Problems of dependability and possible solutions in the context of unique highly vital systems design

Abstract

Aim. The paper examines the problems caused by the conventional interpretation of dependability that prevent the practical use of dependability analysis (assessment) as a tool for engineers involved in the creation of unique highly vital systems and substantiates proposals for their resolution. Methods. The paper analyzes the problem of quantitative estimation of the dependability of unique highly vital systems without the use of probability statistical models. The view of dependability as a physical property of a product (as a result of changes in its internal state) allows at the physical level ensuring lasting capability to fulfil the required functions and quantitatively estimating the criteria of the required functions’ performance, that can be defined by, for instance, specifying a set of parameters for each function that characterize the capability to perform, as well as the permissible limits of such parameters’ variation. Such approach causes the requirement to take the origin of dependability into consideration and examine the causes of unlikely failures that are to be identified by means of additional analysis in parallel with calculations and experiments performed to support dependability. The solution to the problems of fuzzy terminology allows revealing the interrelation between the quality and the dependability, thus enabling using the single information basis of design and process engineering solutions the analysis, synthesis and assessment of the dependability of unique highly vital systems based on performance parameters without the use of probabilistic statistical models. Results. The solution of the above dependability-related problems allows ensuring dependability based on the physicality (causal connections) and physical necessity (consistency with the laws of nature) of the causes of failures. The dependability of unique highly vital systems must be ensured from the very early lifecycle stages based on consecutive execution of certain design, process engineering and manufacturing procedures, as well as application of engineering and design analysis of dependability, which also allows solving problems indirectly related to dependability, e.g. improving the quality and reducing the cost of the manufactured products. Conciusions. The paper shows that the application of design engineering methods for the dependability analysis (assessment) allows within the framework of existing views, yet with certain corrections solving dependability-related problems without the use of the mathematics of the classic dependability theory. High dependability can be achieved by the same ways as undependability comes about, i.e. through design and process engineering solutions. The analysis, substantiation of engineering solutions and specification of necessary and sufficient requirements for the manufacturing process allows achieving the target dependability by engineering means through higher quality of design and process engineering. If we regard dependability as a multiparametric property, parametric models of products can be developed that enable the evaluation of the temporal stability of parameter values using methods of individual design dependability and/or design engineering analysis of dependability. The principles of unity of the design concept and its implementation in manufacture enables the development of products and assessment of their dependability based on a single foundation, i.e. the design and process engineering solutions directly associated with the capabilities of a specific manufacturing facility.