Identification of sources of failures and their propagation in critical infrastructures from 12 years of public failure reports

Thirty years of software problems in the news

“…We began this paper by observing that critical software problems in security [7,10], financial calamities [23], software-based infrastructure [24], and automotive recalls [18] are on the rise. Our findings provide a broader picture for interpreting these domain-specific trends:…”

“…Most water, gas, and electricity failures now involve software failures [24], stock markets can now crash in a flash [27], race conditions can leave millions without power [31], and usability problems can cost lives [16]. From a productivity perspective, software problems can also be quite expensive: one estimate of the annual costs in lost time and labor due to software problems is nearly $40 billion per year [26] and companies like Microsoft regularly spend hundreds of millions of dollars removing defects from a single release of Windows [20].…”

“…Security breaches are now more prevalent and more costly than ever before [7,10]; spreadsheet errors are leading to an increasing number of financial calamities [23]; software infrastructure failures rose between 1994 and 2005 [24]; and cars are recalled more frequently than ever because of software defects [18].…”

Modelling Interdependencies Between the Electricity and Information Infrastructures

“…), c) the couplings among the infrastructures and their effects on their response behaviour (loose or tight, inflexible or adaptive), d) the infrastructure characteristics (organisational, operational, temporal, spatial), e) the state of operation (normal, stressed, emergency, repair), the degree to which the infrastructures are coupled, f) the type of failure affecting the infrastructures (commoncause, cascading, escalating). Other classifications have also been proposed in [4][5][6]29]. In particular, the study reported in [29], based on 12 years public domain failure data, provides useful insights about the sources of failures affecting critical infrastructures, their propagation and their impact on public life, considering in particular the interdependencies between the communication and information technology infrastructure and other critical infrastructures such as electricity, transportation, financial services, etc.…”

“…Other classifications have also been proposed in [4][5][6]29]. In particular, the study reported in [29], based on 12 years public domain failure data, provides useful insights about the sources of failures affecting critical infrastructures, their propagation and their impact on public life, considering in particular the interdependencies between the communication and information technology infrastructure and other critical infrastructures such as electricity, transportation, financial services, etc.…”