Quantum Modeling of Spatial Contiguity Constraints

Amr Magdy; Federico M. Spedalieri; Yunhan Chang

REVIEW 4 major objections 5 minor 1 cited by

Quantum Modeling of Spatial Contiguity Constraints

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims the first quantum model that enforces spatial contiguity among neighboring spatial entities, using flow variables on adjacency edges in DQM and QUBO forms plus a hybrid seed-based extension for larger maps.

desk verdict The contiguity encoding has a real idea, but the root-assignment constraint is never enforced, so invalid partitions with empty regions have zero penalty—this needs fixing before the paper claims to solve p-regions. read the letter →

arxiv 2505.12608 v2 pith:MXKWSRCY submitted 2025-05-19 cs.DC

Yunhan Chang , Amr Magdy , Federico M. Spedalieri This is my paper · ORCID

classification cs.DC

keywords quantumannealingspatialcontiguityp-regionsproblemnetworkflowmodeldiscretequadraticQUBOhybridquantum-classicaloptimizationregionalization

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper sets out to make spatial contiguity—the requirement that all areas in a region form one connected piece—expressible as an optimization model that quantum annealers can accept. The key move is to encode contiguity through network flow: each region has a preselected source area, every other area in the region receives one unit of flow from that source, and flow is allowed only between neighboring areas. This flow constraint is delivered in two forms, a Discrete Quadratic Model (DQM) for small instances and a Quadratic Unconstrained Binary Optimization (QUBO) formulation for binary hardware, together with a hybrid seeding scheme that grows regions classically and uses quantum optimization to move boundary areas. Because flow variables exist only on adjacency edges, the model avoids the quadratic variable blow-up of pairwise formulations. If the formulation is right, the same flow gadget can enforce contiguity in any spatial optimization problem, not just p-regions.

What carries the argument

The load-bearing object is a directed flow network defined on the adjacency graph of the spatial areas. Each area $i$ is a node labeled by its region $d_i$; each directed edge that corresponds to a shared border carries a flow $f_{ij}\ge 0$, so the number of flow variables equals the number of adjacency edges rather than all pairs of areas. The constraint $u_{ij}=|d_i-d_j|$ switches flow off between different regions, and the inflow-minus-outflow equation, with value $1$ at every non-root and $1-|R_k|$ at the region's root, forces every area to be reachable from the root. This is what turns a global topological property (connectedness) into local linear equations, and the paper's variable-reduction argument is what makes the DQM/QUBO encoding feasible on current hardware.

What would settle it

Take a path of four areas with $p=2$ and fix the roots at areas 1 and 2. The connected partition $\{\{1,2\},\{3,4\}\}$, which places both roots in one region, cannot be represented by the model even if it minimizes the dissimilarity objective; constructing the DQM/QUBO for this instance and comparing its ground-state objective with the unrestricted optimum therefore settles whether the rooted restriction changes the problem.

Watch

Extended reading notes

Core claim

The paper claims to present the first quantum model that ensures spatial contiguity constraints among neighboring spatial entities, using the p-regions problem as its testbed. With assignment variables $d_i\in\{1,\dots,p\}$, the model adds nonnegative flow variables $f_{ij}$ on adjacency edges only, and gates them by $u_{ij}=|d_i-d_j|$ so that flow cannot cross region borders. For each region $k$, a root area $r(k)$ is fixed in advance with $d_{r(k)}=k$; every non-root area in the region must satisfy $\sum_{j\in N(i)}f_{ji}-\sum_{j\in N(i)}f_{ij}=1$, and the root must supply $|R_k|-1$ units. The paper proves (Theorem 5.1) that these conservation equations force each region to be connected, and argues that the resulting DQM is small enough to run directly on quantum annealing hardware for small datasets, while a hybrid seed-growing scheme extends the approach to larger maps by restricting quantum optimization to movable border areas.

Load-bearing premise

The model fixes one source area per region in advance, but the original p-regions problem has no such sources, so the contiguity guarantee applies only to the rooted variant and the result quality depends on those preselected sources being compatible with a good partition.

Editorial extensions

If this is right

Any region-based optimization that needs contiguity—electoral districting, forest planning, site selection, wildlife corridors—can in principle use the same flow gadget, since the paper presents the model as general rather than p-regions-specific.
Variable count grows with the number of shared borders, so maps with sparse spatial adjacency produce smaller DQM/QUBO instances and are better suited to current quantum hardware.
Small p-region instances can be encoded directly as a DQM and submitted to quantum annealing hardware; larger maps can be split by seed selection, classical region growing, and a quantum pass over movable border areas.
When the flow conservation equations are satisfied, each region is provably connected (Theorem 5.1), so contiguity is guaranteed by the model rather than checked post hoc.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper's claims, the rooted formulation means the model actually solves a rooted variant of p-regions: a contiguous partition is representable only when each region contains exactly one preselected root. Extending to the original problem means optimizing root choice jointly, unless one can prove some optimal partition contains the scattered seeds.
Beyond the paper's claims, the assertion that the DQM is solvable directly on quantum annealing hardware is not backed by a hardware run or simulation in the paper, so whether annealers find ground states reliably for these energy landscapes remains an open empirical question, likely sensitive to penalty weights.
Beyond the paper's claims, a natural testable extension is to make the roots themselves discrete decision variables; on sparse maps the added assignment dimension may still fit within current qubit limits and would remove the rooted restriction.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: the paper has a real idea, but the central model as written doesn't enforce what it claims. The root-identity constraint d_{r(k)}=k is stated in passing in Section 5 and then never added to either the DQM or the QUBO objective. The stress-test example is right: on the three-node path with p=2, assign all nodes to region 1 and set f12=2, f23=1; every lambda_1 through lambda_4 penalty is zero, so the model accepts a partition where region 2 is empty. That is a load-bearing gap, not a cosmetic one.

What is genuinely new deserves credit: the DQM/QUBO encoding of Shirabe's flow model with flow variables restricted to adjacency edges, the binary expansion of flow variables, and the seeding-based hybrid decomposition. Those encoding details are not in the cited literature, and restricting flows to neighborhood edges is a sensible way to cut qubit count. The flow conservation theorem is standard and correctly stated for fixed roots. The paper also cites the relevant classical work on Shibabe and Duque, so the lineage is clear.

The soft spots go beyond the root gap. There are no hardware runs or simulations anywhere; Section 8 says only that empirical validation is ongoing. Predetermining roots changes the p-regions problem, which has no roots in its original definition, so the model solves a rooted variant unless seed selection is globally optimal. The set notation I\{r(d_1),...,r(d_p)} in the QUBO is ambiguous, which is another sign that the constraint layer is under-specified. Penalty coefficients are said to be 'suitably high' with no guidance on setting them.

None of this kills the underlying idea—a corrected model needs explicit root-assignment penalties (e.g., add a term forcing x_{r(k),k}=1) and at least a small annealer demonstration. But as written, the mathematical claim that the model ensures contiguity for the p-regions problem is not supported. This reads like a workshop paper, and it is one (Q-Data 2025).

My recommendation: a serious referee should see it, not because it is near publishable, but because the flaw is concrete, fixable, and worth documenting. If the authors add the missing constraints and run a small test, this could be a useful contribution to the quantum-spatial optimization literature. I'd send it to review with a major-revision decision. Your reader's REJECT is fair for the current arXiv version; I'd soften that only slightly because the core idea is salvageable.

Referee Report

4 major / 5 minor

Summary. The paper proposes a DQM and a QUBO formulation of the p-regions problem with an explicit spatial-contiguity constraint, using a network-flow model with one predetermined root (source) per region. It claims to be the first quantum model that ensures spatial contiguity among neighboring spatial entities, and it sketches a hybrid quantum-classical seeding approach for larger instances. The central technical content is Theorem 5.1, which states that if flow conservation holds in each region, then the region is connected, and the subsequent DQM/QUBO encodings of the flow constraints, assignment variables, and penalties.

Significance. If the model were correct, the paper would fill a real gap: no previous work encodes spatial contiguity for regionalization in a form directly mappable to D-Wave quantum annealers. The flow-based encoding restricted to adjacency edges is a sensible variable-reduction idea, and the paper correctly identifies Shirabe's flow model as the appropriate starting point. However, the current formulation does not enforce the root-assignment condition, and the QUBO penalty set is not well defined because it depends on the decision variables. These are load-bearing issues: the model as written admits invalid p-region partitions with empty regions, so the central claim of ensuring spatial contiguity for the p-regions problem is not supported. The paper also provides no experimental validation or resource estimates despite asserting direct hardware solvability in the abstract.

major comments (4)

[Section 5, 'Complete DQM Formulation' and Section 6, 'Complete QUBO Formulation']
[Section 6, 'Complete QUBO Formulation']
[Section 5, 'Intra-Region Flow Constraint' and 'Complete DQM Formulation']
[Abstract and Section 8]

minor comments (5)

[Section 5, 'Complete DQM Formulation']
[Throughout]
[Section 3, Definition 3.1]
[Section 4, Figure 2]
[Section 7]

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the flow-based contiguity constraints are adopted from external prior work and re-encoded into DQM/QUBO, with no fitted parameter presented as a prediction.

full rationale

The paper's central derivation re-expresses the established Shirabe/Duque flow-conservation model as a DQM and then as a QUBO. The contiguity condition is not derived from the objective or from any fitted quantity; it is encoded directly as flow-conservation constraints with predetermined roots, and the objective is the standard p-regions heterogeneity function. The proof of Theorem 5.1 relies on summing flow balances over a disconnected component, which is the usual argument for this known flow formulation. No parameter is fitted to data and then called a prediction, and no uniqueness claim is imported from the authors' own prior work. The paper does cite itself as reference [8] for supplementary details, but that self-citation is not load-bearing for the contiguity derivation. The root-preselection assumption ('a unique root area r(j) is predetermined (so that d_{r(j)}=j)') and the observation that this identity is not explicitly enforced in the QUBO are correctness/validity concerns about whether every feasible solution corresponds to a valid p-region partition, not circular reasoning: the model's output is not defined in terms of its own conclusion, and the derivation does not reduce to its inputs. The absence of hardware experiments is an empirical-support gap, not circularity. The derivation is self-contained relative to the cited classical flow model and standard quantum encoding schemes.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical or mathematical entities are postulated; the flow variables, roots, and penalties are standard modeling devices. The free parameters and domain assumptions above capture the unverified choices the central claim depends on.

free parameters (3)

M (flow upper bound)
Used in the intra-region constraint f_ij <= M(1-u_ij) and in choosing binary expansion length L; the paper only says 'suitably large constant,' so the model's behavior depends on an unspecified hand-chosen value.
Penalty coefficients lambda_1..lambda_4
The QUBO objective adds four penalties with weights 'chosen sufficiently high'; no values or scaling rules are given, and feasibility depends on these choices.
L (binary expansion width for flow variables)
Number of binary variables per flow in QUBO, chosen such that 2^L - 1 >= M; depends on M and is not specified.

assumptions (4)

standard math Flow conservation with a single source per region implies the region is connected (Theorem 5.1).
This is the classical Shirabe flow argument; the proof is sketched and essentially correct, but assumes flows are bounded and internal edge flows cancel when summed over a component.
domain assumption Spatial adjacency can be represented as a graph whose edges are shared boundaries, and flow on neighbor edges suffices to characterize contiguity.
Section 4; this is standard in regionalization but assumes areas are connected polygons with shared boundaries and ignores point-adjacency artifacts.
ad hoc to paper A predetermined root per region can be fixed without loss of generality for the p-regions problem.
Section 5 fixes r(k) and scattered seeding; the original p-regions definition has no roots, so this restricts the feasible set and needs justification or optimization over seeds.
domain assumption D-Wave DQM and QUBO solvers can represent the variables and enforce the constraints through penalties in practice.
The paper relies on D-Wave's DQM solver and hybrid solvers (refs 10, 18, 33) without demonstrating that the required domains, M values, and penalty strengths are within hardware limits.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Quantum Modeling of Spatial Contiguity Constraints." pith.science (2026). https://pith.science/paper/MXKWSRCY

@misc{pith2026250512608,
  author       = {Pith},
  title        = {Pith review of: Quantum Modeling of Spatial Contiguity Constraints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MXKWSRCY}},
  note         = {Machine review of arXiv:2505.12608}
}

read the original abstract

Quantum computing has demonstrated potential for solving complex optimization problems; however, its application to spatial regionalization remains underexplored. Spatial contiguity, a fundamental constraint requiring spatial entities to form connected components, significantly increases the complexity of regionalization problems, which are typically challenging for quantum modeling. This paper proposes novel quantum formulations based on a flow model that enforces spatial contiguity constraints. Our scale-aware approach employs a Discrete Quadratic Model (DQM), solvable directly on quantum annealing hardware for small-scale datasets. In addition, it designs a hybrid quantum-classical approach to manage larger-scale problems within existing hardware limitations. This work establishes a foundational framework for integrating quantum methods into practical spatial optimization tasks.

Figures

Figures reproduced from arXiv: 2505.12608 by the authors.

**Figure 2.** Example of Connected and Disconnected Regions [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Spatial Regionalization: A Hybrid Quantum Computing Approach
cs.ET 2025-06 conditional novelty 6.0 of 10

A hybrid quantum-classical pipeline applies D-Wave annealing to seed selection and local optimization in spatial regionalization, reporting preliminary quality and runtime improvements at small scales.

Reference graph

Works this paper leans on

45 extracted references · 19 canonical work pages · cited by 1 Pith paper

[1]

Hussah Alrashid, Yongyi Liu, and Amr Magdy. 2022. SMP: scalable max-P regionalization. InProceedings of the 30th International Conference on Advances in Geographic Information Systems(SIGSPATIAL)(Seattle, Washington). 4 pages. doi:10.1145/3557915.3561011

arXiv 2022
[2]

Hussah Alrashid, Yongyi Liu, and Amr Magdy. 2023. PAGE: Parallel Scalable Regionalization Framework.ACM Transactions on Spatial Algorithms and Systems 9, 3 (2023), 21:1–21:26

work page 2023
[3]

Janikas, Renato Assunção, and Ting-Hwan Lee

Orhun Aydin, Mark V. Janikas, Renato Assunção, and Ting-Hwan Lee. 2018. SKATER-CON: Unsupervised Regionalization via Stochastic Tree Partitioning within a Consensus Framework Using Random Spanning Trees: Research Paper. InProceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. 33–42. doi:10.1145/3281548.3281554

work page arXiv 2018
[4]

Orhun Aydin, Mark. V. Janikas, Renato Martins Assunção, and Ting-Hwan Lee and. 2021. A quantitative comparison of regionalization methods.International Journal of Geographical Information Science35, 11 (2021), 2287–2315. doi:10.1080/ 13658816.2021.1905819

arXiv 2021
[5]

Christian Bauckhage, Nico Piatkowski, Rafet Sifa, Dirk Hecker, and Stefan Wrobel

work page
[6]

Adam Callison and Nicholas Chancellor. 2022. Hybrid quantum-classical algo- rithms in the noisy intermediate-scale quantum era and beyond.Phys. Rev. A 106 (Jul 2022), 010101. Issue 1. doi:10.1103/PhysRevA.106.010101

doi:10.1103/physreva.106.010101 2022
[7]

Rodolfo Carvajal, Miguel Constantino, Marcos Goycoolea, Juan Vielma, and Andres Weintraub. 2013. Imposing Connectivity Constraints in Forest Planning Models.Operations Research61 (08 2013), 824–836. doi:10.2307/23481799

work page doi:10.2307/23481799 2013
[8]

Spedalieri

Yunhan Chang, Amr Magdy, and Federico M. Spedalieri. 2025. Quantum Modeling of Spatial Contiguity Constraints. arXiv:2505.12608 [cs.DC] https: //arxiv.org/abs/2505.12608

arXiv 2025

Show all 45 references

[9]

Church, J

R. Church, J. C. Duque, and D. E. Restrepo. 2020. The p-Innovation Ecosystems Model. arXiv:2008.05885 [physics.soc-ph] https://arxiv.org/abs/2008.05885

2020 arXiv
[10]

2022.A Hybrid Solver for Constrained Quadratic Models

D-Wave Systems Inc. 2022.A Hybrid Solver for Constrained Quadratic Models. Technical Report. D-Wave Systems Inc. https://www.dwavequantum.com/media/ rldh2ghw/14-1055a-a_hybrid_solver_for_constrained_quadratic_models.pdf

2022
[11]

Chakrabarti

Arnab Das and Bikas K. Chakrabarti. 2008. Colloquium: Quantum annealing and analog quantum computation.Rev. Mod. Phys.80 (Sep 2008), 1061–1081. Issue 3. doi:10.1103/RevModPhys.80.1061

2008 doi
[12]

Juan Carlos Duque, Luc Anselin, and Sergio J. Rey. 2012. The max-p-regions problem.Journal of Regional Science52, 3 (2012), 397–419. doi:10.1111/j.1467- 9787.2011.00743.x

2012
[14]

A. B. Finnila, M. A. Gómez, C. Sebenik, C. Stenson, and J. D. Doll. 1994. Quantum annealing: A new method for minimizing multidimensional functions.Chemical Physics Letters219, 5-6 (1994), 343–348. doi:10.1016/0009-2614(94)00117-0

1994 doi
[15]

Gemeinhardt, Robert Wille, and Manuel Wimmer

Felix G. Gemeinhardt, Robert Wille, and Manuel Wimmer. 2021. Quantum k- community detection: algorithm proposals and cross-architectural evaluation. Quantum Information Processing20, 9 (2021), 302. doi:10.1007/s11128-021-03239- 1

2021 doi
[16]

Fred Glover, Gary Kochenberger, Rick Hennig, and Yu Du. 2022. Quantum bridge analytics I: a tutorial on formulating and using QUBO models.Annals of Operations Research314, 1 (2022), 141–183. doi:10.1007/s10479-022-04634-2

2022 doi
[17]

D. Guo. 2008. Regionalization with dynamically constrained agglomerative clustering and partitioning (REDCAP).International Journal of Geographical Information Science22, 7 (2008), 801–823. doi:10.1080/13658810701674970

2008 doi
[18]

D-Wave Systems Inc. [n. d.]. Discrete Quadratic Models (DQM) Documentation. https://docs.ocean.dwavesys.com/en/stable/docs_dqm/dqm.html. Accessed: 2025-02-18

2025
[19]

Quantum Insider. 2024. 8 Companies That Work with Quan- tum Annealing Technology: A Powerful Approach to Optimization. https://thequantuminsider.com/2024/06/10/8-companies-that-work-with- quantum-annealing-technology-a-powerful-approach-to-optimization/

2024
[20]

Tadashi Kadowaki and Hidetoshi Nishimori. 1998. Quantum annealing in the transverse Ising model.Physical Review E58, 5 (1998), 5355–5363. doi:10.1103/ PhysRevE.58.5355

1998
[21]

Yunfan Kang, Yongyi Liu, Hussah Alrashid, Akash Bilgi, Siddhant Purohit, Ahmed Mahmood, Sergio Rey, and Amr Magdy. 2024. Pyneapple-R: Scalable and Ex- pressive Spatial Regionalization . In2024 IEEE 40th International Conference on Data Engineering (ICDE). 5497–5500. doi:10.110...

2024
[22]

Yunfan Kang and Amr Magdy. 2022. EMP: Max-P Regionalization with Enriched Constraints. In2022 IEEE 38th International Conference on Data Engineering (ICDE). 1914–1926. doi:10.1109/ICDE53745.2022.00189

2022
[23]

Krishnageetha Karuppasamy, Varun Puram, Stevens Johnson, and Johnson P. Thomas. 2025. A Comprehensive Review of Quantum Circuit Optimization: Current Trends and Future Directions.Quantum Reports7, 1 (2025). doi:10.3390/ quantum7010002

2025
[24]

King, Jack Raymond, Trevor Lanting, Rami Harris, Alessandro Zucca, Francesco Altomare, et al

Andrew D. King, Jack Raymond, Trevor Lanting, Rami Harris, Alessandro Zucca, Francesco Altomare, et al. 2023. Quantum critical dynamics in a 5,000-qubit programmable spin glass.Nature617, 7959 (2023), 61–66. doi:10.1038/s41586- 023-05867-2

2023 doi
[25]

Kirkpatrick, C

S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi. 1983. Optimization by Simulated Annealing.Science220, 4598 (1983), 671–680. doi:10.1126/science.220.4598.671

1983 doi
[26]

Yaroslav Koshka and M. A. Novotny. 2020. Comparison of D-Wave Quantum Annealing and Classical Simulated Annealing for Local Minima Determination. InProceedings of the 2020 IEEE International Conference on Rebooting Computing (ICRC). 1–8. doi:10.1109/ICRC49039.2020.00012

2020
[27]

Vaibhaw Kumar, Gideon Bass, Casey Tomlin, and Joseph Dulny. 2018. Quantum annealing for combinatorial clustering.Quantum Information Processing17, 2 (Feb. 2018), 1–14. doi:10.1007/s11128-017-1809-2

2018 doi
[28]

Mahmood, Amr Magdy, and Sergio Rey

Yongyi Liu, Ahmed R. Mahmood, Amr Magdy, and Sergio Rey. 2021. PRUC: P-regions with user-defined constraint.Proc. VLDB Endow.15, 3 (2021). doi:10. 14778/3494124.3494133

2021
[29]

Assuncao Martins, Marcos Neves, Gilberto Câmara, and Corina Freitas. 2006. Efficient Regionalization Techniques for Socio-Economic Geographical Units Us- ing Minimum Spanning Trees.International Journal of Geographical Information Science20 (08 2006), 797–811. doi:10.1080/1365...

2006 doi
[30]

Samya Muhuri and Shashank Sheshar Singh. 2024. Quantum-Social Network Analysis for Community Detection: A Comprehensive Review.IEEE Transactions on Computational Social Systems11, 5 (2024), 6795–6806. doi:10.1109/TCSS.2024. 3397967

2024 doi
[31]

Alan Murray, Jiwon Baik, Vanessa Figueroa, Darlene Rini, Max Moritz, Dar Roberts, Stuart Sweeney, Leila Carvalho, and Charles Jones. 2023. Developing effective wildfire risk mitigation plans for the wildland urban interface.Inter- national Journal of Applied Earth Observation ...

2023
[32]

Alan Murray and Richard Church. 2023. Spatial Optimization of Multiple Area Land Acquisition.Computers & Operations Research153 (05 2023), 106160. doi:10. 1016/j.cor.2023.106160

2023
[33]

Eneko Osaba and Pablo Miranda-Rodriguez. 2024. D-Wave’s Nonlinear-Program Hybrid Solver: Description and Performance Analysis.arXiv preprint(2024). arXiv:2410.07980 [quant-ph] https://arxiv.org/abs/2410.07980

2024 arXiv
[34]

Alasdair Rae. 2011. The Geography of Travel to Work in England and Wales: Ex- tracting Functional Regions from a Network of Commuting Flows.Geographical Analysis43, 1 (2011), 49–67. doi:10.1111/j.1538-4632.2010.00810.x

2011
[35]

G. E. Santoro, R. Martonak, E. Tosatti, and R. Car. 2006. Optimization using quantum mechanics: Quantum annealing through adiabatic evolution.Journal of Physics A: Mathematical and General39 (2006), R393–R431. doi:10.1088/0305- 4470/39/36/R01

2006 doi
[36]

Philipp Schworm, Xiangqian Wu, Moritz Glatt, and Jan C. Aurich. 2023. Solving flexible job shop scheduling problems in manufacturing with Quantum Anneal- ing.Production Engineering17, 1 (2023), 105–115. doi:10.1007/s11740-022-01145-8

2023 doi
[37]

Takeshi Shirabe. 2005. A Model of Contiguity for Spatial Unit Allocation.Geo- graphical Analysis37, 1 (2005), 2–16. doi:10.1111/j.1538-4632.2005.00605.x

2005
[38]

Rachel St John, Sándor Tóth, and Zelda Zabinsky. 2018. Optimizing the Geometry of Wildlife Corridors in Conservation Reserve Design.Operations Research66 (09 2018). doi:10.1287/opre.2018.1758

2018
[39]

Robert E. Tarjan. 1972. Depth-first search and linear graph algorithms.SIAM J. Comput.1, 2 (1972), 146–160. doi:10.1137/0201010

1972 doi
[40]

Hamidreza Validi, Austin Buchanan, and Eugene Lykhovyd. 2022. Imposing Contiguity Constraints in Political Districting Models.Oper. Res.70, 2 (March 2022), 867–892. doi:10.1287/opre.2021.2141

2022
[41]

Alonso Viladomat Jasso, Ark Modi, Roberto Ferrara, Christian Deppe, Janis Nötzel, Fred Fung, and Maximilian Schädler. 2023. Quantum and Quantum- Inspired Stereographic K Nearest-Neighbour Clustering.Entropy25, 9 (Sept. 2023), 1361. doi:10.3390/e25091361

2023 doi
[42]

Ran Wei, Sergio Rey, and Elijah Knaap. 2020. Efficient regionalization for spa- tially explicit neighborhood delineation.International Journal of Geographical Information Science35 (05 2020), 1–17. doi:10.1080/13658816.2020.1759806

2020
[43]

Marcin Wierzbiński, Joan Falcó-Roget, and Alessandro Crimi. 2023. Community detection in brain connectomes with hybrid quantum computing.Scientific Reports13, 1 (2023), 3446. doi:10.1038/s41598-023-30579-y

2023 doi
[44]

Kein Yukiyoshi, Taku Mikuriya, Hyeon Seok Rou, Giuseppe Thadeu Freitas de Abreu, and Naoki Ishikawa. 2024. Quantum Speedup of the Dispersion and Codebook Design Problems .IEEE Transactions on Quantum Engineering5, 01 (Jan. 2024), 1–16. doi:10.1109/TQE.2024.3450852

2024
[45]

Kamila Zaman, Alberto Marchisio, Muhammad Abdullah Hanif, and Muhammad Shafique. 2023. A Survey on Quantum Machine Learning: Current Trends, Chal- lenges, Opportunities, and the Road Ahead. (10 2023). arXiv:2310.10315 [quant- ph] This paper has been revised and accepted at the...

2023 arXiv
[2019]

A QUBO Formulation of the k-Medoids Problem. InLWDA. 54–63. https: //publica.fraunhofer.de/handle/publica/405469 This paper has been revised and accepted at the Q-Data 2025 Workshop, in conjunction with ACM SIGMOD 2025. Quantum Modeling of Spatial Contiguity Constraints

2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Hussah Alrashid, Yongyi Liu, and Amr Magdy. 2022. SMP: scalable max-P regionalization. InProceedings of the 30th International Conference on Advances in Geographic Information Systems(SIGSPATIAL)(Seattle, Washington). 4 pages. doi:10.1145/3557915.3561011

arXiv 2022

[2] [2]

Hussah Alrashid, Yongyi Liu, and Amr Magdy. 2023. PAGE: Parallel Scalable Regionalization Framework.ACM Transactions on Spatial Algorithms and Systems 9, 3 (2023), 21:1–21:26

work page 2023

[3] [3]

Janikas, Renato Assunção, and Ting-Hwan Lee

Orhun Aydin, Mark V. Janikas, Renato Assunção, and Ting-Hwan Lee. 2018. SKATER-CON: Unsupervised Regionalization via Stochastic Tree Partitioning within a Consensus Framework Using Random Spanning Trees: Research Paper. InProceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. 33–42. doi:10.1145/3281548.3281554

work page arXiv 2018

[4] [4]

Orhun Aydin, Mark. V. Janikas, Renato Martins Assunção, and Ting-Hwan Lee and. 2021. A quantitative comparison of regionalization methods.International Journal of Geographical Information Science35, 11 (2021), 2287–2315. doi:10.1080/ 13658816.2021.1905819

arXiv 2021

[5] [5]

Christian Bauckhage, Nico Piatkowski, Rafet Sifa, Dirk Hecker, and Stefan Wrobel

work page

[6] [6]

Adam Callison and Nicholas Chancellor. 2022. Hybrid quantum-classical algo- rithms in the noisy intermediate-scale quantum era and beyond.Phys. Rev. A 106 (Jul 2022), 010101. Issue 1. doi:10.1103/PhysRevA.106.010101

doi:10.1103/physreva.106.010101 2022

[7] [7]

Rodolfo Carvajal, Miguel Constantino, Marcos Goycoolea, Juan Vielma, and Andres Weintraub. 2013. Imposing Connectivity Constraints in Forest Planning Models.Operations Research61 (08 2013), 824–836. doi:10.2307/23481799

work page doi:10.2307/23481799 2013

[8] [8]

Spedalieri

Yunhan Chang, Amr Magdy, and Federico M. Spedalieri. 2025. Quantum Modeling of Spatial Contiguity Constraints. arXiv:2505.12608 [cs.DC] https: //arxiv.org/abs/2505.12608

arXiv 2025