| 1. |
- Almgren, Torgny, 1962, et al.
(författare)
-
Optimization of opportunistic replacement activities: A case study in the aircraft industry
- 2007
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- In the aircraft industry maximizing availability is essential. Maintenance schedules must therefore be opportunistic, incorporating preventive maintenance activities within the scheduled as well as the unplanned ones. At the same time, the maintenance contractor should utilize opportunistic maintenance to enable the minimization of the total expected cost to have a functional aircraft engine and thus to provide attractive service contracts. This paper provides an opportunistic maintenance optimization model which has been constructed and tested together with Volvo Aero Corporation in Trollhättan, Sweden for the maintenance of the RM12 engine. The model incorporates components with deterministic as well as with stochastic lives. The replacement model is shown to have favourable properties; in particular, when the maintenance occasions are fixed the remaining problem has the integrality property, the replacement polytope corresponding to the convex hull of feasible solutions is full-dimensional, and all the necessary constraints for its definition are facet-inducing. We present an empirical crack growth model that estimates the remaining life and also a case study that indicates that a non-stationary renewal process with Weibull distributed lives is a good model for the recurring maintenance occasions. Using one point of support for the distribution yields a deterministic replacement model; it is evaluated against classic maintenance policies from the literature through stochastic simulations. The deterministic model provides maintenance schedules over a finite time period that induce fewer maintenance occasions as well as fewer components replaced.
|
|
| 2. |
|
|
| 3. |
- Anevski, Dragi, 1965, et al.
(författare)
-
A general asymptotic scheme for inference under order restrictions
- 2006
-
Ingår i: Annals of Statistics. - : Institute of Mathematical Statistics. - 0090-5364. ; 34:4, s. 1874-1930
-
Tidskriftsartikel (refereegranskat)abstract
- Limit distributions for the greatest convex minorant and its derivative are considered for a general class of stochastic processes including partial sum processes and empirical processes, for independent, weakly dependent and long range dependent data. The results are applied to isotonic regression, isotonic regression after kernel smoothing, estimation of convex regression functions, and estimation of monotone and convex density functions. Various pointwise limit distributions are obtained, and the rate of convergence depends on the self similarity properties and on the rate of convergence of the processes considered. © Institute of Mathematical Statistics, 2006.
|
|
| 4. |
- Anevski, Dragi, et al.
(författare)
-
A stochastic process approach to multilayer neutron detectors
- 2019
-
Ingår i: Scandinavian Journal of Statistics. - : Wiley. - 0303-6898 .- 1467-9469. ; 46:2, s. 621-635
-
Tidskriftsartikel (refereegranskat)abstract
- The sparsity of the isotope Helium-3, ongoing since 2009, has initiated a new generation of neutron detectors. One particularly promising development line for detectors is the multilayer gaseous detector. In this paper, a stochastic process approach is used to determine the neutron energy from the additional data afforded by the multilayer nature of these novel detectors. The data from a multilayer detector consist of counts of the number of absorbed neutrons along the sequence of the detector's layers, in which the neutron absorption probability is unknown. We study the maximum likelihood estimator for the intensity and absorption probability and show its consistency and asymptotic normality, as the number of incoming neutrons goes to infinity. We combine these results with known results on the relation between the absorption probability and the wavelength to derive an estimator of the wavelength and to show its consistency and asymptotic normality.
|
|
| 5. |
- Anevski, Dragi, et al.
(författare)
-
Estimating a probability mass function with unknown labels
- 2017
-
Ingår i: Annals of Statistics. - 0090-5364. ; 45:6, s. 2708-2735
-
Tidskriftsartikel (refereegranskat)abstract
- In the context of a species sampling problem, we discuss a nonparametric maximum likelihood estimator for the underlying probability mass function. The estimator is known in the computer science literature as the high profile estimator. We prove strong consistency and derive the rates of convergence, for an extended model version of the estimator. We also study a sieved estimator for which similar consistency results are derived. Numerical computation of the sieved estimator is of great interest for practical problems, such as forensic DNA analysis, and we present a computational algorithm based on the stochastic approximation of the expectation maximisation algorithm. As an interesting byproduct of the numerical analyses, we introduce an algorithm for bounded isotonic regression for which we also prove convergence.
|
|
| 6. |
- Anevski, Dragi
(författare)
-
Estimating the derivative of a convex density
- 2003
-
Ingår i: Statistica Neerlandica. - : Wiley. - 0039-0402 .- 1467-9574. ; 57:2, s. 245-257
-
Tidskriftsartikel (refereegranskat)abstract
- We obtain a relation between the time between two bird-catchings and the total resting period of a bird, leading to the problem of estimating the derivative of a convex density. We state a fundamental result on the nonparametric maximum likelihood estimator of a convex density. Further, we derive the optimal rate in the minimax risk sense for estimating the derivative of a convex density.
|
|
| 7. |
- Anevski, Dragi
(författare)
-
Functional central limit theorems for the Nelson–Aalen and Kaplan–Meier estimators for dependent stationary data
- 2017
-
Ingår i: Statistics and Probability Letters. - : Elsevier BV. - 0167-7152. ; 124, s. 83-91
-
Tidskriftsartikel (refereegranskat)abstract
- We derive process limit distribution results for the Nelson–Aalen estimator of a hazard function and for the Kaplan–Meier estimator of a distribution function, under different dependence assumptions. The data are assumed to be right censored observations of a stationary time series. We treat weakly dependent as well as long range dependent data, and allow for qualitative differences in the dependence for the censoring times versus the time of interest.
|
|
| 8. |
|
|
| 9. |
- Anevski, Dragi, et al.
(författare)
-
Limit properties of the monotone rearrangement for density and regression function estimation
- 2019
-
Ingår i: Bernoulli. - 1350-7265. ; 25:1, s. 549-583
-
Tidskriftsartikel (refereegranskat)abstract
- The monotone rearrrangement algorithm was introduced by Hardy, Littlewood and Po ́lya as a sorting device for functions. As- suming that x is a monotone function and that an estimate xn of x is given, consider the monotone rearrangement xˆn of xn. This new estimator is shown to be uniformly consistent. Under suitable as- sumptions, pointwise limit distribution results for xˆn are obtained. The framework is general and allows for weakly dependent and long range dependent stationary data. Applications in monotone density and regression function estimation are detailed.
|
|
| 10. |
- Anevski, Dragi, et al.
(författare)
-
Monotone regression and density function estimation at a point of discontinuity
- 2002
-
Ingår i: Journal of Nonparametric Statistics. - : Informa UK Limited. - 1048-5252 .- 1029-0311. ; 14:3, s. 279-294
-
Tidskriftsartikel (refereegranskat)abstract
- Pointwise limit distribution results are given for the isotonic regression estimator at a point of discontinuity. The cases treated are independent data, phi- and alpha-mixing data and subordinated Gaussian long range dependent data. Pointwise limit results for the nonparametric maximum likelihood estimator of a monotone density are given at a point of discontinuity, for independent data. The limit distributions are non-standard and differ from the ones obtained for differentiable regression and density functions.
|
|
