New optimization framework streamlines mega‑constellation deployment

Large Low Earth Orbit (LEO) constellations can provide round-the-clock, high-performance information services, but under constraints such as launch cadence, payload capacity, and orbital mechanics, completing efficient, low-cost batch deployments remains a major challenge. In a recent study published in the Chinese Journal of Aeronautics, Junru Lin, Tiantian Zhang, and Min Hu of Space Engineering University (Beijing) propose a scalable optimization framework that preserves temporal fidelity while dramatically shrinking problem size, offering engineering-ready decision support for mega-constellations.

Their work was recently published in the Chinese Journal of Aeronautics (https://doi.org/10.1016/j.cja.2025.103813).

The team builds a “partial time-expanded network” that prunes over 90% of redundant links through feasibility pruning and hierarchical aggregation, and generates time nodes dynamically according to task durations—avoiding the exponential blow-up of traditional “full time-expanded networks” tied to global time discretization. The model separates a “transport layer” (active transitions such as launch and orbital maneuvers) from a “waiting layer” (state holding such as parking-orbit residence) and integrates a dual-channel strategy—direct injection plus parking-orbit indirect injection—to jointly optimize multi-configuration, multi-batch missions using primary and auxiliary launch vehicles.

For solution methodology, the study adopts a hybrid “column generation + heuristic A* pricing” approach with a dual-variable-based subproblem filtering mechanism, which prioritizes capacity-constrained windows and critical arcs to accelerate convergence. Compared with full time-expanded networks, the network size is reduced by about 84.7%, 99.4%, and 88.9% for the representative Telesat, OneWeb, and Starlink constellations, respectively. In computational performance, the subproblem filter reduces pricing problems by up to ~20% (OneWeb) and ~12% (Starlink), with overall solve-time reductions of roughly 30% across scales; A* consistently outperforms Dijkstra in both pricing counts and runtime.

The framework also supports multi-objective trade-offs: by weighting cost against time-to-service, it automatically yields optimal combinations of launcher selection, batching, and injection paths. Simulations show that a “dual-launcher, dual-path” strategy can shorten deployment time by about 9.34% versus a one-to-one scheme, at the expense of an approximately 8.19% increase in total cost—reflecting the balance between the higher throughput/lower specific cost of primary launchers and the responsiveness of auxiliary launchers.

According to the authors, this method provides a unified paradigm for end-to-end constellation deployment planning: it jointly optimizes when to launch, which vehicle to use, which orbit to inject into, and when to transfer and phase, all while respecting launch windows, payload capacities, and transfer feasibility. Looking ahead, the team will address uncertainty and real-time replanning by incorporating production/manifest disruptions and weather, and will focus on modeling practical multi-plane dispersion tactics (e.g., distributing to multiple orbital planes from a single launch via J2 effects) to further enhance the framework’s generality and real-world utility.

Original Source

Junru LIN, Tiantian ZHANG, Min HU. Optimization strategy for batch launch deployment of large-scale low earth orbit constellations based on multimodal transportation network model [J]. Chinese Journal of Aeronautics, 2025, https://doi.org/10.1016/j.cja.2025.103813.

About Chinese Journal of Aeronautics 

Chinese Journal of Aeronautics (CJA) is an open access, peer-reviewed international journal covering all aspects of aerospace engineering, monthly published by Elsevier. The Journal reports the scientific and technological achievements and frontiers in aeronautic engineering and astronautic engineering, in both theory and practice. CJA is indexed in SCI (IF = 5.7, Q1), EI, IAA, AJ, CSA, Scopus.