Keynote

"The Role of Data for Safety Critical Systems Development and Validation"

Speaker: Nuno Silva - Program Manager at Critical Software, Coimbra, Portugal

 

Keywords: reliability and security data, development, validation, certification, RAMS, RCM

Abstract

The development and validation of safety critical systems is uniquely related to several sources of data. Not only the certification of these systems relies on data collection, historical data, components data reliability, risks quantification, etc, but also the existing data (reliability, risks, historical data, data for functional comparison, data for/from planning, data from testing and metrics) shapes the safety critical architectures and most decisions taken during the specification and development phases.

Projects or products data becomes extremely useful when dealing with safety critical systems, where not only quality, but also time-to-market, efficiency and dependability assurance are key to success. At Critical Software we base our engineering on several important sources of data:

  • Management related data (effort, time, risks quantification and effects, delays),
  • Results of the engineering processes (namely the issues found or the number of RIDs raised, then accepted by the customer, data from prototype usage and data collection from usage – keystroke dynamics as an example),
  • External data and field data (historical reliability/failure data, maintenance data, NPRD, EPRD, MIL-HDBK-217, etc).

The talk will briefly discuss how these sources of data are useful on an industrial perspective, namely for all the system development lifecycle phases, ranging from specification to maintenance, with special focus on RAMS (Reliability, Availability, Maintainability and Safety), RCM (Reliability Centered Maintenance) and certification support.

About Critical Software

Critical Software is a CMMi Level 5 company established in Coimbra, Portugal, and with offices in 4 continents. Being a level 5 company denotes a high process maturity and management level, where process optimization and continuous process improvement are based on data continuously collected and analysed. This management, process and project data addresses statistical common causes of process variation and changing the processes to improve performance.