The focus of this thesis is on the development of a conceptual design evaluation model that can be used in engineering design and can be implemented as a computer tool. A prerequisite to achieve this objective is a proper understanding of the initial phases of the design process, using an adequate framework. Hence, a brief examination of Theory of Technical Systems associated to a comprehensive study of the conceptual design stage, based on academic design methodologies and a survey amongst British industries, is presented. Additionally, evaluation issues at the early phases of design and a number of approaches for evaluating alternative solutions are investigated and relevant characteristics to be included in a prospective conceptual design evaluation model are compiled. A novel evaluation model based on function metrics has been proposed. The approach provides an intermediate evaluation, indicating which solutions have the potential to progress further in the design process The core of the model is the composition of evaluation matrices and computation of partial indices, which will originate an overall index used to classify the alternatives. The model assumes the existence of an explicit function structure on which the development of the organ/component structure is going to be based. A unique feature of this model is that it does not depend on designers' preferences or judgement in assigning values. From the formalised solution the designer has to identify which organ/component implements which function. An initial prototype of a computer tool (LiberSolutio), which embodies the above model, is presented. In addition to being an evaluation system, LiberSolutio can record the design history of the set of solutions generated for a particular functional decomposition/ structure. A preliminary evaluation of the model and computer system is also presented with conclusions drawn from the results obtained.