IDENTIFYING, MEASURING AND ANALYZING KEY PROCESS VARIABLES

The identification and measurement of key process variables is an important element of quality improvement. One of the major objectives of the Quality Department is to ensure that products are fit for the intended use, do not pose any form of hazard to the end users and comply with its marketing authorizations. This places great emphasis on user needs such that the point of departure for systematic quality work and for deducing quality indicators is the user needs. The user demands products of the best quality but producers must consider the cost or efficiency of processes. A precondition for improvement of processes to yield cost efficiency is the study of such processes, which is done through identification, measurement and analysis of key process variables (KPV).  

Identification of KPV has 3 steps 

• Identify critical product characteristics, Develop a process flow chart and Determine KPVs 

Measuring KPV simply means evaluating the measurement capability of the procedure/tool/system that has been deployed to measure KPVs 

Analysis of KPVs has 2 steps 

• Determine the stability of critical processes and Determine system capability.  

Finally, there is need to establish a system for continuous monitoring of processes.  

IDENTIFICATION OF KPV  

1.  IDENTIFY CRITICAL PRODUCT CHARACTERISTICS  

 Customer Satisfaction Survey and/or Consumer Complaint Register are valuable tools that can be used to identify critical product characteristics. It is important to know the quality attributes end users seek for in a product as well as knowing the definition of the quality of a product from customer’s perspectives. This will help note the strengths and weaknesses of a product and to monitor them for improvement. It must be borne in mind, however, that due to a wide variety of consumer tastes it is difficult to establish accurate target levels of product quality in consumer terms. A producer must set it’s standards of quality and strive hard to not just attain but also to improve on them.  

2.  DEVELOP A PROCESS FLOW CHART 

Once a process has been identified for monitoring, the next step is to map the process by developing a detailed and comprehensive flow chart that can be used to identify sources of variations in the process. In some cases it may be necessary to carry out on site confirmation of the process flow chart so as not to leave anything to chance. Such process charts must have these 3 components 

1. The sequence of processes must be delineated indicating decision points of the flow process i.e. what goes in and out of the process must be identified.  

2. The persons responsible for each process are identified. It’s important to know who executes the process and who they interact with.  

3. The actions to be taken by those responsible for the process i.e. noting the variables and their controls.  

This task can be carried out by a cGMP audit team which every company should have.  

3. DETERMINE KPVs 

KPVs are those factors that vary every time a process is repeated and could have an effect on product characteristics. KPVs should not be confused with quality indicators, which are measures that give indications of output quality. All processes have numerous variables,  so it’s important to identify the most relevant ones e. g. temperature, volume/weight, mixing time etc. Note that relevant KPVs are those variables which changes in, no matter how little, will have the greatest impact on product quality. For instance, temperature and time of pasteurization of milk is a relevant KPV.  

A good tool that can be used to identify KPV is the Pareto diagram. It was designed for the identification of KPVs when numeric information on the variable is not available. It is also called cause and effect or fishbone diagram. The review team picks out the causes that have the greatest effect on the process and these are labelled KPVs.  

EVALUATE MEASUREMENT CAPABILITY 

KPVs must be accurately measured so it is crucial to know how KPVs are to be measured. If there are errors in the measurement of KPVs, results will be unreliable and may invalidate any analysis of process stability or capability. Evaluating measurement capability is therefore critical for process improvement. Two commonly used measurement systems are 

• Customer Satisfaction Survey: this has been discussed earlier. It is a poor system of measurement and the errors therein limit it’s use for continuous improvement of quality. Another limitation of this method is that it increases the cost of poor quality because by the time feed back is coming from the customers they may have formed an impression that the product is of poor quality.  
• Sampling and Verification: this is more reliable. It involves the use of a spot check control form by supervisors who check samples at regular intervals. This is the technique commonly used in assessing product quality. It can also be called Line Inspection where QA personnel do acceptance sampling: sampling products and checking them against acceptance limits with a view to reject if below limit or accept if in line with limit. This approach reduces the cost of poor quality as Quality indicators are checked and corrected if defects are noticed.  

ANALYSING KPVs 

1.  DETERMINE STABILITY OF CRITICAL PROCESSES 

A process control variable is termed stable if it displays a consistent variability about a constant mean. Nonetheless, that a process is stable does not imply it is operating well enough to meet requirements. Stability only means that no unusual process variation are being recorded.  

The tools that can be used to determine process stability are Pareto charts and control charts (pie chart, bar chart, histograms). To use these tools, lower control limits (LCL), control limits (CL) and upper control limits (UCL) must be set for products. A tomato sachet with a net weight of 70gm may have LCL of 68gm and UCL of 72gm. As long as sachets meet this standard the process is said to be under control.  

There are special and common causes of variation. Distinction has to be made between the two types of variations. Special cause variations are one-off events that are corrected whilst common cause variations affect all processes and may require system change.  

2. DETERMINE SYSTEM STABILITY  

For a process that is under control,  there is need to determine the capability of the system to maintain control of process over time. How long it takes the system to produce variations that are below the LCL or above the UCL must be determined. This can be achieved by analyzing data generated while process is running.  

ESTABLISH A SYSTEM FOR CONTINUOUS MONITORING OF PROCESSES  

That stable and capable processes have been set up is just half the job. Continuous monitoring is needed to provide information on feedback, control process variations and improvement. Monitoring in itself will not bring continuous improvements. Procedures and plans must be put in place for managing the implementation of improvements identified. Internal cGMP audit teams can plan alongside QA Department to set up effective procedures of monitoring and executing improvements that best suit the facility.  

BENEFITS OF PROCESS CONTROL MONITORING  

Some of the benefits of systematically monitoring processes are as follows  

1. To find cost-efficient ways of improving product quality  

2. It gives Managers the opportunity to be responsive to problems  

3. It allows objective measuring and monitoring of quality over time  

4. Insight is given to future improvement projects  

5. There is effective feedback which in turn is used to train staff  

6. Customers are given the best quality products