Quality through quantification: how understanding helps generate profit

But how exactly do they go about developing this understanding?

Experimentation is all very well, and of course will remain a core activity for all companies operating in this sector. However, increasing the quantity of information from experimentation does not necessarily result in increasing knowledge or understanding of the system from which the measurements are taken, particularly where complex processes are involved. It is necessary to evolve a framework in which to capture and analyse this information in such a way that it can be put to work to improve design and enhance operations. Understanding how processes work is fundamental to quality and consistency.

Advanced Process Modelling

The key technology here is Advanced Process Modeling, or APM. Advanced process models, used for some time to accelerate process innovation in the chemical industry, capture detailed knowledge of physical and chemical phenomena in the form of mathematical relationships within a modelling software framework.

For example, chemical engineering relationships can be used to model the material and energy flows to and from a process to provide mass and heat balance information. So far, so good, and quite simple; this technology has been around for some time. But if one goes a step further, and adds detailed reaction information in the form of kinetic relationships, and the micro-scale multi-component diffusion that governs the rate of processes such as reaction and crystallisation, the picture alters. The more detail that is added in the form of fundamental, first-principles physical and chemical relationships, the closer the model can come to a true representation of what is happening inside the process at any time, and the greater the predictive accuracy.

However capturing such ‘engineering’ knowledge within model detail is only one half of the story. Even when working with fundamental relationships, there are always empirical values – in this case in the form of model parameters such as reaction kinetic constants and heat transfer coefficients – that need to be determined. This is the second, essential, ingredient of an advanced process model: accurate parameter information derived from real-world data, typically laboratory or pilot plant experiments. This data is used within mathematical optimization-based parameter estimation techniques to determine multiple model parameters to the highest degree of accuracy possible given the level of information contained in the experimental data.

It is the combination of these two components: rigorous first-principles models and accurate model parameters that provides such models with a highly-accurate predictive capability over a wide range of operating conditions. Once you have a fully-validated model, the sky is the limit. You can optimize the size and numerous other attributes of reactors; you can optimize recipes and operating policy to reduce batch times and maximize quality; you can scale-up crystallizers secure in the knowledge that they will actually produce crystals of the right size and shape; you can determine the cooling profiles that give the optimal crystal size distribution; you can use the model within the automation framework to monitor batches using inferential measurement, or to generate linearised models for batch model-predictive control; and many others.

Moreover, the very act of creating the model enhances understanding. Modelling forces the modeller to think about the fundamental process drivers; to examine the accuracy and appropriateness of the available data (and think about where effort should be expended in further experimentation); to consider the accuracy and completeness of the proposed reaction mechanisms given the fit with observed data; to question the validity of recipes that may have been devised without the benefit of accurate quantification; and to think of ways in which future designs and operations can be improved.

Example

A typical example of APM in action is in the manufacture of pharmaceutical-grade lactose, where the product must have a narrow crystal size distribution. PSE worked with the customer to customise library models to reflect the exact process configuration and include the key crystal growth phenomena. The models were then extensively validated against experimental data in order to ensure that the growth kinetics reflected the observed behaviour in all respects. The resulting models were then used to determine optimal processing conditions to ensure growth of uniform crystals of the right size and purity, as well as to optimise the batch recipe in order to minimise batch time and obtain greater throughput.

Modelling and experimentation

Modelling not only provides a ‘live’ tool to generate quantitative predictive information about processes, it is also a framework in which to capture information from the laboratory.

It is easy see how experimental data can be used to improve models, but less well known is that models can be used to improve experimentation with great effect. By building a relatively simple model of the experimental apparatus, complete with relationships describing the reactions you are studying, for example, you have a tool that allows you to do many things never previously possible.
Model-based data analysis will tell you immediately whether you need additional data in certain areas – for example, if the available data are not sufficient to determine the values for two parameters independently of each other – or if your data is sufficiently inaccurate to pose a significant design or operational risk. Using the same model in a model-based experiment design framework will enable you to design the optimal subsequent experiment – that is, the experiment that will yield the most information within the least time, or at the lowest ingredient cost.

This is the basis of a set of techniques called model-based innovation, which are helping to drive change through the chemical and other process sectors. It is also fundamental to another exciting new area, model-based risk management.

Model-based risk management

Model-based data analysis may tell you that a parameter value is likely to be, say, 2.5, but the confidence analysis of the data fit says that the real value could lie anywhere between 2.4 and 3.2. What if the true value is actually 2.4 or 3.2? Will your downstream equipment cope? Will you achieve required conversion and quality? If the answer is “no”, you know that you need to invest in refining the data. This may involve simply doing a few more experiments, or it may involve building an entire new R&D facility. The key here is that the model provides you with a quantification tool that allows you to support risk and investment decisions with hard numbers.

PSE’s gPROMS

Process Systems Enterprise is a pioneer and a leader in the application of this technology. The company’s gPROMS Advanced Process Modelling environment package has powerful modelling and solution capabilities that can handle highly detailed models covering multiple scales, from the micro-scale diffusion of molecules to the macro-scale circulation of cooling water in a jacket. Dynamic simulation and optimisation capabilities mean that gPROMS is equally suited to batch process optimisation – for example, determining minimum batch times for required product quality by direct calculation rather than repeated simulation – and decision support for continuous operation. State-of-the-art parameter estimation, model-based data analysis and experiment design capabilities mean that it can be applied across the model-based innovation cycle.

In addition, through many years of working collaborative working with customers, PSE has developed a set of industry-leading high-accuracy model libraries for many common reaction, crystallisation and separation processes. These minimise the overhead – and risk – in building models, and help ensure rapid payback on projects.

The company has also evolved well-tested methodologies for collaborative working with R&D and engineering personnel, including transfer of modelling know-how to in-house teams in order to build experience within customer organisations. Model-based innovation techniques, which formally combine modelling and experimentation to accelerate innovation, help to integrate R&D and engineering design functions in a way that has not been possible in the past using models as the medium for knowledge capture and transfer.

APM has been applied in recent years to great effect in the chemical industries. Pharmaceutical manufacture, with its equally complex reaction, crystallisation and biotechnology processes, stands to gain just as much, if not more – particularly when the time-to-market imperatives and the necessity to change rapidly are taken into account.

Mark Matzopoulos is a director of modelling technology company Process Systems Enterprise, suppliers of gPROMS modelling software and ModelCare services.