Models of construction attributes selection process in ordered construction families

Purpose: The main aim of research was to elaborate models of construction attributes selection process in ordered construction families. Design/methodology/approach: Based on selection models of quantitative construction attributes in reference to unified characteristic attributes it is possible to generate ordered construction families like: series of types, modular systems. Findings: Selection models of quantitative construction attributes were analyzed. Those methods are based on construction congruence. Research limitations/implications: The neural network model limitations come from the neural network structure. The accuracy of neural network work results vary of neural network quality. The constructional similarity model can be applied only to that states which are described by similarity theory. Practical implications: Presented method was applied in grippers series of types generation process. Originality/value: Described models give simple and quick ways to select construction attributes in ordered construction families.