Session 2: MBSA (Optimisation)

Session chair: Frank Ortmeier 

14:00 - 16:00 Wednesday 4th September 2013
14:00 Combined Optimisation of System Architecture and Maintenance
Authors: Shawulu Hunira Nggada, Yiannis Papadopoulos, David J. Parker
Abstract: The dependability analysis of safety-critical engineering systems during the early stages of design helps ensure that requirements are met and also reduces cost that may be incurred due to modifications later in the process. One attribute that can be analysed during the infancy of design is scheduled preventive maintenance (PM) through optimisation. PM is normally optimised in isolation to other system design improvement mechanisms. It is however helpful if PM could be optimised in combination with other mechanisms such improving the system's architecture. By combining these optimization objectives more design possibilities can be simultaneously explored, therefore helping to make better design decisions. In this paper we explore this type of optimization by using HiP-HOPS a scalable dependability analysis and optimisation tool.
14:30 Multi-objective Architecture Optimisation Modelling for Dependable Systems
Authors: Zhibao Mian, Leonardo Bottaci
Abstract: The design of dependable systems must address both cost and dependability (i.e. safety, reliability, availability and maintainability) concerns. For large systems, the design space of alternatives with respect to both dependability and cost is very large and automation is essential to explore this space. The model-based approach to the development and analysis of complex dependable systems is increasingly popular and recently, the Architecture Analysis and Design Language (AADL) has emerged as a potential future standard for model-based development of dependability-critical systems. The paper tackles the problem of describing, within an AADL model, the design space of alternatives. A new AADL property set is proposed for modelling component and system variability for cost and dependability optimisation. The proposed method is illustrated with an example of an AADL model of a safety critical embedded system. 
15:00 Implementing the functional requirements for determining the optimal arrangement of a distributed charging infrastructure
Authors: Tamás Kurczveil, Eckehard Schnieder
Abstract: The optimized operation of future traffic by intelligent control systems will need to take into account boundary conditions that arise from alternative drive concepts. New challenges will need to be mastered when it comes to corresponding energy systems, control of operations, and communication interfaces, such as needed for the sufficient energy supply of traffic participants. However, they will need to be conformed to existing systems, technologies, and infrastructure to allow the common operation and positioning of charging elements with minimum interference between different modes of transport. Funded by the German Federal Ministry of Transport, Building and Urban Development (Bundesministerium für Verkehr, Bau und Stadtentwicklung) the project emil (Elektromobilität mittels induktiver Ladung - electric mobility via inductive charging) will integrate an inductive vehicle charging system and a compatible prototype bus fleet into Braunschweig's traffic infrastructure. This paper describes the methodic approach and the implementation of functional requirements in a traffic simulation tool that are required for an evaluation of future urban road traffic with an increased rate of electric vehicles. The modifications can subsequently be used to determine the optimal placement of the corresponding charging infrastructure with consideration of conventional traffic demand.
15:30 Integrated Monitoring Tasks for the Safety of Critical Systems
Authors: Amer Dheedan, Ajith Kumar Parlikad
Abstract: Nuclear power plants, chemical processes and means of transportation are seen as critical systems, the failure of which may hazard lives and assets. Thus, the safety of such systems is rigorously considered and established during the design and operational stages. In the design stage, an off-line safety analysis investigates, retrofits and affixes whenever necessary fault-tolerant means and reliable components. In the operational stage, the functionality of systems is monitored through the delivery of three safety tasks: fault detection and diagnosis, alarm annunciation and fault controlling. However, systems still showing malfunctions and hazardous failures continue to be recorded. To address this issue, this paper develops a distributed on-line safety monitor. The monitor aims to achieve an effective integration among the delivery of the three safety tasks through the exploitation of the thorough and cost-effective abstraction of the off-line safety analysis and the distributed reasoning of a multi-agent system.