Faced with high levels of demand and disruptive market conditions, pharmaceutical manufacturers, like others across the asset-intensive industries, are looking at technologies to drive agility and profitability, while at the same time helping to ensure security of supply
These challenges can effectively be addressed through digitalisation, reports Gerardo Munoz, Product Marketing Manager at AspenTech.
Pharmaceuticals companies can make more informed decisions by modelling outcomes against different scenarios that consider asset utilisation, changing supply and demand, product line pricing and other factors affecting quality, profitability and delivery.
By migrating to a hybrid modelling approach, they can harness a combination of first principles modelling and AI and machine learning techniques to bring new products to market faster and support decisions to implement efficiency initiatives.
Traditional first principles modelling draws on decades of experience from scientists, researchers and industry. It is an approach known and well-respected for its ability to accurately process key asset-intensive industry procedures, based on engineering fundamentals.
To extract even greater accuracy from the models generated, operational data are employed to calibrate them to observed conditions and performance, something which typically also requires considerable time, expertise and experience.
Hybrid modelling uses AI and machine learning, alongside first principles, to capture unknown or unmeasurable details of phenomena, while recalibrating models using this data to changing process conditions more easily.
Unlike a conventional modelling approach, in which expressions (reaction kinetics) need to be known or hypothesised before tuning their parameters, the machine learning algorithm learns how to predict the equipment performance from the available operating data.
This results in a more efficient workflow and enables easy recalibration to new operating conditions, in addition to a model that is typically more accurate over a wider range of conditions.
AI and machine learning allow manufacturers to build a model analysing a broader set of data while leveraging advanced data science techniques for model prediction. When combined with engineering principles and domain expertise, models can be built and maintained with less time and effort than traditional methods.
The other key benefit is that hybrid modelling provides a better representation of the plant, which keeps the model relevant for a longer period of time.
This reduces the barrier to entry for using modeling for asset optimisation by requiring less effort and expertise. With the models in place, the connected worker becomes free to perform higher value-added and strategic work.
Across the asset-intensive industries, there are a range of applications for this new approach: from plant digital twins right through to the use of process and laboratory data to help predict equipment breakdowns and optimise processes.
This hybrid modelling approach enables the creation of accurate, robust, and fast high-fidelity models that can be used to support real-time decisions in operations helping manufacturers, in turn, to optimally design, operate and maintain assets.
The new approach also helps balance profitability and sustainability. On the one hand, it can help operators react more quickly, be more agile and make informed, strategic decisions.
On the other, it can assist them in making choices that save energy, reduce cycle times and increase product quality. With hybrid models, the benefits of process simulation are expanded to assets and processes that traditionally don’t have a good simulation approach.
Hybrid modelling is rapidly emerging. Operators and manufacturers that don’t see themselves as early adopters might see the benefits of the approach but are waiting for it to become mature.
By so doing, they may miss out on the immediate benefits the methodology would bring them. For many organisations, the time is now for hybrid modelling. They can’t afford to let it pass them by.