Complexity estimation service for change management in industrial automation systems using Digital Twin

Abstract

Industrial automation systems are becoming increasingly complex. This complexity is caused by the higher proportion of software, networking, and cyber capability. Modern software-defined industrial automation systems provide higher flexibility and adaptivity against market uncertainty in dynamic and heterogeneous environmental conditions in industrial automation. Consequently, different conventional assets need changes during their life cycle to be adaptable. There are different ways to implement the required changes to reach the previously defined requirements such as systems’ software changes which have easier implementation. These options cause different levels of complexity that require a different amount of resources mainly time and cost. Achievement of the new requirements concurrently induces a large amount of complexity. The omnipresent complexity gives rise to many consequences on different abstraction levels in software-defined systems. In order to manage these software-defined systems properly, an automatic method for estimating this complexity due to the importance of a proper selection among different possible modifications is required. This paper proposes a novel concept for a complexity estimation service before applying real changes in the systems. The presented method enables the simulation of required changes using the Intelligent Digital Twin concept to decrease the cost of real changing and implementation of the software and their unknown consequences besides acquiring the future requirements of asset management. By automatic complexity drivers’ detection, this service simulates the new asset and predicts how it would be complex to apply the required changes by proper complexity metrics. To validate this service, an assistance system for a model car is implemented to show the complexity of software changes for adding new functionalities to the car.