Performance Evaluation of Thermal Insulation Materials from Sheep's Wool and Hemp Fibres.

In the current work, the performance properties of natural-fibre-based thermal insulation materials were examined. For this purpose, three different compositions of natural fibres were prepared: pure sheep wool (SW), wool and industrial hemp (SW/HF) fibres, and pure industrial hemp (HF) fibres. Low-melt bicomponent polylactide (PLA) fibres were used as a binding material. For specimens prepared from natural fibres, the dependence of the thermal conductivity, the tensile strength along and across the direction of product formation, and the short-term water absorption on the density of the specimens and the flammability parameters were determined. In addition, to reduce the water absorption and flammability, the specimens were coated with hydrophobic agents and flame retardants. The obtained research results were also statistically processed. The analysis of the results showed that the thermal conductivity of natural-fibre-based thermal insulation materials varied within the range of 0.0333 ÷ 0.0438 W/(m·K), the tensile strength varied from 2.5 to 130 kPa, the short-term water absorption varied from 0.5 to 8.5 kg/m2, and the water vapour diffusion resistance factor varied from 2.537 to 2.667. It was additionally determined that all the studied products were flammable. The water absorption and flammability values were significantly reduced by the use of hydrophobic agents and flame retardants.

Stochastic MCDM with normal approximation of the uniformly distributed variables for assessing sustainable insulation materials.

The sustainability-related decision-making oftentimes involves uncertain information. One of the key solutions in representing the interval information is the use of the interval information (numbers). This paper proposes a multi-criteria decision-making approach that relies on the algebra of random variables in handling the interval information. The interval information is provided in the form of the uniform distributions that are represented by mean and variance parameters. Following the central limit theorem, the normal approximation is involved. Then, the pair-wise comparisons are facilitated to establish the probabilities of dominance and rank the alternatives accordingly. The proposed approach allows for effectively handling the uncertainty and is user-friendly. The empirical application dealing with selection of the sustainable insulation materials is presented to validate the proposed approach. The technological, economic, environmental, and social facets of sustainability are considered when comparing the insulation materials. Sensitivity of the results is then checked via the Monte Carlo simulation.