The simplest characterization of the transaction failure distribution ( tilde{D}left t,rright) ) is exponential and for current purposes, the single parameter exponential distribution provides an adequate characterization of Internet transaction failure over time.
The dispute rate (i.e., the rate at which some sort of transaction failure takes place that is sufficient for the traders to dispute the proper functioning of the market and its transactions. This will be represented by random variable ( tilde{mathrm{D}}left(mathrm{t},mathrm{r}right) ) which depends on the time of the transaction t and time between failures r. 2.
This implies an increased number of SIP packets to be managed (original packets and retransmissions), which first experience higher delays (in the order of hundreds of ms, since first retransmission occur after at least 500 ms), and then are discarded (service transaction failure), thus having a nearly constant delay independently on the offered load.
However, it should be emphasized that there are some particular applications such as mission critical operations, applications with high transaction failure costs such as merchant banking, or covert cyberse-curity applications where the benefits of using TPM and its attestation protocol overwhelms the costs.
Evidently, to avoid transaction failures in a Greenfield concession, the port authority needs to identify clearly the objectives undertaken.
Less frequent failures are desirable, and reducing the Internet transaction failures to zero are optimal.
Time between electronic market transaction failures ( tilde{mathrm{D}}left(mathrm{t},mathrm{r}right) ).
Time between electronic market transaction failures ( tilde{mathrm{D}}left(mathrm{t},mathrm{r}right) ) 2.
As expected, mean time to failure is r for the exponential distribution, and optimization on r implies that discounting of all future periods, and reducing the Internet transaction failures to zero are optimal.
