Asymptotic expansions of defective renewal equations with applications to perturbed risk models and processor sharing queues (Q604808)

In this paper an asymptotic expansion for the defective and excessive renewal equations, playing an important role in a number of applied probabilistic settings, are considered. The case of defective renewal equations as the defective parameter is close to one, so-called ``defectively perturbed renewal equations'', is analysed. The introduced asymptotic expansion of defective renewal equations can be applied to the analysis of processor sharing (PS) queues and the perturbed risk models. It allows us to obtain a numerical Laplace inversion to the transformation of the steady-state workload distribution. By differentiation all moments have been derived. This result yields a two-term corrected diffusion approximation for the workload distribution. In the author's opinion the resulting formulae are much easier and faster to compute than the original distribution of \(V(t)\).