As demand for broader shade ranges and more inclusive complexion products continues across the United States beauty market, manufacturers and suppliers are reassessing traditional formulation workflows.

Dassault Systèmes, through its BIOVIA portfolio, is working with beauty brands to apply scientific modeling, virtual twin technology and AI-driven simulation to complexion product development. In this Q&A, Nick Reynolds, Industry Process Consultant Director, BIOVIA, Dassault Systèmes, discusses how predictive modeling is being integrated into R&D, its impact on shade inclusivity and what it may mean for manufacturers over the next five years.

CDU: From a formulation and R&D standpoint, what specific challenges in developing inclusive complexion products can specific modeling and simulation address more efficiently than traditional bench work?

NR: Two barriers many brands face when looking to develop inclusive complexion products are high costs for extensive testing and the technical challenges involved with formulating for diverse skin needs. Advanced simulation and modeling techniques solve both of these problems.

For example, cosmetic chemists can utilize data-rich virtual twin models, which are scientifically accurate digital replicas of real-life counterparts, to model all skin types based on real-world data. These virtual models reduce the need for repetitive physical testing, speed up ingredient screening, and enable faster decision-making, especially when creating expansive product shade ranges and analyzing complex skin interactions.

Models can integrate physics-based simulations to predict properties like solubility, while using formulation models that leverage existing data to predict the performance of new, untested formulations.

CDU: How does Dassault Systèmes’ simulation technology account for real-world variables such as undertone diversity, skin texture, sebum levels and environmental conditions when predicting how foundation will look, feel, and wear across different consumers?

NR: Dassault Systèmes’ simulation software is used to create a unified, collaborative environment for scientific and data-driven organizations, particularly in the life sciences, materials science, and chemicals spaces. Brands can pair this simulation technology with real-world data and scientific formulations to build digital skin models tailored to individual skin profiles, including different types of melanated skin.

They can also assess how various formulations perform on these different models taking into account shade and undertone diversity, aging properties, and sebum levels and continuously adjust their formulas. Through Dassault Systèmes’ cloud-based 3DEXPERIENCE collaboration platform, brands can virtually screen thousands to millions of potential formulations virtually and optimize them to tailored criteria.

From there, a select few formulations are chosen and can be tested in a laboratory. Brands can then re-input this physical feedback into the software platform to further augment current and future skin models.

CDU: What types of data inputs are required for accurate virtual testing, and how are beauty brands integrating these datasets into their existing product development workflows?

NR: Accurate virtual testing requires comprehensive datasets like ingredient properties, environmental parameters, skin type profiles, and lab results. Virtual testing does not replace the need for physical testing but drastically decreases it while being able to understand the chemistry happening within these trials around permeation, solubility, and personalization at a molecular level.

Brands should feed physical results from past and current experiments to help build robust virtual models. As a first step, we recommend beauty brands standardize functions like materials management and R&D on a unified cloud platform that key stakeholders can continuously collaborate, build, and contribute data into.

Creating this digital infrastructure ensures simulations are informed by real-world data while enabling key stakeholder visibility, so insights can be swiftly integrated into product development cycles.

This also creates a basis of legacy knowledge to be easily stored and accessed. It’s not uncommon for brands to need to redo experiments they’ve done previously because they’ve lost the initial data, so creating this repository of knowledge ensures all data is captured, accounted for, and leverageable.

CDU: For manufacturers and suppliers, where do you see the most immediate impact of simulation technologies: reducing the number of physical iterations, improving raw material selection, accelerating go-to market timelines, or something else entirely?

NR: Simulations reduce the number of physical iterations which leads to accelerated product launch timelines. Ahead of lab-scale production, brands can virtually screen ingredient interactions to predict formula performance, potential toxicity, and shelf life, decreasing potential safety hazards, costly errors tied to physical missteps, or long-term stock issues.

Reducing physical asset disposal can also help brands be more sustainable in their practices. Collectively, these efficiencies lower R&D costs and support more agile and responsive product development. Simulations are also valuable in root cause analysis when manufacturing issues arise, providing a fundamental understanding of material properties.

CDU: How might this approach influence claims substantiation and regulatory documentation, particularly as brands rely more heavily on predictive modeling during formulation?

NR: Funneling virtual models through one unified platform connected to the cloud ensures regulatory checking is completed early and on an ongoing basis, primarily in the design phase. Dassault Systèmes’ 3DEXPERIENCE platform is equipped with global regulatory and compliance information for brands to easily reference during product design.

These platforms can also leverage AI to automatically translate relevant data into regulatory documentation while ensuring full source traceability.

As consumers increasingly look for more ethical products, modeling helps substantiate these claims. This approach decreases the amount of physical testing needed and provides an accessible alternative for brands to move away from animal cosmetics testing for a cruelty-free product.

Source traceability gives brands the ability to ensure only ethical and sustainable materials are used in their products while meeting regulatory compliance. Simulation of properties such as toxicological endpoints is a valuable way to screen ingredients, while eliminating animal testing.

CDU: Looking ahead, how do you envision simulation and AI shaping cross-functional decisions – from ingredient innovation and shade range development to scaling manufacturing across the next five years of beauty product development?

NR: Everything we’ve discussed in our conversation will even further advance in the next five years to set new industry standards. Traditionally, discovering a new active ingredient took years of “wet lab” testing.

In the next five years, molecular simulation will allow chemists to test thousands of compounds virtually before a single beaker is touched. Advanced AI models, or AI advisors, will predict toxicology and allergenicity with such high accuracy that the beauty industry will collectively move past animal testing, as mentioned above.

Scaling a 1-liter lab sample to a 1,000-liter production batch is notoriously difficult in cosmetics because “shear” and “heat transfer” change at scale. Virtual twins of physical factories will solve this. Simulation software will model the fluid dynamics of a new cream within a specific industrial mixer.

This prevents broken emulsions and saves millions in wasted batches. AI will also build more resilient supply chains by monitoring global raw material fluctuations and suggesting new formulas in real time to maintain consistency without halting production.

AI will further augment shade range development with hyper-inclusive spectral accuracy, making inclusivity not just a buzzword, but a technical reality. Instead of physical prototypes, AI will accurately simulate how pigments reflect light on different skin textures and undertones, enabling product and marketing teams to align on a launch range that truly leaves no consumer behind.

We will see “mini-factories” at retail counters where AI scans a customer’s skin and triggers a simulation to mix a bespoke formulation on the spot, bridging the gap between a digital scan and a physical bottle.