At each stage of the product development process, the designers are facing an important task which consists of decision making. Two cases are observed: the problem of concept selection in conceptual design phases and, the problem of pre-dimensioning once concept choices are made. Making decisions in conceptual design phases on a sound basis is one of the most difficult challenges in engineering design, especially when innovative concepts are introduced. On the one hand, designers deal with imprecise data about design alternatives. On the other hand, design objectives and requirements are usually not clear in these phases. The greatest opportunities to reduce product life cycle costs usually occur during the first conceptual design phases. The need for reliable multi-criteria decision aid (MCDA) methods is thus greatest at early conceptual design phases. Various MCDA methods are proposed in the literature. The main criticism of these methods is that they usually yield different results for the same problem. In this work, an analysis of six MCDA methods (weighed sum, weighted product, Kim & Lin, compromise programming, TOPSIS, and ELECTRE I) was conducted. Our analysis was performed via an industrial case of solar collector structure development. The objective is to define the most appropriate MCDA methods in term of three criteria: (i) the consistency of the results, (ii) the ease of understanding and, (iii) the adaptation of the decision type. The results show that TOPSIS is the most consistent MCDA method in our case.< Réduire