Rheology, Texture Analysis and Tribology for Sensory Prediction and Sustainable Cosmetic Design

Abstract

The cosmetic industry is undergoing a deep transformation driven by rapid innovation, evolving consumer expectations, and increasing demands for sustainability. Formulators are required to design products that combine functional efficacy, stability, and appealing sensory properties while adopting environmentally responsible strategies. Traditional empirical and sensory-based approaches, though valuable, are often limited by high costs, time, subjectivity and lack of reproducibility. In this context, instrumental techniques provide an objective and predictive means to optimize product performance. Rheology, texture analysis, and tribology offer complementary insights into the structure, mechanical behavior, and interfacial phenomena of cosmetic formulations, all of which are closely linked to application behavior and sensory perception. Their integration enables a quantitative correlation between formulation composition, process conditions, and tactile performance. This review critically examines recent advances in the integrated use of rheology, texture analysis and tribology in cosmetic science, highlighting their role in sensory prediction, stability assessment, scale-up and eco-design. Together, these instrumental approaches support a more data-driven and innovation-oriented formulation paradigm, enabling database development and predictive modeling. Future research should prioritize database expansion, in vivo validation and machine learning integration to further improve sensory prediction and accelerate the design of advanced cosmetic formulations.

1. Introduction

Innovation represents a key driving force in the modern cosmetic industry, shaped by the need for accelerated development, shifting consumer demands, and growing regulatory constraints [1]. Cosmetic companies are continuously required to introduce new formulations with distinctive textures, responding to evolving market trends while ensuring product safety, stability, and compliance. At the same time, the sector faces the challenge of reformulating existing products to replace critical raw materials with the aim of reducing environmental impact and embracing eco-design principles, all without compromising performance or sensory appeal [2,3].

From a consumer perspective, sensory attributes, such as spreadability, thickness, stickiness, and after-feel, are among the first determinants of product acceptance and appreciation, playing a decisive role in purchasing behavior and fostering long-term loyalty and repeat purchases [4,5]. Consequently, manufacturers devote significant effort to fine-tuning the texture and application behavior of their formulations, which depends largely on ingredient selection, formulation architecture, and microstructural organization [6].

Traditionally, product development has relied on trial-and-error approaches, requiring numerous experimental iterations to achieve the desired texture, sensory feel, and stability. However, this empirical strategy is no longer compatible with the current demands of innovation and sustainability goals. Modern formulation development must be faster, data-driven, and cost-effective, integrating predictive analytical tools capable of correlating composition, structure, and sensory performance. While sensory panel testing remains indispensable, it suffers from inherent limitations related to subjectivity, high cost, and limited reproducibility [7,8].

Within this framework, the adoption of instrumental analytical techniques has become essential to support formulation design and optimization. Rheology has long been recognized as a fundamental tool for assessing the structure, flow, and deformation behavior of cosmetic products. By quantifying viscoelastic properties under controlled conditions, rheological tests provide insight into the internal microstructure, the spreadability, and the physical stability of soft matter systems [9,10,11]. Complementarily, texture analysis, originally applied in the food industry, measures application-related parameters such as firmness, adhesiveness, and cohesiveness, which are strongly associated with sensory perception [12,13]. Rheology and texture analysis, when used together, offer a comprehensive picture of a product’s mechanical and application performance, enabling more rational design of textures that meet consumer expectations [14,15].

Although less widespread, tribology, i.e., the study of friction and lubrication phenomena at skin–formulation interfaces, has emerged as a supportive technique, useful for interpreting attributes like slipperiness and after-feel [16,17]. When integrated with rheological and textural data, tribological analysis contributes to a deeper understanding of the mechanisms that govern tactile perception during product application [18]. The adoption of predictive rheological and mechanical models accelerates product development while minimizing empirical trial-and-error, resource consumption, and experimental waste [19]. Moreover, such models can inform the scale-up process, allowing early detection of stability issues and deviations between laboratory and production batches. These instrumental methodologies are thus required not only in R&D but also in manufacturing and quality control, where predictive formulation tools are essential for operational success [20,21].

Despite extensive experimental research, the cosmetic literature still lacks a unified and predictive framework linking physico-mechanical and sensory data. Existing studies and recent reviews [19,22] focus on individual techniques, with limited cross-comparison, insufficient attention to interfacial phenomena and reduced applicability to formulation design. This review integrates rheology, texture analysis and tribology within a single analytical framework, emphasizing friction and lubrication mechanisms in tactile perception and adopting an application-oriented perspective. The combined approach is discussed as a key enabler of predictive, data-driven and sustainable formulation strategies supporting eco-design, raw-material substitution and scale-up.

2. Rheology: Structure and Flow Behavior

Rheology, broadly defined as the science of deformation and flow of matter, investigates how materials respond to applied stresses or strains over time [9]. Cosmetic products are typically multicomponent colloidal systems exhibiting non-Newtonian, time-dependent behavior due to the presence of structured networks of polymers, surfactants, and dispersed phases. Their rheological fingerprints reflect not only the internal organization (e.g., droplet packing, polymer entanglement, liquid crystalline phases) but also determine key technological and sensory properties such as stability, spreadability, pick-up, and skin feel. By quantifying parameters such as viscosity, elastic storage (G′) and viscous loss (G″) moduli, yield stress, and thixotropy, rheological analysis allows formulators to predict a product’s response during both processing and use [23,24].

2.1. Continuous Shear Rheology

In steady-shear tests, the shear stress (τ) is measured as a function of the imposed shear rate ($\stackrel{·}{\gamma }$), and the flow curve obtained describes viscosity evolution and provides direct insight into material flow behavior and mechanical integrity. Most cosmetic formulations exhibit shear-thinning behavior, where viscosity decreases with increasing shear rate due to progressive alignment and disruption of internal structures during deformation [25]. From a quantitative standpoint, the steady-state response can often be modeled by constitutive equations such as the Power Law:

$$\tau =\mathrm{K}·{\stackrel{·}{\gamma }}^{ n}$$

(1)

where K is the consistency index and n (<1 for shear-thinning fluids) is the flow behavior index. The presence of a minimum stress threshold before flow occurs, the yield point (τ₀), is characteristic of gels, pastes, and suspensions, and it strongly correlates with their spreadability [26,27]. From a formulation standpoint, the flow curve provides critical information for processing and end-use performance. High zero-shear viscosity (η₀), i.e., the viscosity at rest, and yield stress ensure good physical stability during storage, whereas pronounced shear thinning facilitates ease of spreading and pleasant sensory perception [28,29]. Brummer and Godersky [30] demonstrated that the tactile sensations experienced when emulsions are spread onto the skin, the so-called primary and secondary skin feeling, can be directly correlated with distinct regions and parameters of the flow curve (Figure 1). The primary skin feeling, describing the initial impression upon contact, was found to correlate with the shear stress at the onset of flow (η₀). Products exhibiting moderate values of η₀ were perceived as smooth and rich, without excessive drag. Conversely, the secondary skin feeling, corresponding to the final stage of rubbing when the film becomes thin, was related to the stationary viscosity measured at the high shear rates (η_∞) typical of skin application (≈500 s⁻¹ for creams and ≈5000 s⁻¹ for lotions).

While steady-shear tests provide valuable information on flow behavior and yield phenomena, they are intrinsically destructive methods. Continuous rotation imposes large and cumulative deformations that disrupt the internal microstructure of the formulation, often leading to irreversible breakdown of polymeric networks or droplet associations. Moreover, the quantitative interpretation of the rheological parameters requires the adoption of empirical or semi-empirical constitutive models (e.g., Herschel–Bulkley, Carreau–Yasuda, Casson), which introduce uncertainties and fitting dependencies [31]. In addition, reproducibility is further limited at high shear rates, where instrument inertia, sample slippage, and thermal effects become significant, particularly in complex cosmetic matrices.

2.2. Oscillatory Rheology

Oscillatory rheology offers a non-destructive and highly sensitive approach for probing the complex internal architecture and the viscoelastic properties of cosmetic systems. In these tests, the sample is subjected to a sinusoidal strain (γ) and responds with a stress signal (τ) that, in general, is also sinusoidal but phase-shifted by an angle δ due to the coexistence of viscous and elastic components. From this relationship, two independent quantities are derived: the storage modulus (G′) and the loss modulus (G″). The storage modulus represents elastic energy storage; the loss modulus is associated with viscous dissipation. The relative magnitudes of G′ and G″ provide a quantitative fingerprint of internal coherence and elasticity, parameters directly related to product stability, texture, and sensory perception [32]. Their ratio, the so-called loss factor describing the dominant viscoelastic nature, constitutes the fundamental descriptors of a material’s mechanical identity:

tan δ = G″/G′

(2)

If tan δ < 1, the system has a solid-like or gel behavior; if tan δ > 1, the system has a liquid-like behavior.

In an amplitude sweep, G′ and G″ are recorded as strain increases at constant frequency. The plot reveals the extent of the Linear Visco Elastic Region (LVER), whose width denotes the material’s structural robustness and strongly depends on polymer type and microstructure [33]. The critical strain (γ_c) or flow point indicates the onset of irreversible structural breakdown (G′ ≈ G″) followed by the transition to viscous dominance (G″ > G′). The amplitude sweep thus provides a quantitative measure of microstructural integrity, identifying formulation resistance to deformation and predicting structural breakdown under application-like stresses [34].

In frequency sweep experiments, G′ and G″ are measured across several decades of frequency at fixed strain (within LVER), generating the mechanical spectrum. This spectrum reveals the relaxation behavior of the material (Figure 2). For strong gels such as carbomer dispersions or sodium polyacrylate cross-linked networks, G′ ≫ G″, and both are nearly frequency-independent, indicating a percolated and highly elastic structure with long relaxation times [35]. Weak gels (e.g., xanthan or scleroglucan dispersions) show G′ > G″, but both moduli depend on frequency, consistent with physical flexible, reversible interactions [36,37]. Viscoelastic fluids, such as hyaluronate or galactomannans water dispersions, display G″ > G′ at low frequencies, indicating flow-dominated behavior but partial elasticity at high frequencies [38,39].

At the low frequencies of the mechanical spectra, the slope of G′ and the relative distance between G′ and G″ provide valuable information on the long-term physical stability of an emulsified system [40,41]. Oscillatory parameters describing a product’s microstructure are also linked to application properties and sensory perception. Higher G′ and lower tan δ associated with a predominant elastic structure correlate with perceived firmness, body, and richness. Increased G″ and tan δ values are associated with fluid systems with smooth spreadability and lighter after-feel [42,43].

Rheology also represents a powerful tool for evaluating the thermal sensitivity and physical stability of cosmetic systems. The temperature sweep test, conducted under oscillatory conditions at fixed frequency and strain (within the LVER), is useful to monitor the evolution of G′ and G″ moduli as the temperature gradually increases. Variations in the viscoelastic moduli reflect microstructural rearrangements and the weakening of intermolecular interactions, providing predictive insight into a formulation’s response to storage and thermal stress [32,44].

2.3. Evaluation of the Application Properties

The relationships between the product’s viscoelasticity and application properties extend also to functional performance and efficacy. In sunscreen formulations, Infante et al. demonstrated that optimized rheological profiles, corresponding to lower tan δ and medium-high G′, improved film-forming behavior and enhanced in vivo SPF without increasing UV filter content, due to more uniform surface coverage, greater film thickness on the skin, and better structural stability [10]. The ability of cosmetic formulations to recover their internal structure after deformation is crucial for both their application performance and film-forming homogeneity on the skin. Rheological protocols such as the three-interval thixotropy test (3ITT) and creep–recovery tests are widely employed to quantify this behavior under stress conditions that mimic extrusion and spreading.

In a 3ITT test, the sample is exposed to three sequential stages: (i) a low-shear oscillatory step within the LVER to characterize the initial microstructure; (ii) a high-shear rotational pulse that induces controlled structural breakdown; and (iii) a final low-shear oscillatory step that monitors the time-dependent recovery of viscoelastic moduli. The extent and rate of recovery, calculated from the G′ and G″ evolution, provide a direct measure of thixotropic rebuilding, giving information about the product’s permanence at the application site [45]. As reported by Yilmaz and Vatansever [46], systems with higher polymer concentrations or stronger associative cross-linked networks (e.g., glucomannan or carbomer) showed faster and more complete regeneration after stress, mirroring better shape retention and reduced phase separation during processing. 3ITT appeared to be an effective tool to simulate and analyze effects of real pumping, mixing and instant stirring processes during production and handling steps of polymer films [47].

The creep–recovery test provides complementary and highly informative data on elastic resilience and structural memory. In this experiment, a constant shear stress (τ₀) is applied on the material for a defined time interval (creep phase), and the resulting strain (γ) is recorded as a function of time. When the stress is subsequently removed (recovery phase), the decrease in strain reveals the material’s ability to restore its initial configuration. The response is expressed as the creep compliance:

J(t) = γ(t)/τ₀

(3)

The ratio between recovered and total deformation quantifies the percentage of structural recovery (R%). According to Jayabal et al. [48] and He et al. [49], in cosmetic formulations, high elastic recovery and low irreversible compliance correlate with improved film uniformity, spreadability, and sensory smoothness after application, confirming the test’s value for predicting application performance and structural durability. Overall, 3ITT and creep–recovery protocols provide complementary insight into the resilience, reversibility, and reorganization kinetics of cosmetic systems, bridging rheological data with in-use performance and perceived texture [50].

While rheological measurements provide essential information on bulk flow and viscoelastic behavior, they do not fully describe the mechanical response experienced during product handling and initial contact with the skin. This gap is addressed by texture analysis, which captures deformation phenomena under conditions closer to real consumer interactions.

3. Texture Analysis: Linking Mechanical Properties to Cosmetic Performance

Texture analysis represents a fundamental instrumental approach to characterize the mechanical and surface properties of cosmetic products. Originally developed within the food science field to quantify sensory attributes such as firmness, chewiness, or cohesiveness [51,52], texture analysis was progressively adapted to cosmetic formulations to simulate real-use conditions and to bridge instrumental measurements with sensory perception. This methodological transposition has proven particularly valuable since cosmetic application inherently involves surface deformation, film formation, and tactile feedback, all phenomena that cannot be fully captured by conventional rheological tests alone.

3.1. Texture Parameters

The immersion/de-immersion test, also known as back-extrusion, is one of the most versatile instrumental methods for the mechanical characterization of semi-solid and viscous cosmetic formulations [12]. The principal textural parameters are extracted from the resulting force–time or force–distance curve (Figure 3) [53,54]. Firmness is the maximum compression force, related to sample viscosity and structural rigidity. Consistency represents the total work of deformation calculated from the area under the positive curve. These two parameters can be related to the ease of spreading on the skin, as the higher the necessary force to reach a given deformation, the poorer the product spreadability [55]. Cohesiveness corresponds to the maximum negative force during withdrawal, representing internal bonding strength of the structure. Adhesiveness, or work of cohesion, expresses the total work required to overcome internal forces during detachment and corresponds to the area under the negative curve. Both parameters, obtained during the de-immersion phase, are closely related to the formulation’s retention or permanence at the site of application. When not properly balanced, they can markedly influence the perceived tackiness and stickiness of the product during and after spreading. Stringiness is defined as the distance at which the material detaches from the probe, derived from the final portion of the texture curve, where the characteristic tailing occurs during the de-immersion phase. From an applicative standpoint, this parameter is particularly informative of the pick-up behavior and of the perceived “stringy” character of the formulation, a feature that, when excessively pronounced, can negatively affect consumer acceptance and sensory perception [56,57].

3.2. Influence of Composition on Product Physico-Mechanical Properties

Texture parameters provide a quantitative framework linking formulation structure to consumer-perceived mechanical attributes. Early studies on semi-solids [58] demonstrated that firmness and adhesiveness reflect the strength of polymer–polymer and polymer–solvent interactions, a relationship later confirmed for cosmetic formulations. Tai et al. [12] systematically quantified mechanical parameters across a wide range of cosmetic raw materials and formulations. They found that cross-linked acrylate gels displayed higher firmness and lower stringiness than surfactant-based systems, highlighting the strong link between polymer network structure and surface mechanical response. Similarly, Hurler et al. [59] observed that increasing polymer concentration in carbomer, poloxamer, or chitosan hydrogels enhanced network strength, increasing firmness, cohesiveness, and adhesiveness, while humectants such as glycerol improved spreadability and reduced stickiness. Expanding this methodology, Tamburic et al. [60] compared semi-solid emulsions thickened with carbomer or xanthan gum and demonstrated that the inflection point in the compression curve corresponds to the yield stress measured by rheometry, confirming linear correlations between polymer concentration and flow onset.

The comparative evaluation of synthetic and natural rheological modifiers both in water dispersions [57,61] and in emulsions [62] clearly demonstrated that the microstructure and cross-linking degree of the polymer network govern both rheological and texture properties of cosmetic systems. Carbomer-based systems exhibit higher firmness, consistency, and a strong-gel rheological behavior (G′ ≫ G″) due to their covalently cross-linked microgel structure, while polymers like xanthan- or cellulose-based formulations display softer textures with moderate adhesiveness and stringiness, resulting from weaker hydrogen-bonded or entangled networks. Increasing polymer concentration systematically enhances firmness, cohesiveness, and adhesiveness, whereas stringiness decreases as the network becomes more compact and less deformable. In contrast, viscoelastic gels formed by galactomannans such as guar, carob, and tara gums show high stringiness and adhesiveness, which increase with polymer concentration depending on the degree of galactose substitution that prevents the formation of hydrogen bonds and inhibits the aggregation of the polysaccharidic main chains. The results established a clear correlation between the rheological parameters and textural responses, highlighting that these instrumental analyses are complementary tools for predicting application behavior. Firmness and consistency are correlated with G′ and viscosity, while adhesiveness and cohesiveness are linked to viscoelastic moduli. Stringiness, instead, depends mainly on the viscous component (tan δ), increasing in weakly structured, long-flow systems and decreasing in highly cross-linked or stiff matrices.

Further formulation studies have consolidated the relationship between composition, structure, and mechanical response in cosmetic systems [13,63]. Filipović et al. [64] analyzed creams stabilized with nonionic emulsifiers and showed that higher internal phase and thickener content enhanced inner packing, contributing to higher firmness and consistency, in line with increases in G′ and yield stress. A more recent study by Martins et al. [65] introduced innovative bacterial cellulose–carboxymethyl cellulose composites (BC:CMC) as natural stabilizers and texturizing agents in surfactant-free creams. The authors correlated the strong elastic (G′ > G″) and gel-like behavior of BC:CMC emulsions with enhanced firmness and reduced stickiness, attributing these properties to the formation of a 3D fibrillar network stabilizing the continuous phase. From a formulation efficacy standpoint, Shirata and Maia Campos [66] investigated how UV filters reduce samples’ consistency due to an increased oily phase, whereas the insertion of film-forming polysaccharides counterbalanced these effects by enhancing surface smoothness and elasticity.

Beyond polymeric thickeners, several formulation components can modulate the rheological and textural performance of cosmetic systems. Andonova et al. [67] demonstrated that the fatty acid composition of emollients determines mechanical strength: oils rich in α-linolenic and linoleic acids produce firmer, more cohesive creams, whereas saturated oils yield softer, more spreadable textures. The authors, through the correlation between the consistency index (K) from the Power law and firmness from the texture profile curve, highlighted the predictive value of texture analysis for rheological performance [68]. Bogdan et al. [69] found that surfactant blend composition in cleansers controls foam consistency and viscosity, with fatty-acid-based surfactants increasing structure and glycerol/betaine improving elasticity and spreadability. Among lipid structuring systems, Da Costa-Sanches et al. [70] formulated an organogel structured with 12-hydroxystearic acid and enriched with hyaluronic acid, showing high mechanical strength and a predominant elastic behavior (G′ > G″) typical of semi-solid elastic gels. The incorporation of hyaluronic acid increased the texture parameters of firmness and cohesiveness, evidencing molecular cross-linking within the lipid network.

Rheological and texture data show consistent correlations across all systems. This combined analytical approach allows a mechanistic understanding of how composition, ingredient chemistry and concentration, and microstructure govern both the functional stability and application performance of cosmetic products [71]. However, neither bulk rheology nor texture analysis alone can account for interfacial phenomena governing slip, drag and after-feel during application. These aspects are directly addressed by tribological measurements, which probe frictional behavior under skin-relevant contact and lubrication conditions.

4. Tribology for the Assessment of Spreading Behavior

Tribology is a scientific discipline that examines friction, lubrication, and wear between two surfaces in relative motion. It originated from classical mechanical engineering, especially for metal–lubricant interactions, and was later extended to soft materials and biological interfaces [72]. Its relevance to cosmetic science arises from the inherently interfacial nature of topical formulations: during application, products form thin lubricating films between the fingers and the skin, and the resistance to sliding modulates key sensory perceptions such as glide, smoothness, drag, richness, and the progressive transition toward after-feel. While rheology quantifies the bulk viscoelastic response of a formulation and texture analysis describes surface mechanical attributes, tribology captures the dynamic interfacial behavior emerging during spreading and massaging, conditions that more closely mimic real use [22].

At the core of tribological analysis lies the coefficient of friction (COF), defined as the ratio between the tangential frictional force (F) and the applied normal load (N) [73].

COF = F/N

(4)

In cosmetic films, the COF reflects the ease of sliding between model surfaces, often PDMS, Teflon or biological surrogates, and the formulation acting as a lubricant. High COF values are typically associated with sensory impressions of resistance and insufficient slip, whereas lower COF values correspond to a smoother, more lubricious application. A central interpretative tool in tribology is the Stribeck curve, which describes the evolution of the COF as a function of sliding speed (Figure 4). The curve reveals distinct lubrication regimes [74]. At low speeds, the lubricant film is thin, and surfaces remain in close contact, resulting in the boundary lubrication regime, characterized by high friction. As sliding speed increases, lubricant entrainment begins to separate the surfaces, reducing friction and leading to a mixed lubrication regime. At sufficiently high speeds, a continuous fluid film is formed, producing hydrodynamic lubrication, where the COF reaches a minimum before rising again due to the viscous resistance of the fluid.

Within cosmetic applications, these regimes correspond to the phases of product use [75,76,77]. The early boundary regime reflects the initial contact and “break-in” of the formulation, often associated with tack or initial drag. The mixed regime corresponds to spreading, when the film becomes more uniform and smoother. The hydrodynamic contribution emerges transiently during fast hand movements or when the product contains highly lubricious emollients, and its decay reflects the return to boundary-like behavior associated with dry-down and after-feel.

Integrating tribological data with rheology and texture analysis enables a more complete physico-chemical understanding of application behavior and enhances the capacity to rationally design formulations with targeted sensory profiles. Farias et al. [78], using bulk rheology and soft tribology on PDMS contacts, investigated model systems of increasing complexity: hydrophobically modified polymer gels, polymer–phospholipid mixtures, and O/W emulsions. Polymer concentration mostly controlled viscosity and gel strength, with higher G′ and yield stress, whereas the addition of phospholipids significantly lowered boundary friction due to adsorbed interfacial films of phospholipids and polymer chains. Confocal microscopy confirmed the formation of adsorbed layers on PDMS surfaces, explaining the enhanced slip at low entrainment speeds. Cyriac et al. [18] expanded this framework to a set of 56 commercial creams, lotions and gels, integrating instrumental texture analysis, friction measurements and both linear viscoelasticity and non-linear rheology (LAOS). Principal component analysis indicated that LAOS parameters and friction measurements cluster together with instrumental texture attributes such as firmness, adhesion, cohesiveness and spreadability, highlighting their shared relevance during high-strain application. Their findings support the conclusion that tribology and non-linear rheology describe the interfacial and structural changes occurring during rub-out and after-feel more accurately than steady or small-amplitude rheology alone.

The complementary nature of rheology, texture analysis and tribology naturally leads to integrated frameworks aimed at predicting sensory perception. By combining bulk, mechanical and interfacial descriptors, several studies have demonstrated improved correlations with in vivo sensory attributes.

5. Instrumental Prediction of Sensory Performance

Sensory characterization has traditionally relied on methodologies such as consumer tests, trained panel evaluations, and structured descriptive approaches including quantitative descriptive analysis (QDA) and spectrum descriptive analysis (SDA), which allow the systematic quantification of perceptual attributes of cosmetic products. Even if these methods remain the gold standard for assessing consumer-relevant sensory profiles, their inherent variability, dependence on panelist expertise, and operational costs have stimulated the adoption of instrumental tools capable of objectively probing the physico-mechanical behaviors underlying sensory perception [19]. The growing body of evidence indicates that specific rheological, texture and tribological parameters can serve as reliable predictors of sensory attributes, offering mechanistic insights into the phenomena governing product pick-up, skin application, film formation, and post-application tactile perception [22] (Table 1). Establishing such predictive relationships not only enhances formulation rationalization, overcoming the traditional trial-and-error approach, but also enables data-driven product development, contributing to greater efficiency, reproducibility, and innovation within the cosmetic industry.

Gilbert et al. [79], considering laboratory-prepared O/W emulsions and commercial products, demonstrated that sensory descriptors involving simple mechanical actions, i.e., penetration force, compression force, stringiness and spreading, were predicted with high accuracy using imitative texture tests, in which the instrumental protocol replicated the panel procedure. Penetration force and compression force were strongly correlated with the positive area of the force–time texture curves (Pearson’s correlation coefficient R_p > 0.98). Stringiness, defined as filament continuity during finger separation, significantly correlated with the breaking length of the filament in the stretching test (R_p > 0.97). Notably, more structurally complex sensory attributes, such as integrity of shape, which is the degree to which product holds the given shape, required a combination of rheological parameters to achieve high prediction accuracy: the best predictors were G′ in the LVER, the stress at the end of the LVER, and the work of extrusion (R_p > 0.97). Imbart et al. [80], analyzing a broader set of commercial emulsions, confirmed and extended these correlations using multivariate statistical principal component analysis (PCA). Their results showed that firmness and cohesiveness from texture extrusion tests aligned closely with sensory visual residue (immediate and after 1 min), the former positively and the latter negatively. On the other hand, yield stress and apparent viscosity at low shear were inversely associated with adherence and greasiness 1 min after application, indicating that perceived tack and residue derive from the combined effects of structural rigidity and restricted flow mobility. From a mechanistic perspective, higher network packing limits film leveling and relaxation at the skin surface, promoting localized accumulation and persistent residue. Importantly, the 60-s structural recovery percentage from 3ITT tests correlated with softness and visual residue, demonstrating that sensory after-feel depends on the kinetics of microstructural rebuilding following shear-induced breakdown during spreading.

In the work of Calixto et al. [81], conducted on O/W emulsions structured with waxes and polymers, apparent viscosity and texture parameters of consistency and cohesiveness showed significant associations with panel-assessed viscosity, consistency, and cohesiveness (R_p > 0.92). Their subsequent study [82] confirmed these relationships in a broader formulation matrix, highlighting that wax-rich emulsions displayed higher firmness and cohesiveness, both instrumentally and sensorially, due to the formation of rigid, percolated crystalline wax domains, while polymer-rich matrices reduced instrumental work of shear as a result of more deformable, shear-responsive polymeric structures, aligning with higher sensory spreadability scores. Estanqueiro et al. [55] examined a series of creams thickened with different hydrophilic polymers and showed that instrumental firmness was positively correlated with sensory firmness (R_p = 0.610) and negatively with spreadability (R_p = −0.843), confirming that higher penetration resistance translates into poorer spreading perception. Moreover, instrumental adhesiveness showed a weak negative correlation with sensory spreadability (R_p = −0.617). Kulawik-Pióro et al. [83] and Vergilio et al. [84] reported analogous findings, confirming that rheological resistance (yield stress and viscosity) and adhesive/cohesive balance strongly govern tactile perception during both spreading and after-feel phases. A similar trend was observed by Vieira et al. [85], who compared sensory and texture data of oils, starches, butters, and glycolic extracts. Their PCA further confirmed the clustering of high-consistency materials (e.g., shea butter, tapioca starch) with sensory attributes such as stickiness and reduced fluidity, reflecting the presence of structured or semi-solid networks that resist deformation and promote cohesive contact with the skin surface, whereas oils clustered with slipperiness, fluidity and oily residue, aligning well with their low mechanical resistance, rapid interfacial spreading and boundary lubrication during application.

Additional studies further reinforce that structural and flow-governing properties are the primary determinants of sensory performance in semi-solid and emulsion-based cosmetic systems, even when no formal statistical correlations are reported [86,87,88]. Among the most recent and application-relevant contributions, the work of Špaglová et al. [89] provides a particularly valuable model, as it investigates sunscreens and SPF boosters, one of the fastest-growing formulation categories in the cosmetic industry. The authors demonstrated that increases in viscosity, firmness and adhesiveness, driven by variations in emulsifier type, reduced droplet size and particle-based booster concentration, reduced spreadability, and enhanced tackiness and residual feel, with direct consequences for film uniformity and, ultimately, SPF-boosting efficiency. Their findings highlighted that improvements in photoprotective performance did not necessarily align with enhanced sensory attributes, reinforcing the need to balance sensory pleasantness and functional efficacy through careful ingredient selection, optimized concentrations, and objective characterization of application-related properties.

Across these studies and formulation types, a coherent pattern emerges: sensory spreadability, fluidity, and early tactile perception can be easily predicted by instrumental parameters related to a system’s structure. In contrast, stickiness, residue and after-feel show markedly higher variability, reflecting sensitivity to interfacial phenomena, film evolution and individual perception. Differences in experimental protocols, including shear history, application rate, contact geometry and substrate selection, further contribute to inconsistencies between reported correlations, limiting direct cross-study comparability. The integration of rheology, tribology and sensory evaluation has been explored to identify which physical parameters govern tactile perception across the different phases of product use. Savary et al. [90] examined five O/W emulsions formulated with different emollients, assessing spreading through sensory evaluation and an instrumental sliding-friction test using a texture analyzer. Despite similar rheological profiles, emulsions showed distinct frictional responses. A significant correlation emerged between sensory spreading and instrumental work of friction (R_p ≈ −0.78), demonstrating that friction-based metrics better captured perceived ease of spreading than viscosity alone. In subsequent work [91], the authors extended the approach to six commercial topical products (gels and emulsions), integrating sensory analysis, rheology, texture, and in vivo tribology via a frictiometer. Optimal spreadability and low perceived drag were associated with viscosities of approximately 0.1–0.3 Pa·s at application-relevant shear rates and yield stress values typically below 50–100 Pa. Texture analysis further supported this prediction, as instrumental work of spreading values below ~200 g·s and moderate firmness correlated with high in vivo spreadability and low stickiness. However, only tribology captured the time-dependent evolution of the residual film: emulsions maintained low friction due to the oil phase, while gels showed a sharp friction increase driven by water loss and film thinning. Crucially, the sensory “amount of residue” attribute correlated negatively with the area under the friction curve (R_p = −0.850), directly linking friction dynamics with after-feel. The study of Lee et al. [92] confirmed these results. They investigated 23 commercial facial moisturizers covering a wide range of textures and compositions, combining steady rheology, texture analysis, tribology on artificial skin, and a descriptive sensory panel covering 17 tactile attributes. Their results clearly showed that bulk rheology dominates early sensory stages such as appearance and pick-up. On the other hand, tribology becomes increasingly relevant as the film thins during rubbing, eventually becoming the primary predictor of residue and after-feel. Although the assessment of residue attributes was more challenging for expert panelists, as evidenced by the larger rating variations compared to those for other attributes, the authors found that the coefficient of friction at very low speed (0.16 mm/s) was negatively correlated with slipperiness and silicone-like feel, and positively correlated with thickness-related attributes such as perceived residue and greasiness. Formulations whose friction coefficient was below ~0.3 in the boundary lubrication regime were perceived as smoother and less prone to residue after application

Comparison across studies reveals several limitations that constrain the generalization and predictive power of instrumental–sensory correlations. Many studies rely on small formulation sets, or narrowly defined compositional variations, which limits the robustness and transferability of the reported relationships to broader formulation spaces. Correlations are often derived from simple linear models or qualitative multivariate associations, without systematic validation on independent datasets or across different product categories. Furthermore, the heterogeneity of sensory methodologies, ranging from trained panel descriptive analysis to semi-quantitative consumer evaluations, introduces variability that complicates cross-study comparison. Within this context, linear and multivariate regression frameworks, including multiple linear regression and partial least squares regression, have been widely employed to quantitatively predict sensory attributes such as spreadability, stickiness and residual feel from rheological parameters, texture descriptors and friction measurements, using in vivo sensory scores as dependent variables [43,79,91]. While fully mechanistic models capturing skin–product–perception coupling are not yet available, these regression-based mathematical frameworks provide the current state-of-the-art for predicting in vivo sensory performance. The quantitative parameter ranges and integrated rheology–texture–tribology signatures identified through these models define critical targets that must be preserved during sustainable reformulation and eco-design, providing a direct bridge to the application-oriented strategies discussed in the following section.

Table 1. Summary of statistical correlations between sensory attributes evaluated during different application phases and instrumental rheological, texture and tribological parameters: (+) stands for positive correlations, (−) for negative ones.

Sample Type	Application Phase	Sensory Attributes	Instrumental Parameters	Ref.
Lab-prepared and commercial O/W emulsions	Pick-up	Penetration force	Positive area under texture curve (+)	[79]
		Stringiness	Breaking length of the filament (+)
		Integrity of shape	G′ LVER, τ end LVER (+)
Commercial O/W emulsions	After-feel	Residue	Firmness (+), Cohesiveness (−)	[80]
		Softness	Cohesiveness (+), Firmness, G′ (−)
		Greasiness	tan δ, Cohesiveness (+), Viscosity (−)
O/W emulsions structured with waxes and polymer	Pick-up	Cohesiveness	Viscosity, Cohesiveness, Consistency (+)	[81,82]
	Rub-out	Consistency	Viscosity, Cohesiveness, Consistency (+)
		Spreadability	Viscosity, Firmness, Consistency (−)
Creams thickened with different hydrophilic polymers	Rub-out	Spreadability	Firmness, Adhesiveness, Viscosity (−)	[83]
Commercial sunscreens	Rub-out	Spreadability	Viscosity, Consistency, Cohesiveness (−)	[84]
	After-feel	Stickiness	Firmness, Consistency, Cohesiveness (+)
		Residue	Firmness, Cohesiveness (+)
Raw cosmetic ingredients	Rub-out	Fluidity	Consistency, Cohesiveness, Viscosity index (−)	[85]
O/W emulsions with different oily phase	Rub-out	Spreadability	Area under the friction curve (−)	[90]
Commercial topical products	Pick-up	Firmness	Positive area from the texture curve (+)	[91]
	After-feel	Stickiness	Negative area from the texture curve (+)
		Residue	Area under the friction curve (−)
Commercial face moisturizers	Pick-up	Thickness	Viscosity, Friction coefficient at low speed (+)	[92]
	Rub-out	Spreadability	Viscosity, Friction coefficient at low speed (−)
	After-feel	Stickiness	Negative area from the texture curve (+)
		Greasiness	Friction coefficient at low speed (+)
		Silicone-like feel	Friction coefficient at low speed (−)

Table 1. Summary of statistical correlations between sensory attributes evaluated during different application phases and instrumental rheological, texture and tribological parameters: (+) stands for positive correlations, (−) for negative ones.

Sample Type	Application Phase	Sensory Attributes	Instrumental Parameters	Ref.
Lab-prepared and commercial O/W emulsions	Pick-up	Penetration force	Positive area under texture curve (+)	[79]
		Stringiness	Breaking length of the filament (+)
		Integrity of shape	G′ LVER, τ end LVER (+)
Commercial O/W emulsions	After-feel	Residue	Firmness (+), Cohesiveness (−)	[80]
		Softness	Cohesiveness (+), Firmness, G′ (−)
		Greasiness	tan δ, Cohesiveness (+), Viscosity (−)
O/W emulsions structured with waxes and polymer	Pick-up	Cohesiveness	Viscosity, Cohesiveness, Consistency (+)	[81,82]
	Rub-out	Consistency	Viscosity, Cohesiveness, Consistency (+)
		Spreadability	Viscosity, Firmness, Consistency (−)
Creams thickened with different hydrophilic polymers	Rub-out	Spreadability	Firmness, Adhesiveness, Viscosity (−)	[83]
Commercial sunscreens	Rub-out	Spreadability	Viscosity, Consistency, Cohesiveness (−)	[84]
	After-feel	Stickiness	Firmness, Consistency, Cohesiveness (+)
		Residue	Firmness, Cohesiveness (+)
Raw cosmetic ingredients	Rub-out	Fluidity	Consistency, Cohesiveness, Viscosity index (−)	[85]
O/W emulsions with different oily phase	Rub-out	Spreadability	Area under the friction curve (−)	[90]
Commercial topical products	Pick-up	Firmness	Positive area from the texture curve (+)	[91]
	After-feel	Stickiness	Negative area from the texture curve (+)
		Residue	Area under the friction curve (−)
Commercial face moisturizers	Pick-up	Thickness	Viscosity, Friction coefficient at low speed (+)	[92]
	Rub-out	Spreadability	Viscosity, Friction coefficient at low speed (−)
	After-feel	Stickiness	Negative area from the texture curve (+)
		Greasiness	Friction coefficient at low speed (+)
		Silicone-like feel	Friction coefficient at low speed (−)

6. Instrumental Approaches for Eco-Design Formulation

The current transition toward green, sustainable and biodegradable cosmetic systems, driven by growing environmental concerns and regulations, is forcing formulators to rethink the design logic traditionally applied to product formulation [93,94]. Substituting petrochemical-derived polymers, synthetic thickeners, microplastics or high-impact emulsifiers with bio-based, renewable and eco-friendly raw materials can deeply affect the architecture, flow behavior, and sensory performance of formulations [95,96]. Here, instrumental physical characterization, such as rheology, texture analysis and tribology, becomes essential for accelerating development cycles and reducing reliance on empirical, time- and resource-intensive approaches.

Recent advances highlighted the central role of polysaccharides as renewable rheology modifiers for eco-designed formulations, offering viable alternatives to synthetic acrylate-based polymers [97,98]. However, while synthetic rheology modifiers have been engineered to deliver highly specific texture signatures alongside robust stability, natural polymers often display lower structuring efficiency and elasticity, weaker network connectivity, and inconsistent sensory profiles, as shown in comparative rheological-textural mapping of natural versus synthetic rheology modifiers [57] and in the systematic correlations between viscoelastic fingerprints and sensory attributes of polymer-thickened emulsions [62]. Variations in polymer chemistry, concentration and environmental conditions have an impact on the rheology, texture, and tribology signatures of cosmetics, and cross-linked synthetic polymers typically induce a strong increase in elastic modulus and yield stress with concentration, leading to reduced spreadability, stringiness and limited pick-up, and increased friction. In contrast, polysaccharides exhibit a less linear response: at higher concentrations, they may promote network heterogeneity and syneresis, accompanied by pronounced stringiness and less homogeneous tactile perception [57]. These systems are also significantly more sensitive to environmental conditions; polysaccharide-based gels display strong dependence on shear and thermal history, as weak, reversible physical junctions can be irreversibly disrupted under high shear or temperature fluctuations, resulting in loss of structure, altered recovery behavior [36] and changes in lubrication regime. This intrinsic performance gap represents one of the major barriers to eco-design: consumers expect the same pick-up, spreadability, and after-feel, regardless of whether the formulation is synthetic or natural. A viable strategy to narrow this performance gap is the combination of polysaccharides with complementary rheological behavior, pairing highly structuring, rigid-coil hydrocolloids (e.g., xanthan gum or gellan gum) with more flexible, adhesive or viscoelastic modifiers (hyaluronic acid, pectins, or galactomannans) or protein-based fibers to build networks that more closely reproduce the elasticity, yield stress and network recovery typical of synthetic polymers [99,100,101,102]. Rheology and texture analysis are essential to quantify these synergistic effects, identify optimal ratios, and verify whether the resulting systems approach the target viscoelastic and sensory profile required for eco-designed formulations. Even with a growing body of evidence supporting polysaccharide-based and hybrid biopolymer systems, these studies remain limited to a restricted concentration range and to simplified model formulations, making it difficult to fully extrapolate the reported rheological and textural synergies to complex, multicomponent cosmetic products.

Green reformulation is rarely a one-to-one substitution process [103]. Eco-design must also include the substitution of the lipid phase, a key driver of spreadability, lubrication and sensory identity, with bio-derived, low-impact alternatives [79,104], considering that rheological and textural responses can vary markedly depending on the polarity and fatty-acid composition of the added oil [67]. De Castro et al. [105] investigated a set of six biosynthesized branched esters as sustainable substitutes for cyclopentasiloxane (D5) and low-viscosity silicone oils, an especially relevant target given the high environmental impact and regulatory pressure associated with volatile silicones. When incorporated into O/W gel-cream, the esters produced rheological and texture profiles comparable to those of the silicone-containing control, indicating preserved structural integrity, making them promising low-impact alternative emollients. However, these investigations assess samples’ performance under laboratory conditions or under limited shear histories, without systematically addressing the impact of processing, filling or repeated application on lubrication regimes and film formation.

Instrumental characterization also offers a powerful framework for evaluating novel classes of sustainable functional ingredients, whose introduction often perturbs the rheological and sensory equilibrium of a formulation [106]. Recent studies demonstrated how natural deep eutectic solvents (NaDES), although attractive as green solubilizers and humectants, require careful structural tuning to maintain desirable application properties [107,108]. The incorporation of NaDES-based botanical extracts into hydrogels, emulsions and emulgels not only enhanced SPF and antioxidant activity but also altered network strength, elasticity and cohesiveness, as highlighted by texture and flow measurements. Lukić et al. [109] specifically addressed the use of upcycled plant-derived actives obtained from agro-industrial waste streams, demonstrating that the extraction process can induce non-trivial modifications in emulsion microstructure and application properties. Lipid extracts modified droplet size and reduced shear stress, affecting spreadability and occlusivity, reflecting their direct integration into the dispersed phase and the consequent modification of interfacial packing and internal structuring. Ethanolic extracts produced the opposite rheological trend and affected stickiness. A key limitation lies in the strong formulation and solvent dependence of the observed effects, which reduces transferability between systems and highlights the need for standardized characterization protocols to ensure reproducibility and scalability.

Tribology has emerged as a crucial tool for eco-design, in combination with rheology and texture analysis, enabling formulators to quantify lubrication regimes and frictional behavior of new, sustainable formulations that directly underpin sensory perception and application performance. Ajayi et al. [76] showed that a novel amino-lipid surfactant, proposed as a biodegradable alternative to conventional poliquaternium, exhibited tribological and wet-lubrication performances comparable to standard conditioning agents, while offering improved environmental compatibility. The integration of combing tests, yield stress measurements and friction curves clearly illustrated that tribology could discriminate subtle effects such as film formation, sliding behavior and shear resistance in ways not captured by flow curves alone [110]. More recently, Rischard et al. [111] demonstrated that the integration of viscosity, texture and friction measurements can generate robust predictive models of in vivo tactile properties, enabling the screening of novel biobased emollients. These studies showed that key attributes such as spreading, residual film thickness and slippery after-feel can be accurately predicted, providing a powerful tool for accelerating the eco-design of next-generation sustainable emollients. However, tribological tests are still affected by a lack of methodological harmonization, as friction coefficients and lubrication regimes depend strongly on probe geometry, substrate choice and testing conditions, complicating cross-study comparison and industrial translation.

Rheology-, texture- and tribology-guided formulation becomes indispensable not only for substituting classical synthetic ingredients, but also for rationally optimizing sustainable formulations without compromising sensory qualities expected by consumers. These considerations highlight how instrumental-based approaches not only support current eco-design strategies but also open new perspectives for addressing additional challenges related to formulation robustness, process variability and scale-up.

7. Process Optimization and Scale-Up of Cosmetic Emulsions

The transition from laboratory-scale prototyping to industrial-scale manufacturing represents one of the most critical phases in the development of cosmetic emulsions. This phase introduces changes in hydrodynamics, mechanical power input and heat transfer that can alter the droplet size distribution, viscoelasticity and physical stability of emulsions. In this context, instrumental rheology and texture analysis emerge as quantitative tools for monitoring structural evolution throughout the production steps and identifying the process parameters that most strongly influence the final microstructure. Their application goes far beyond the prediction of physical stability: these methods enable the rational optimization of emulsifying conditions, homogenization intensity, thickener activation, and cooling profiles, ultimately reducing batch failures, material waste, and energy demand, as well as CO₂ emissions in industrial production. From a life-cycle perspective, in fact, the manufacturing stage of cosmetic products is a major contributor to environmental impact, primarily through energy consumption, water use and associated greenhouse-gas emissions, often comparable to the impact of raw material choice itself [112].

Tamburic et al. [113] provided a compelling demonstration of how rheological and textural data can be used to validate low-energy manufacturing routes, specifically hot–cold and cold process emulsification, while ensuring structural equivalence to conventional hot processing. Through parallel testing at laboratory and semi-industrial scale of O/W and W/O emulsions, the authors showed that achieving comparable viscosity, yield stress, thixotropy and firmness across scales is essential not only for product performance but also for the accurate quantification of energy savings and related CO₂-equivalent reductions. Their results explicitly linked structural optimization with sustainable process design, showing that energy-efficient emulsification protocols can reduce thermal energy input by >80% while maintaining microstructural integrity. The LCA data showed that optimizing the manufacturing process has a far greater impact on cradle-to-gate emissions than ingredient substitution alone.

Through a rigorously controlled multiscale analysis, Calvo et al. [114] showed that impeller tip velocity and pumping capacity are the dominant scale-up determinants and that keeping these parameters constant across geometries enables structurally equivalent emulsions. Their results clearly demonstrated that rheology (G′ and viscosity) and texture parameters (firmness, consistency, cohesiveness) measured the impact of process hydrodynamics and dispersed-phase concentration with high sensitivity, providing a quantitative basis for predicting deviations arising during scale transfer. In a more recent contribution, the authors expanded the multiscale framework by coupling rheology, texture, formulation variables and droplet size distribution [115]. Their modeling approach described how droplets break under inertial and viscous forces during emulsification, allowing the prediction of the entire droplet size distribution as a function of thickener concentration and agitation rate. Key macroscopic properties, including elastic modulus and firmness, allowed establishing a direct quantitative link between microscale structure and product-level mechanical behavior.

Several limitations still restrict the broader applicability of the proposed approaches. Most investigations are conducted on a limited set of formulation types and under tightly controlled processing conditions, which may not fully capture the structural complexity of industrial cosmetic systems. Moreover, time-dependent phenomena such as coalescence, structural aging and process-induced heterogeneities remain largely unexplored. Addressing these gaps will likely require the integration of multiscale structural descriptors with advanced data-driven modeling strategies, enabling more robust and predictive control of product structure and performance during scale-up, supporting the development of low-impact, resource-efficient manufacturing pathways.

8. Future Perspectives and AI

Current research demonstrates that artificial intelligence (AI) and machine learning (ML) can fundamentally reshape the scientific workflow of cosmetic R&D by integrating high-resolution physico-chemical data derived from rheology, texture analysis and tribology. Cosmetic products are intrinsically multiscale systems, where formulation variables generate large, high-dimensional datasets, including droplet size distribution, flow indices, viscoelastic moduli, interfacial properties, and texture parameters, that are particularly well suited for ML architectures capable of extracting complex structure–property relationships. Rheological and textural fingerprints could serve as high-resolution digital markers to detect batch-to-batch variability, anticipate long-term stability, and monitor microstructural drift during processing or storage [116].

The construction of supervised-learning datasets enables direct mapping between instrumental mechanical parameters and human-perceived sensory attributes, providing quantitative foundations for predictive modeling in formulation design [117,118]. A recent work established a multivariate ML framework trained on an extensive dataset combining rheological, tribological and texture measurements with expert sensory evaluations of cosmetic creams, covering 22 sensory descriptors [49]. By testing multiple supervised algorithms, the authors showed that more than 80% of the sensory attributes could be predicted with over 95% accuracy, with instrumental parameters such as viscoelastic moduli, nonlinear LAOS metrics, creep-recovery features and friction coefficients as model inputs. The study provides one of the clearest demonstrations that ML and data-driven mathematical models, despite their non-mechanistic nature, can translate complex rheology and tribology signatures into sensory-relevant outputs, offering a scalable alternative to time-consuming and variable panel testing and enabling earlier, data-driven decision-making during formulation screening. Lee at al. [119] incorporated non-linear rheology and extensional texture parameters into a predictive model trained on the largest rheology–sensory database reported to date for cosmetic formulation. The authors demonstrated that parameters describing viscoelastic deformation under high shear and extensional thinning/break-up behavior were among the most critical determinants of perceived sensory texture, further supporting the role of formal predictive models in bridging instrumental measurements and in vivo perception. Evidence of mathematically formalized predictive modeling was already provided by Franzol et al. [120], who developed artificial neural network (ANN) models to predict consumer sensory perceptions of cosmetic emulsions from rheological inputs, including viscoelastic moduli, yield stress and thixotropy. Using a dataset of 39 commercial emulsions evaluated by consumer panels, the authors achieved prediction accuracy between 60% and 84% for key attributes such as spreadability, consistency and general impression.

Together, these studies exemplify the growing potential of AI models trained on rheology-, texture- and tribology-derived datasets to support formulation design, accelerate sensory optimization and reduce experimental burden. However, current AI- and ML-based approaches in cosmetic formulation remain subject to important limitations. As discussed in the cited literature, the reported models are trained on relatively constrained and formulation-specific datasets and the risk of overfitting remains significant, particularly for high-dimensional datasets combined with relatively small sample sizes. Moreover, the lack of standardized experimental protocols and harmonized sensory methodologies limits direct comparison between datasets and reduces model transferability. The studies discussed above should be regarded as proof-of-concept demonstrations, showing the feasibility of linking instrumental descriptors to sensory outcomes under controlled conditions rather than providing universally deployable solutions. Translation toward industrial applications will require larger and standardized datasets, robust experimental design, external validation across formulation spaces, and integration with existing R&D workflows. Addressing these challenges represents a critical step toward reliable, scalable and industry-ready predictive formulation platforms.

9. Conclusions

Rheology, texture analysis and tribology represent an integrated analytical framework that allows cosmetic researchers to quantitatively connect formulation structure with application behavior and sensory performance. Their combined use provides reliable criteria for prototype comparison, ingredient substitution and formulation refinement, ensuring that material decisions are grounded in measurable functional outcomes. However, the reliable prediction of after-feel and post-application sensations remains a key limitation, as these attributes are strongly influenced by skin–formulation interactions and individual subjectivity. At the same time, the sensitivity of these methods to formulation composition and processing history makes them essential tools for eco-design, supporting the identification of sustainable ingredients and the design of efficient manufacturing strategies. Predictive AI and ML models, built on large rheological and mechanical datasets, are beginning to anticipate texture, stability and microstructural evolution, enabling faster screening and more rational choices for product development and process optimization. Future progress will require larger rheology–tribology–sensory datasets and the development of advanced data-driven and neuroscience-based approaches, which were not addressed in the present review.