Nonconvex Separable Programming Problem for Optimal Raw Material Mix in Flexible Polyurethane Foam Production

Abstract

Nonconvex Separable Programming (NSP)  for selecting optimal raw material mix for flexible polyurethane foam production (FPFP) was developd. With unit cost function () as objective function;,Density(P₁); compression set (P₂); elongation(P₃); hardness-index(P₄) and tensile strength (P₅) and other boundary conditions as constraints, an NSP foam raw material mix problem was defined and solved. Twelve existing formulation were used for validation .The cost and physical properties were determined and compared to the existing products using t-test. The optimal raw material mix 1.00, 0.4366, 0.0398, 0.0066, 0.0115, 0.0026, 0.0046 kg of polyol, toluene-di-isocynate, water, amine, silicone-oil, stannous-octoate and methylene-chloride respectively, were significantly different from  the existing formulations. The validated values of P₁, P₂, P₃, P₄, P₅ from optimally formulated foams were 23.83kgm^-3, 8.6%, 159.35%, 143.19N, and 117.33kNm^-2, respectively and conformed to standard. The associated costs per metric tonne of the optimal mixes were lower than that of existing mixes.

Keywords: Polyurethane foam, Optimal-mix, Separable  programming , nonconvex, elongation.