Inter-arrival distribution

Definition 1.4 Unless explicitly stated, the inter-arrival distribution, or return distribution, is a discrete sub-probability density K -) on N which may be written as... 

We give now a scaling limit result in the strongly delocalized regime under the condition that if(-) decays sufficiently rapidly. One is certainly tempted to conjecture the validity of such a result in the whole class of inter-arrival distributions that we consider and everywhere in T>. We will come back to this issue in the next section (with no definite answer). 

Denote by Pb the law of the renewal process with inter-arrival distribution Kb -)- We will show below that... 

Consider, for example, a trial in four centres in which it is determined to recruit 96 patients in total. Suppose that in each centre recruitment follows a Poisson distribution with mean arrival rate of 24 patients per year. Mean inter-arrival time is then 1/24 years or 0.042 year. The time to recruit the 24th patient in a given centre will be given by the gamma distribution with parameters 0.042 and 24. The mean recruitment time for a centre will be 1 year and the median will be 0.986 year. The probability of completing in one year or less in a given centre will be 0.527. (Details of the necessary calculations to support these assertions are not given here. Let us accept that they are true.)... 

It is assumed that each centre contributes patients according to a common stable Poisson process with intensity A per year. It is required to recruit N patients in total in k centres. Thus inter-arrival times in a given centre are exponential with mean 1 fX and we require the time, t, to recruit patient number in a given centre i. This is given by the gamma distribution with parameters B= /X and C = n. The probability density is... 

Table A. 14 Distributions of inter-arrival time and deliver quantities for external customers / suppliers...

In [34], Chen and Simchi-Levi consider the infinite horizon model with stationary parameters and general demand processes. They show that in this case, the (s, 5, p) policy identified by Thomas is optimal under both the average and discounted expected profit criteria. They further consider the problem with continuous inventory review in [35], and show that a stationary (s, S,p) policy is optimal for both the discounted and average proft models with general demand functions and general inter-arrival time distribution. 

The next two variables are represented for an Exponential PDF that is symbolized by the equation (9), where x represents the continuous variable of the distribution and is equivalent to the impulses duration w, while the inter-arrival times is expressed by r. 

Fig. 14. Exponential function distribution for inter-arrival times between impulses a) PDF b) Exponential probability...

The two can be related using a concept called a Poisson process. From the Wikipedia s article on the Poisson Process In probability theory, a Poisson process is a stochastic process which counts the number of events and the time that these events occur in a given time interval. The time between each pair of consecutive events has an exponential distribution with parameter G the parameter is the occurrence rate per unit time] and each of these inter-arrival times is. .. independent of other... 

Now the renewal, impulse process excitation is considered where the inter-arrival times are the sum of two independent, negative expcMiential distributed variables q and Td, with probalnlity density functions given, respectively, by (for t > 0)... 

Key reference for this section is [Asmussen (2003)]. Given the discrete probability density K -) on N, t = Tj j=o,i,... is the sequence of partial sums of an IID sequence of variables distributed according to K -), that is To = 0 and the sequence of inter-arrival times vj — Tj i j=i 2,... is HD, Ti K -). We call r renewal process on N associated to K -). r may be also viewed as a point process, that is as a random subset of N U 0, so, for example, n t means that there exists j such that Tj = n. Actually, the points in r will be called renewals or renewal points. If tq is a N-valued random variable independent of r, then we call tq + r delayed renewal process and tq is the delay. Delayed or not, renewal processes enjoy the... 

We call r-fc/V the law of r n 0,1,..., N when the inter-arrival times are distributed like Kh -)- We want to compute the relative entropy density of Vb,N with respect to z/q Af. 

---