department of Chemical Engineering, Trondheim, Norway

Stefan Krämer, Ineos, Köln, Germany

Andreas Weiß, Ineos, Köln, Germany

Sebastian Engell, Technical University of Dortmund, Department of Chemical Engineering, Dortmund, Germany

Efstratios N. Pistikopoulos, Imperial College London, Department of Chemical Engineering, London, United Kingdom

David B. Cameron, IBM Global Business Services, Kolbotn, Norway

Read more about this topic in Ullmann's

This article is an excerpt from the Ullmann's Encyclopedia of Industrial Chemistry which celebrates its 100^th anniversary in 2014. More about the topic can be found in the encyclopedia article on Process Systems Engineering, 5. Process Dynamics, Control, Monitoring, and Identification. More concept articles on general interest topics in industrial chemistry and chemical engineering can be found on the Ullmann's Academy homepage!