DIANA
is designed to pursue investigation on complex systems reliability at any levels
of designing and engineering service.
Diana provides good facilities for solving the following problems:
Evaluation of all the standard reliability indices with due regard to all failure and repair distributions adopted in reliability theory, which are based on statistic data on components and systems test results;
Optimization of system cost through close execution of reliability requirements as well as reliability optimization under limitation on joint cost;
Statistic Date Processing which allows to draw a histogram and to estimate parameters for any a priori given distribution using sample (including censored one).
A component of a system represents an object which can be in two states:
"operability state" and "breakdown state". Reliability work of a lower level
component is characterized by settings:
failure distribution;
repair distribution (if a component is repairable);
repairable component maintenance discipline.
Both failure distribution law and repair distribution law can be
specified by its distribution law:
exponential distribution;
Weibull;
normal (Gaussian) truncated;
lognormal;
gamma;
uniform;
or by a sample (including censored one).
Repairable component maintenance discipline consists of the list of the components
to be repaired in the particular repairing unit:
limitation on number components maintenance by the repairing unit at a time;
components priority specifying ordered service discipline.
There is a possibility of absolute priority assignment for a component,
guarantying urgent component maintenance. Interrupted remedial action component
can be put on a general queue depending on its priority or on a separate queue to resume.
The following reliability parameters can be used for optimization:
average mean-time-between-failures;
failure distribution function of nonrepairable system;
stationary and nonstationary system readiness indices;
failure and repair distribution functions of repairable system recovery;
no-failure system operation probability within the specified time interval;
stationary and nonstationary instantaneous system readiness indices.
DIANA allows to solve the following design probability problems:
reliability indices calculation;
solving direct and reverse problems of optimal redundancy, i.e. construction
of the cheapest system with specified reliability and of the most reliable
system within the limits of given cost;
choice of the optimal maintenance strategy in repairable system.
DIANA provides the following statistic to estimate reliability of the tested and operated systems:
plotting histogram of a failure distribution function taking into account
data on the tests results. Utilizing information on those tests which
has not been completed by component failure is admitted;
approximation of statistic data by any a supported distribution.
DIANA is designed by JV Dialogue and Operations Research Department's workers.
