This study combines multi-frequency dynamic mechanical analysis (DMA) with machine learning (ML) to characterize and predict the viscoelastic properties of a commercial polycarbonate/acrylonitrile–butadiene–styrene (PC/ABS) blend. DMA temperature sweeps at four frequencies (1–10 Hz) in single cantilever mode yielded a glass transition range of 115.8–$123.2$ °C ($E^{″}$ peak), frequency sensitivity of $7.18$ °C/$\mathrm{decade}$, and an apparent activation energy of $335\pm 85 \mathrm{k}\mathrm{J} {\mathrm{mol}}^{-1}$. Time–temperature superposition master curves were parameterized with a six-term Prony series ($R^2=0.998$). Four data-driven models (RF, XGB, SVR, MLP) and a physics-informed NeuralWLF model were evaluated through a hierarchical validation framework. Temperature-blocked CV ranked MLP ($\overline{R^2}=0.989$) above RF (0.950) for interpolation; LOFO validation revealed that NeuralWLF achieved the best cross-frequency generalization ($R^2>0.92$ for all targets) with interpretable WLF parameters ($C_1\approx 12.2$, $C_2\approx 51.7 °\mathrm{C}$). A systematic block size sweep (5–30 $°\mathrm{C}$) revealed a validation inflation effect in which MLP $tan\delta$$R^2$ dropped from 0.986 to 0.592 as the gap-to-FWHM ratio increased from 0.5 to 3.1, establishing the gap/FWHM ratio as a quantitative validation stringency criterion. A physics–data crossover was identified at gap/FWHM $\approx 2$: beyond this threshold, NeuralWLF outperformed all data-driven models in $tan\delta$ prediction by up to $+0.300$ in $R^2$, while curriculum learning (freezing the WLF layer for 300 epochs) further improved the most stringent 30 $°\mathrm{C}$ validation from $R^2=0.660$ to 0.731. The integrated framework demonstrates that honest evaluation of DMA–ML models requires validation gaps exceeding the characteristic feature width and introduces a quantifiable physics-data crossover criterion for selecting between data-driven and physics-informed architectures.

Multi-sensor systems require precise time synchronization for accurate data fusion. However, currently prevalent software time synchronization methods often rely on clocks provided by the Global Navigation Satellite System (GNSS), which may not offer high accuracy and can be easily affected by issues with GNSS signals. To address this limitation, this study introduces a novel laser pulse-driven time synchronization (LPTS) method in our custom-developed Light Detecting and Ranging (LiDAR) system. The LPTS method uses electrical pulses, synchronized with laser beams as the time synchronization source, driving the Micro-Controller Unit (MCU) timer within the control system to count with a timing accuracy of 0.1 $\mathsf{\mu }$s and to timestamp the data from the Positioning and Orientation System (POS) unit or laser scanner unit. By employing interpolation techniques, the POS and laser scanner data are precisely synchronized with laser pulses, ensuring strict correlation through their timestamps. In this article, the working principles and experimental methods of both traditional time synchronization (TRTS) and LPTS methods are discussed. We have implemented both methods on experimental platforms, and the results demonstrate that the LPTS method circumvents the dependency on external time references for inter-sensor alignment and minimizes the impact of laser jitter stemming from third-party time references, without requiring additional hardware. Moreover, it elevates the internal time synchronization resolution to 0.1 $\mathsf{\mu }$s and significantly improves relative timing precision.

Die swell is the dominant source of dimensional deviation in rubber profile extrusion. Because it is driven by recoverable elastic strain, a purely viscous baseline flow field cannot reproduce its speed dependence; a viscoelastic correction is required. This study presents, to the best of our knowledge, the first controlled comparison of a Carreau–Arrhenius baseline flow field against a fractional-order viscoelastic correction for carbon-black-filled EPDM across an industrial speed window. The viscoelastic correction (PyCFD-FMM) is a post-processing fractional-order viscoelastic swell correction built on the shared non-isothermal Polyflow Carreau–Arrhenius flow field, derived from a six-mode fractional Maxwell model parameterized from dynamic mechanical analysis via the Laun rule and closed through the Tanner recoverable-strain theory. Three carbon-black-filled EPDM compounds (Shore A 60–80) were extruded at four screw speeds (15–30 rpm) under instrumented conditions. Experimentally, swell ratios of 1.12–1.15 increase monotonically with screw speed (Fisher-combined $p=0.007$; measurement repeatability CV $\le 0.27$% across $n=4$ replicates per condition). The purely viscous baseline output gives a decreasing apparent swell–speed trend—opposite to experiment—whereas PyCFD-FMM recovers the correct increasing trend for all compounds. Under single-anchor hold-out evaluation at 20/25/30 rpm, the non-anchor MAPE decreases from 0.99% for the baseline flow-field output to 0.30% (PyCFD-FMM); an anchor-sensitivity check over all four rpm choices keeps the compound-averaged non-anchor MAPE within 0.27–0.39% and preserves the correct slope sign in every case. Swell decomposition into geometric baseline and net correction factor ($B_{\mathrm{PyCFD}}=B_{\mathrm{geom}}\times f_{\mathrm{corr}}$) confirms that the viscous baseline flow field captures flow-geometry effects but carries no elastic memory. Within the tested window, the viscoelastic correction meets a dual-gate criterion—correct slope sign and reduced non-anchor MAPE—which the purely viscous baseline cannot satisfy by construction.

Filled elastomers often exhibit a low-frequency power-law storage modulus (G-prime), yet quantitative links between molecular interfacial structure and macroscopic reinforcement remain unresolved. This gap is addressed using a hierarchical multiscale framework that integrates coarse-grained molecular dynamics (MD) and dynamic mechanical analysis (DMA). Overall, MD contributes transferable structural descriptors rather than direct macro-rheology prediction. MD simulations yield a bound-layer scaling relation for chain length $N=50$ in coarse-grained simulations serving as a structural probe. For EPDM master curves, the single-phase fractional Maxwell model is statistically preferred (Delta AICc > 147, n = 56), reflecting limited statistical power; larger datasets (e.g., PC/ABS, n = 952) favor the dual-phase formulation. For cross-scale prediction, an MD-derived effective-volume-fraction baseline (MAPE = 54.1%) provides a structural prior; the regime-partitioned bridge model absorbs relaxation physics not resolved at the MD scale, reducing error to 7.3% (blocked-CV MAPE = 9.5%, with a 2.3% fold-to-fold spread). Linear-viscoelastic constraints improve nonlinear PTT calibration, reducing die-swell error by 87%.

Rubber extrusion has evolved through overlapping advances in equip design, rheological characterization, numerical modeling, sensing, and control. This structured critical narrative review synthesizes 180 sources published from 1972 to 2026, assembled through iterative keyword searching and backward and forward citation tracing. The conventional three-zone theory of solid conveying, compression, and metering is used as a bounded analytical framework, while the literature is organized into four overlapping analytical periods spanning empirical design, constitutive and numerical modeling, engineering-scale simulation, and multiphysics and data-enabled methods. Evidence is distinguished among direct rubber-extrusion validation, rubber-material or rheological studies, transferable general polymer extrusion studies, and enabling computational, sensing, or control research. Five persistent challenges are identified: formulation-dependent nonlinear rheology; incomplete representation of filler-network evolution and wall slip; limited cross-machine and cross-formulation validation; high computational cost; and the lack of standardized datasets and reporting protocols. Digital twins, physics-informed neural networks, and neural operators are promising but remain insufficiently validated for industrial rubber extrusion. Priorities include transparent benchmark datasets, evidence-graded, uncertainty- and latency-aware validation, multimodal sensing, transferable reduced-order and learned models, and measurable sustainability indicators. Together, these priorities define a validation-oriented roadmap for more reliable, transferable, and sustainable rubber-extrusion modeling and control.

The polymer–filler interphase in filled elastomers is often represented by a single thickness, obscuring internal heterogeneity. Coupling coarse-grained molecular dynamics with dynamic mechanical analysis of EPDM/carbon-black compounds, we resolve a bimodal bound-rubber layer with a dense inner zone set by surface adsorption and a looser outer zone sustained by chain connectivity. Heating contracts the outer zone about twice as strongly as the inner zone (outer: 26.5%, 95% confidence interval 17.4–34.8%; inner: 13.3%). Per-layer mean-squared displacement analysis shows a modest mobility gradient between the 1–2 nm outer zone and the bulk. Dynamic mechanical analysis at 120–140 °C shows a flatter reinforcement factor at higher temperature, consistent with interphase-linked thermal contraction. Lengthening the chain at fixed filler loading markedly enlarges the bridging fraction and the cumulative excess thickness, signaling a transition from adsorption-limited to connectivity-limited reinforcement. These results show that a single interphase boundary can miss a dynamically active outer zone relevant to reinforcement and thermal aging in filled elastomers.