The pharma and biotech industries are driven by quality, and are highly regulated to assure that the required quality is delivered, consistently. A pharma company’s performance can be improved by systematically using data from operations to find more efficient and effective methods of accomplishing tasks, whilst also cutting down on the associated cost, rework, waste and time. This same data also allows for the demonstration of regulatory compliance and can point to remedial actions when violations do occur. JMP offers a comprehensive set of capabilities to assist you in managing your product quality and the underlying processes that deliver this.
JMP for SPC
SPC charts in JMP can flag out-of-control processes and provide ongoing monitoring of process stability. You can add phase and blocking variables and alerting rules to control charts. The control chart platforms can display continuous, discrete and rare event data. Control Chart Builder allows you to build charts via dragand-drop, so that you can easily experiment with different chart types and rational subgrouping to find the most effective control strategy for your process. Control Chart Builder also allows you to make charts that work when there is more than one source of random variation.
Analysis of measurement systems
You need to be able to trust the measurements you make. So, Measurement Systems Analysis (MSA) is fundamental to any data-driven approach. JMP provides you with everything you need to determine how much of the variation in your data is due to the way it is measured. In MSA, the precision, consistency, and bias of your measurement system is evaluated. Using EMP, a relatively new approach to MSA, it is possible to analyse the performance of your measurement equipment or an analytical method visually and quantitatively. EMP not only gives you a figure of merit, but it also gives a fair and effective assessment, suggesting how to improve your measurement system so you can spot real changes better.
Using the MSA platform in JMP, you can also carry out more typical GR&R studies (also accessible via the Variability/Attribute Gauge Chart platform). Using the AIAG criteria, this determines how much variation is accounted for by your measurement method in total. Your measurement systems can be analysed with confidence using JMP’s support for main effects, crossing and nested, as well as other types of models. The Variability/Attribute Gauge Chart can also quickly build charts that display the major sources of variation without requiring any formal study.
Process capability analysis
It’s important to see how the ‘voice of the process’ measures up to the ‘voice of the customer,’ and JMP allows you to carry out a variety of capability analyses. The capability platform lets assess many variables simultaneously in a single plot. The action needed to improve the performance of a variable is immediately apparent, making it easy to rapidly identify which variables need your attention through re-targetting, reducing variation, or both. Specification limits can be specified on one or both sides, and the dynamic linking in JMP provides the details you need on demand.
The process capability platform computes both ‘within’ and ‘overall’ estimates of variation to provide Cpk and Ppk as pacity indices. Non-normal and non-parametric distributions can be specified for each variable and the capability indices are calculated accordingly. If you like, you can use JMP to automatically select the best fitting distribution from your data. There are various possibilities for subgroup nesting too.
Process monitoring and stability at scale
You may quickly review hundreds of time-based variables using JMP’s process screening platform and identify the most critical that need attention to improve their stability and performance. You can easily sort the variables by key indicators like warning rates, stability ratios, mean shifts and drift indices to rank them as needed. You can then create the vital few control charts by selecting the variables of interest, giving capability analyses and a dashboard for further review.
Six sigma capabilities
◆ Understanding and controlling process variation visually
Six Sigma is usually seen as an overall approach to control variation in relation to requirements. JMP provides an extensive collection of graphical displays that can be used individually or in combination to interactively assess the structure of your data and literally see the dominant sources of variation. Working this way often allows you to identify and isolate the sources of variation that have a practical impact on your process, irrespective of technical statistical considerations. These linked dynamic representations allow you to go beyond what is possible with static graphs, and they become increasingly useful as your data becomes more complex through an increasing number of variables and rows in your data set.
◆ Determining sources of variation
With highly dimensional data, the effective application of dynamic visualisation, together with your understanding of the process, will frequently disclose a subset of the inputs, or X’s, that, either alone or in combination, seem to be responsible for the outcomes, or Y’s. In circumstances where this approach is not useful or workable, JMP provides sophisticated statistical tools that can effectively reduce dimensionality while maintaining information. When utilised in the spirit of finding relationships, JMP platforms such as partition, cluster and discriminant are typically highly successful in isolating the hot X’s that may then be used for more definite statistical modelling if this is appropriate.
Getting actionable results
Data collection and analysis always require time and involve costs that are explicit or hidden. Unless your insights are used to inform subsequent decisions and actions, your efforts to work with data will consume rather than provide value to your organisation. So, communication of your findings to a broader group of stakeholders is critical to success. Furthermore, because most real-world situations entail making trade-offs and compromises, you will almost always be required to generate consensus rather than simply communicate. JMP’s profilers and simulators allow teams to interactively review findings and explore what-if scenarios while drawing on contextual knowledge to create new process understanding – without becoming mired in the technical requirements of statistical modelling.
Resources
White Paper – Visual Six Sigma: Making Data Analysis Lean
https://www.jmp.com/en_us/whitepapers/jmp/visual-six-sigma.html
Webinar – Quality Engineering Tools in Bio/Pharma
In this on-demand webinar, learn how JMP is used in Statistical Process Control and Six Sigma workflows. Also learn how to import data, en-masse, into JMP, manage specification limits, create control charts and perform capability analysis in JMP.
Download your free trial of JMP®
Used by hundreds of thousands of data explorers worldwide, JMP data analysis software reveals insights that raw tables of numbers or static graphs tend to hide. Get more out of your data by downloading a free, fully functional 30-day trial now.
I am interested to author and/or co-author articles for publication such as manuscripts, book chapters, posters, abstracts, letter to editor and so on. Thanks.
Very Nice and Informative Blog you have shared with us.