pith. sign in

arxiv: 2605.02895 · v2 · pith:ENSSUIMInew · submitted 2026-03-20 · 🧮 math.OC · math.PR

Preventive Maintenance of a Two-Unit Priority Standby System with Repair

Alexandros Carballo (UH) , Marelys Crespo , Jos\'e E Vald\'es (UH) This is my paper

Pith reviewed 2026-05-15 09:08 UTC · model grok-4.3

classification 🧮 math.OC math.PR
keywords preventive maintenancestandby redundancymean time to failurebathtub hazard rateoptimal maintenance timemean residual lifestochastic ordering
0
0 comments X

The pith

Preventive maintenance improves mean time to system failure in two-unit priority standby systems when the priority unit has a bathtub or upside-down bathtub hazard rate.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper examines a two-unit priority standby system that includes a repair facility and preventive maintenance on the priority unit. It derives necessary and sufficient conditions under which performing maintenance increases the overall mean time to system failure. When the priority unit's hazard rate follows a bathtub or upside-down bathtub shape, the analysis identifies a threshold time after which maintenance becomes beneficial and an optimal maintenance time, both given explicitly through the hazard rate and the mean residual life function. Stochastic ordering methods are applied to compare mean lifetimes across independent systems and to rank their threshold and optimal times.

Core claim

In the general case necessary and sufficient conditions are obtained for maintenance to raise the mean time to system failure. When the priority unit hazard rate is bathtub-shaped or upside-down bathtub-shaped, a threshold time exists beyond which maintenance improves that mean time, together with an optimal maintenance time; both quantities admit explicit expressions in terms of the hazard rate and the mean residual life function. Stochastic ordering further permits direct comparison of mean lifetimes, threshold times, and optimal maintenance times between two independent copies of the system.

What carries the argument

The bathtub or upside-down bathtub shape of the priority unit's hazard rate, which governs the sign changes of the difference between the mean residual life with and without maintenance and thereby determines the threshold and optimal maintenance times.

Load-bearing premise

The priority unit must have a hazard rate that is either bathtub-shaped or upside-down bathtub-shaped.

What would settle it

A concrete counter-example in which a priority unit with non-bathtub hazard rate still shows maintenance improving mean time to failure exactly when the derived conditions predict it should not.

Figures

Figures reproduced from arXiv: 2605.02895 by Alexandros Carballo (UH), Jos\'e E Vald\'es (UH), Marelys Crespo.

Figure 1
Figure 1. Figure 1: BFR example with γ1 = 0.001, γ2 = 4 and λ = 1.0. 4.2 Upside down bathtub-shaped hazard rate Unlike the BFR case, several classical distributions exhibit an upside-down bathtub-shaped hazard rate, including the lognormal and Pareto distributions. We now consider the following example. Let Z1 and Z2 be independent exponential random variables with rate parameters β1 > 0 and β2 > 0, respectively. That is, Zi … view at source ↗
Figure 2
Figure 2. Figure 2: UBFR example with γ1 = 0.01, γ2 = 6 and λ = 0.1. 5.1 Ordering results for mean time to failure The following proposition establishes comparisons in terms of the mean time to failure, assuming that maintenance for both systems is scheduled at time T > 0 and that the standby unit lifetimes, repair times, and maintenance times are identically distributed. We denote by Mi(T) the mean time to failure of the sys… view at source ↗
read the original abstract

Optimal maintenance policies play an important role in the reliability analysis of repairable systems. This paper examines a two-unit priority standby system with a repair facility, where the priority unit is subject to preventive maintenance. In the general case, we derive necessary and sufficient conditions under which maintenance increases the mean time to system failure. When the hazard rate of the priority unit has either a bathtub-shaped or an upside-down bathtub-shaped form, we establish conditions for the existence of both a threshold time, beyond which maintenance improves the mean time to system failure, and the optimal maintenance time. Expressions to find these quantities are provided in terms of the hazard rate and the mean residual life function. Furthermore, stochastic ordering techniques are used to compare two independent systems under this model, along with their respective threshold times and optimal maintenance times.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript examines preventive maintenance in a two-unit priority standby system with repair. In the general case it derives necessary and sufficient conditions under which maintenance improves mean time to system failure. When the priority unit has a bathtub-shaped or upside-down bathtub-shaped hazard rate, it establishes conditions guaranteeing a threshold time beyond which maintenance is beneficial and an optimal maintenance time, together with explicit expressions for both quantities in terms of the hazard rate and mean residual life function. Stochastic ordering is applied to compare mean lifetimes, threshold times, and optimal times across independent systems.

Significance. If the derivations hold, the paper supplies concrete, usable conditions and closed-form expressions for when and how preventive maintenance improves system reliability in a standard standby model. The explicit dependence on hazard rate and mean residual life, combined with stochastic-order comparisons, strengthens the practical value for maintenance scheduling under common aging patterns.

minor comments (3)
  1. [§2.2] §2.2: the system-state definitions and transition rates would benefit from an accompanying state-transition diagram to make the priority and standby logic immediately visible.
  2. [Eq. (15)] Eq. (15): the integral expression for the improvement in MTTSF contains an implicit assumption that the mean residual life is differentiable at the maintenance epoch; a brief remark on the required regularity would remove ambiguity.
  3. [§4.3] §4.3: the proof that the optimal time is unique when the hazard is upside-down bathtub relies on a sign-change argument; adding a short sentence confirming that the second-derivative test or monotonicity of the derivative is verified would strengthen readability.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary of our manuscript and for recommending minor revision. The referee's description accurately captures the derivation of necessary and sufficient conditions for preventive maintenance to improve mean time to system failure, the explicit results for bathtub and upside-down bathtub hazard rates, and the stochastic ordering comparisons.

Circularity Check

0 steps flagged

No significant circularity; derivations follow from general hazard rate properties

full rationale

The central results derive necessary and sufficient conditions for maintenance to improve mean time to system failure, plus existence of threshold and optimal times, directly from the bathtub or upside-down bathtub shape assumptions on the priority unit's hazard rate combined with the mean residual life function. These steps use standard stochastic ordering and reliability identities without any fitted parameters renamed as predictions, without self-definitional loops, and without load-bearing self-citations that reduce the claim to prior unverified work by the same authors. The model assumptions (independent lifetimes and repairs) are stated explicitly and the explicit expressions are obtained by algebraic manipulation of the hazard and residual life, keeping the derivation self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper rests on standard reliability-theory assumptions about hazard-rate shapes and system independence; no free parameters or new entities are introduced in the abstract.

axioms (2)
  • standard math Hazard rate functions are continuous and admit a mean residual life function.
    Required to derive explicit expressions for threshold and optimal times.
  • domain assumption Units fail and are repaired independently with the priority unit having higher importance.
    Core modeling choice for the two-unit priority standby system.

pith-pipeline@v0.9.0 · 5442 in / 1359 out tokens · 52795 ms · 2026-05-15T09:08:10.689948+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.