improve customer outcomes in spec chem

How to Ensure Process Consistency for the Best Customer Outcomes

July 17, 2018

The recently published white paper “Manufacturing Excellence in Specialty Chemicals: Six Essential Levers” speaks to six elements manufacturers should consider in their quest to deliver bottom-line results from a manufacturing excellence program. These are:


  • Improve customer outcomes

  • Increase customer responsiveness

  • Improve asset effectiveness

  • Increase production throughput

  • Reduce operating costs

  • Achieve sustainable compliance


Let’s focus on the first lever — improve customer outcomes. All manufacturers want to produce right-first-time products for their customers with maximum efficiency and at minimum cost. An important first step towards doing this consistently is to ensure there are processes in place that enable the manufacturer to:


  • Know as soon as possible when there is a problem emerging

  • Assess whether the problem is more likely due to “not knowing what do,” versus “knowing what to do, but not doing it.”


Around for a long time as a concept, the establishment of daily management systems is foundational to the achievement of both early warning and root cause analysis of process problems. Daily management systems define standard operating procedures, critical process metrics and their targets, control and monitoring processes, and continuous improvement strategies for manufacturing unit operations.  Without this in place, it is difficult to execute with excellence, and to know when perfect execution didn’t lead to perfect results.


Enablers of effective daily management systems include procedural and recipe control solutions, such as Aspen Production Execution Manager™ (APEM). Particularly with manual or semi-manual process workflows, workflow management solutions can ensure that the steps that need to be followed per established SOPs are indeed executed per those requirements. This reduces the probability of poor quality, throughput or yield, and if any of these problems do surface, removes the question of whether or not the cause was failure to follow procedures per specifications.


Additional enablers are advanced process controls and real-time quality and process monitoring using univariate and multivariate statistical process controls (SPC/MSPC) and/or golden batch profiling. Whether your process is continuous or batch, controlling and/or monitoring critical process variables, as well as final product quality metrics, is vital to ensuring customer outcomes are within expected ranges of variability around prescribed targets, while maintaining internal process efficiencies and optimal first time performance.



A Solution That’s Right for You


Where the conversation becomes interesting is around what to choose. When is univariate SPC the solution? When is multivariate SPC the solution?  When do I use advanced controls?  This selection process will be the topic of a future blog, but for now, suffice it to say that AspenTech supports virtually every option within the aspenONE® Process Explorer™, aspenONE Process Explorer Analytics and Aspen ProMV™ software solutions. When the best approach is selected, there is a solution to meet the need.


In summary, if you know what variables matter and what processes must be followed, there are enabling solutions to help your operators stick to procedures and to know when you are off course.


But, if the problem is that you don’t know what is causing your quality, or throughput, or yield problems, to establish cause-and-effect relationships, one effective approach is design of experiments. But this can be time-consuming and expensive, so analysis of historical data is understandably and practically the approach that is taken most frequently.


Traditional regression analysis approaches, however, have a lot of pitfalls — particularly when it comes time to draw reliable cause and effect conclusions. An alternative approach is the use of multivariate statistical analysis techniques for offline process model building using historical data. Using multivariate statistical techniques, like those available in Aspen ProMV (offline), happenstance historical data can actually yield a predictive model that can be used to correct previously observed process problems and set optimal targets for the future. 


As noted in the white paper, specialty chemicals producers have achieved quality improvements of 10 to 20 percent by implementing the best practices described above, enabled by manufacturing execution, advanced process control and asset performance management solutions from AspenTech. The reduced variability resulting from these systems has directly contributed to improved outcomes for their customers.

There was a problem storing your subscription

Leave A Comment