Quality Analysis Framework based on Complexity for Change Management Using Intelligent Digital Twin

Abstract

Software attributes analysis in industrial automation systems due to the higher proportion of using software-defined systems is highly significant. The requirement of higher flexibility, adaptivity, quality, less cost, and time to market as main classes of manufacturing attributes in dynamic and heterogeneous environments induce a large amount of complexity, which directly deteriorates the external software quality attributes. Complexity is a comprehension of quality management factor that influences the quality attributes of an asset such as testability, reusability, reliability, and maintainability. In order to manage the continuous requirement of software-defined systems’ updating efficiently, an automatic method for estimating these quality attributes is required. This paper proposes a novel approach for quality attributes analysis with the factor of complexity estimation. The proposed approach is realized in a framework that exploits the operational simulation capability of the intelligent Digital Twin concept to decrease the cost of untoward consequences of complexity increase on software external attributes by using operational data. It uses the simulated asset to predict how complexity due to applying new changes would impact the software quality attributes. The quality analysis process contains the intelligence process of monitoring the complexity metrics vector and quality attributes vector, the analysis of training models, the prediction of attributes, and the evaluation of measurements to enhance the quality attributes in advance. A comprehensive evaluation using the example of a Robotino 3 Premium from Festo, an autonomous mobile robot, is used to show the consequences of changes in software quality attributes for validation of the proposed framework.