Propuestas Tecnológicas de Autocorrección de ejercicios de modelado 3D = Technological Proposals for self-assessment of 3D modelling exercises

Nowadays, there are several contrasted procedures and other proposals [1] for the self-assessment of exercises or exams of subjects which are evaluated using numerical exercises. Intermediate or final values are compared, and an automatic qualification of self-assessment is assigned. It is possible to extend this classic correction procedure by the teacher [2]. The automatic assessment of exercises based on texts is more complicated because the appearance of certain keywords or their synonyms could offer a possible approach as a mechanical assessment of those exercises. However, the difficulty in the assessment of these exercises is the interpretation of their meaning [3].In the case of 2D graphic exercises, which are typical of technical drawing, the problem is very different, since there are no alphanumeric chains to compare. Similarities between images and the comparison of primitive entities (vector objects) may be possible ways for evaluation [4]. The problem is more complicated when we want to evaluate 3D models mechanically.This article presents a compilation of possible procedures to use in the generation of a self-assessment tool for industrial solid modelling exercises, that is, of mechanical parts [5]. In these cases, certain parameters such as volumes, surfaces, centres of gravity or moments of inertia can be a first approximation to their corrections [6]. These evaluations could continue with the analysis of the constructive operations that exist in the modelling of the object, such as solid parts, emptying, holes, threading, etc., all of them included in their modelling trees or lists of operations. The generation of a utility that helps in the correction of 3D modelling exercises would be of great interest, since it would bring effectiveness and agility to the evaluation process, as well as greater objectivity when using a computer system that isolates similarity factors and implements rules of measurable evaluation automatically.