Codis

Across the pharmaceutical industry, the development of new drug products is increasingly challenged by the poor solubility of modern active pharmaceutical ingredients (APIs). Approximately 40% of the top 100 oral dosage drugs are poorly soluble in water, and up to 90% of new chemical entities face the same limitation. To ensure adequate bioavailability without increasing dosage- and therefore avoiding additional side‑effect . Spray drying is widely used to convert crystalline APIs into amorphous forms with improved solubility.

However, scaling spray‑drying processes from early clinical phases to late‑stage and commercial manufacturing is notoriously complex. Thermodynamics, atomisation behaviour, evaporation rates, and heat‑transfer dynamics all influence whether a new API can be successfully spray dried at scale. During clinical development, API availability is often extremely limited, making traditional design‑of‑experiments (DoE) approaches costly, time‑consuming, and environmentally burdensome. Each experimental batch consumes valuable API, large quantities of organic solvents, and significant energy.

Recognising both the scientific challenge and the sustainability opportunity, the Data Science team at Codis Haverhill developed a suite of advanced thermodynamic and physical‑property mathematical models to transform the way spray‑drying processes are designed and scaled. This initiative was conceived, developed, and implemented within a year, driven entirely by the team’s combined expertise in organic chemistry, chemical engineering, and computer science.

Innovation and Technical Achievement:

The team created sophisticated predictive models capable of simulating how a new API will behave under a wide range of spray‑drying conditions. These models incorporate:

1. Thermodynamic behaviour of solvent/API systems

2. Atomisation and droplet‑drying dynamics

3. Heat and mass‑transfer modelling

4. Predictive quality‑attribute outcomes, including amorphous content and particle morphology.

By integrating historical manufacturing data with deep technical knowledge, the team built a tool that allows process scientists to identify optimal operating parameters before entering the plant- eliminating the need for multiple experimental batches.

The models were not purchased or adapted from existing commercial tools; they were designed, coded, and validated internally. The team pushed beyond the capabilities of existing modelling frameworks, developing new algorithms and computational approaches tailored specifically to Codis’ spray‑drying platforms.

Environmental and Sustainability Impact-The sustainability benefits of this initiative are both significant and measurable.

1. Reduction in solvent usage

Most APIs are spray dried from organic solvents. Traditional DoE‑based scale‑up requires multiple experimental batches, each consuming large solvent volumes. By enabling accurate simulation of process conditions, the models have reduced the number of experimental batches by up to 100% for several customer projects. This directly translates into:

-Lower solvent consumption

-Reduced hazardous‑waste generation

-Decreased environmental emissions associated with solvent handling and disposal

2. Reduction in energy consumption

Spray drying is an energy‑intensive process. Every avoided experimental batch reduces:

-Dryer run‑time

-Heating and evaporation energy demand

-Ancillary energy use (HVAC, utilities, plant support systems)

This aligns strongly with Codis’ sustainability goals to reduce energy usage across manufacturing operations.

3. Conservation of scarce active pharmaceutical ingredients

During clinical development, API availability is often extremely limited. The modelling tools allow process scientists to optimise conditions without consuming any API at all, preserving material for clinical studies and reducing the environmental footprint associated with API synthesis.

4. Reduction in development timelines

Fewer experimental batches mean faster scale‑up, reducing the overall resource burden of bringing a new drug product to market. This accelerates patient access while lowering the cumulative environmental impact of development activities.

Tangible Results:

Since implementation, the Data Science team has successfully modelled multiple new products, consistently enabling customers to eliminate the need for traditional DoE‑based experimental campaigns.

In every case:

The predicted operating parameters matched or exceeded customer expectations.

The number of experimental batches was reduced by 100%, delivering immediate environmental and economic benefits.

Solvent usage and energy consumption were significantly reduced compared with historical scale‑up approaches.

These achievements have exceeded all initial targets set for the modelling initiative.

Cultural and Organisational Impact

Beyond the technical innovation, the team has demonstrated exceptional autonomy, creativity, and cross‑disciplinary collaboration. They independently researched existing modelling approaches, identified their limitations, and developed enhanced computational tools tailored to Codis’ needs. Their work has been shared across the wider Codis organisation, enabling other teams to adopt similar modelling‑led approaches to process development.

The initiative has also strengthened Codis’ reputation with customers, who now benefit from faster, more sustainable, and more reliable scale‑up of spray‑dried APIs.

To conclude:

The Codis Haverhill Data Science team has delivered a transformative sustainability initiative that directly reduces solvent usage, energy consumption, and API waste while improving process efficiency and product quality. Their internally developed mathematical modelling tools represent a step‑change in how spray‑drying processes are designed and scaled, with clear, measurable environmental benefits.

This nomination recognises not only their technical excellence but also their commitment to embedding sustainability at the heart of pharmaceutical manufacturing.