Physics-guided multi-objective mixture optimization for functional cementitious composites containing microencapsulated phase changing materials

[Display omitted] •A physics-guided optimization procedure is developed for the mixture design of functional cementitious composites.•Physics-based models are used to elucidate the composition-structure–property correlations.•A multi-attribute decision-making process is developed to account for competing performance and cost indices. A physics-guided multi-objective optimization procedure is developed for the mixture design of functional cementitious materials containing microencapsulated phase change materials (MEPCM). The mixture design procedure combines physics-based models with multi-objective optimization and decision-making methods to meet user’s demands on material’s mechanical and thermal properties, as well as the requirements for sustainability, functionalities, and cost. Physics-based models were utilized to draw the linkage between design variables and objective functions, including a hydration model to capture the hydration kinetics of slag-blended cement and a multiscale sub-stepping homogenization model to obtain the properties of cementitious composite. The multi-objective feasible enhanced particle swarm optimization (MOFEPSO) algorithm and the technique for preference by similarity to an ideal solution (TOPSIS) algorithm are used for mixture optimization and decision-making. The material design method is demonstrated through the design of functional cementitious composite materials containing two MEPCMs – i.e., a polymer encapsulated paraffin wax (PolyPCM) and a recently developed fly-ash cenosphere encapsulated PCM (CenoPCM). The design decision-making charts show the trade-offs among mechanical, thermal, and economic performances of cementitious composites containing MEPCMs. The mixture optimization and decision-making method can be used to assist the design of a variety types of functional cementitious composite and concrete.