A practical dependability measure for embedded computer systems
Artikel i vetenskaplig tidskrift, 1993
Dependability is commonly described by a number of attributes, such as reliability, availability, safety and security. Quantitative measures are found for each
separate attribute e.g. reliability and availability, but are not defined for the totality of all attributes. This paper suggests a vectorized measure based on Markov processes.The measure covers reliability, safety and a modified version of the security attribute. It should be used for practical dependability trade-offs and is especially applicable to autonomous systems with embedded computers, such as aerospace vehicles and
control systems. Key issues are the concepts of degradability, subservice and service
level. The measure is based on the expected operating time on an operational service
level and the total operational time before failure for failed service levels.
embedded computer systems