Spare Strategy Analysis and Design for Large-Scale Satellite Constellation Using Markov Chain
Seungyeop Han, Zachary Grieser, Shoji Yoshikawa, Takumi Noro, Takumi Suda, Koki Ho · 2025
Read Spare Strategy Analysis and Design for Large-Scale Satellite Constellation Using Markov Chain by Seungyeop Han, Zachary Grieser, Shoji Yoshikawa, Takumi Noro, Takumi Suda, Koki Ho (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-09957.txt.
Introduction
Seungyeop Han and 5 co-authors use Markov chains to plan spare satellites for giant constellations in a 2025 aerospace paper. A mega constellation cannot afford grounded service while replacements crawl upward, so operators park spares in holding orbits and shuttle them into working planes as failures strike. Judging such a two-storey stock system over the long run is the paper's first job.
Failures and restocking in each storey become separate Markov chains, joined by a fixed-point loop until their steady states agree. That joint picture captures the random interplay honestly, without the flattening shortcuts older studies needed. A genetic algorithm then searches design choices on top of the fast analysis. A realistic case study closes the paper, showing the machinery guiding real architecture decisions.
Catalog record
| CREATOR | Seungyeop Han, Zachary Grieser, Shoji Yoshikawa, Takumi Noro, Takumi Suda, Koki Ho |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 69 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-09957
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Spare Strategy Analysis and Design for Large-Scale Satellite Constellation Using Markov Chain Seungyeop Han, Zachary Grieser, Shoji Yoshikawa, Takumi Noro, Takumi Suda, Koki Ho Abstract This paper presents a method for analyzing and designing an optimal spare-management policy in large-scale satellite constellations using a Markov chain model. To capture the stochastic nature of satellite failures and launch vehicle lead times, we adopt Markov chains to model both failure and replenishment processes. We reinvestigate an indirect spare strategy, modeled as a multi-echelon periodic-review reorder-point/order-quantity policy, in which spares are first delivered to parking orbits and then transferred to constellation planes. The stock levels in constellation and parking orbits are each modeled as independent Markov chains, and a fixed-point iteration yields a consistent joint stationary solution that describes the strategy’s average behavior. This approach accurately captures the stochastic interplay within a multi-echelon model driven by orbital mechanics, avoiding the aggregation assumptions of prior works and remaining valid across a wider operating domain. Building on this fast, accurate analysis, we formulate an optimization problem and solve it via a genetic algorithm. Finally, we demonstrate the practical value of both the analysis method and the optimization framework in a real-world mega-constellation case study. 0 0 footnotetext: This paper is a substantially revised version of a portion of Paper AAS 24-165 presented at the AAS/AIAA Astrodynamics Specialist Conference, Broomfield, CO, August 11-15, 2024. Nomenclature \tau_{\text{mc}} = Time step of the discrete-time Markov process, in days \lambda_{\text{sat}} = Failure rate of a satellite, in failures per unit time \mu_{\text{lv}} = Mean interval between launches, in days \tau_{\text{lv}} = Constant launch order processing time, in days \tau_{\text{c}}/\tau_{\text{p}} = RAAN alignment period from the perspective of in-plane/parking orbits q_{\text{c}}/q_{\text{p}} = Replenishment quantity for in-plane/parking spares r_{\text{c}}/r_{\text{p}} = Reorder point for in-plane/parking spares N_{\text{sat}_{\text{c}}}/N_{\text{sat}_{\text{p}}} = Maximum in-plane/parking state level including operational and spare satellites \bar{N}_{\text{sat}} = Nominal number of operational satellites per in-plane orbit N_{\text{orbit}_{\text{c}}}/N_{\text{orbit}_{\text{p}}} = Number of in-plane (constellation)/parking orbital planes P_{f_{\text{c}}}/P_{f_{\text{p}}} = Failure transition matrix for in-plane/parking states per review period P_{q_{\text{c}}}/P_{q_{\text{p}}} = q -unit replenishment transition matrix for in-plane/parking states \pi^{q_{\text{c}}}/\pi^{q_{\text{p}}} = Expected in-plane/parking state distribution immediately after q -unit replenishment \pi^{r_{\text{c}}}/\pi^{r_{\text{p}}} = Expected in-plane/parking state distribution at the r -reorder point \pi^{\text{io}_{\text{c}}}/\pi^{\text{io}_{\text{p}}} = Expected in-plane/parking state distribution during the inter-order (IO) period \pi^{\text{lt}_{\text{c}}}/\pi^{\text{lt}_{\text{p}}} = Expected in-plane/parking state distribution during the lead-time (LT) period \pi^{\text{rc}_{\text{c}}}/\pi^{\text{rc}_{\text{p}}} = Expected in-plane/parking state distribution over the full replenishment cycle (RC) 1 Introduction T he satellite constellation is a coordinated group of satellites that together enable continuous global coverage. Constellations with thousands of satellites are becoming central to space-based communication infrastructure. Starlink has deployed over 9,000 satellites as of 2025 and plans to expand to more than 42,000 [ 1 , 2 ] . Amazon’s Project Kuiper began launching in 2025 and targets 3,236 satellites [ 3 ] . OneWeb, now under Eutelsat, has completed its initial deployment of 648 satellites and is planning expansion [ 4 ] . China’s Guowang network is expected to include about 13,000 satellites and has already begun initial deployment [ 5 ] . These systems rely on advanced constellation architecture, frequency coordination, and inter-satellite communication, with a comparative analysis of their designs provided by Pachler et al. [ 6 ] . Over time, satellite failures reduce operational capacity and degrade constellation performance. To maintain service, an effective population management strategy is required. Two representative approaches are spare deployment and on-orbit servicing. The former inserts a spare satellite to replace a failed one [ 7 ] , while the latter uses a servicer to repair the failed satellite [ 8 ] . Spare deployment is generally more suitable for mega-constellations with small, low-cost satellites, whereas servicing is more appropriate for smaller systems with high-value assets. In the context of low Earth orbit (LEO) mega-constellations, launching replacement satellites is widely regarded as more cost-effective than performing on-orbit repairs. This is driven by several factors: declining launch costs [ 9 ] , economies of scale from mass production, and the high expense of designing satellites to be serviceable. Additionally, small satellites tend to have higher failure rates due to limited testing and lack of redundancy [ 10 ] , further reinforcing the need for rapid spare deployment. These trends motivate the study of optimal spare-management strategies, which is the focus of this paper. Early work on spare deployment modeled satellite replenishment using an (s,S) inventory framework with exponential lead times, establishing a probabilistic basis for maintaining constellation performance [ 11 ] . Subsequent research analyzed various spare deployment strategies—including in-orbit, parking orbit, and ground-based options—across different phases of the constellation life cycle [ 7 ] . Discrete-time Markov decision processes were later applied to determine cost-optimal satellite replacement policies over finite planning horizons [ 12 ] . A more scalable formulation for mega-constellations was introduced through a multi-echelon reorder-point/order-quantity policy, commonly referred to as the (s,Q) inventory model or the (r,q) model, that considers parking orbits as intermediate warehouses between the ground and constellation planes [ 13 ] . Beyond inventory modeling, other studies have explored resilience and value-based assessment. A value model was developed to examine trade-offs between satellite reliability and system attributes such as cost or mass [ 14 ] . A quantitative resilience framework was proposed, incorporating robustness and recovery metrics to support mission-driven constellation optimization [ 15 ] . Resilience under solar weather radiation was also analyzed using a time-based simulation approach that evaluated reconstitution strategies and highlighted the influence of manufacturing and launch timelines on system performance [ 16 ] . Building on previous work [ 17 ] , this paper introduces a novel Markov-based methodology for analyzing spare management strategies. The approach employs Markov processes to represent state transitions and analyze long-term behavior through stationary state distributions. The framework incorporates key physical realities often simplified in other models, such as the non-negativity of physical stock and state-dependent failure rates. The central contribution is the high-fidelity modeling of the multi-echelon replenishment dynamics. Unlike prior works that use aggregation assumptions, our approach models the stochastic interplay between the probability distribution of in-plane demand and that of parking orbit spare availability, a dynamic physically constrained by the periodic lead times from orbital mechanics. By computing the full state probability distribution, a step often intractable in prior analytical models, the framework enables a rigorous analysis of operational costs and constellation availability. We demonstrate the framework on a real-world mega-constellation case study. The remainder of the paper is organized as follows. Section 2 introduces the modeling preliminaries. Section 3 presents the proposed analytical method for evaluating the indirect resupply strategy. Section 4 applies this method to assess system performance. Section 5 validates the model through comparison with Monte Carlo simulations, and Section 6 illustrates its application in a design optimization context. Finally, Section 7 concludes the paper and suggests directions for future research. 2 Preliminaries 2.1 Spare Management Policy 2.1.1 Indirect Resupply Strategies The indirect strategy uses a large launch vehicle (LV) to deliver spares to a parking orbit, from which they are transferred to the in constellation orbit via RAAN drift. Although this method benefits from batch discounts and lower launch cost, the RAAN drift introduces a longer replenishment delay. By contrast, the direct strategy employs