Article 3/3
After the promises, a return to reality is required.
Is quantum computing already usable in geomatics? What experiments actually exist? And on what timeline could concrete contributions to GIS realistically be expected?This final article provides a critical overview. It reviews existing research and demonstrators, analyzes current technological and methodological limitations, and offers a realistic reading of short-, medium-, and long-term perspectives. An essential conclusion for approaching quantum computing in GIS with curiosity, but also with rigor and pragmatism.
Concrete Examples and Existing Experiments
A field that remains largely experimental
At this stage, there are no operational GIS tools directly based on quantum computing. Current work mainly consists of academic research, industrial prototypes, or proof-of-concept studies conducted on simplified problems.
The goal of these experiments is not to replace existing GIS processing chains, but to test whether certain classes of geographic problems could benefit — even marginally — from quantum or hybrid approaches.
Spatial optimization and quantum annealing
The most advanced experiments concern combinatorial optimization, particularly through quantum annealing. Several studies have explored:
- optimal facility location on spatial grids,
- allocation of zones to constrained land uses,
- simplified territorial coverage problems.
In these cases, geographic data are heavily aggregated or discretized to make the problems compatible with the current capabilities of quantum machines. Results show that, for certain formulations, quantum annealing can quickly provide solutions comparable to those obtained with classical heuristics, without demonstrating a systematic advantage.
Networks and graphs: early experiments
Exploratory work has also focused on graphs inspired by geographic networks: small transportation networks, synthetic graphs, or sub-networks extracted from real-world data.
The objectives are often methodological:
- testing the translation of a spatial graph into a quantum formulation,
- comparing performance with classical algorithms,
- assessing the quality of the solutions obtained.
These studies show that problem formulation is at least as important as computational power itself. Poor encoding leads to weak results, even when using quantum hardware.
Remote sensing and quantum learning: demonstrators
In remote sensing, experiments are even more preliminary. They mainly involve:
- very small datasets,
- simple classification tasks,
- conceptual comparisons between classical and quantum models.
These works belong more to machine learning research than to operational geomatics. Nevertheless, they make it possible to explore how spectral signatures or spatial patterns can be represented in quantum spaces.
Platforms and access to quantum technologies
Several actors now provide cloud-based testing environments, allowing researchers and engineers to conduct experiments without access to on-site quantum hardware. Notably, platforms such as IBM Quantum and D-Wave Systems are widely used.
For the GIS community, these platforms mainly offer:
- a framework for technology monitoring,
- opportunities for interdisciplinary research,
- a pedagogical experimentation environment.
Results that must be interpreted with caution
It is important to emphasize that, in most cases:
- the problems tested are highly simplified,
- data sizes are not comparable to real-world GIS datasets,
- observed gains are not always significant compared to classical methods.
This does not diminish the value of these works, but it calls for caution against excessive extrapolation. Current experiments primarily serve to understand the conditions under which quantum computing might become relevant for geomatics.
A primarily methodological contribution at this stage
For GIS, the main contribution of these experiments may be more conceptual than technological. They force researchers and practitioners to:
- clearly reformulate spatial problems,
- explicitly define constraints and objectives,
- distinguish what belongs to intensive computation from what belongs to modeling.
In this sense, quantum computing already acts as a complexity revealer, even before any operational adoption.
Current Limitations and Points of Caution
Technologies that are still immature
Despite rapid research progress, quantum computing remains at an experimental stage. Current machines have a limited number of usable qubits and are highly sensitive to noise, computational errors, and environmental conditions.
Under these circumstances, it is unrealistic to envision complete GIS workflows running on quantum computers in the short term. Realistic use cases are limited to very targeted sub-problems on highly simplified data.
The challenge of translating GIS problems
One of the main obstacles is not technological, but conceptual.
Geographic problems must be reformulated into mathematical models compatible with quantum computing (cost functions, constraints, binary graphs). This step is often more difficult than the computation itself.
Many GIS problems are:
- continuous rather than discrete,
- multi-scale,
- context-dependent,
- highly interpretative.
The loss of geographic richness during this translation represents a real risk, especially when complex spatial models are forced into overly simplistic formulations.
Cost, accessibility, and technological dependency
Access to quantum resources is currently almost exclusively cloud-based and operated by major industrial actors. This raises several issues:
- access and scaling costs,
- dependence on proprietary solutions,
- limited control over infrastructure,
- digital sovereignty concerns for territorial data.
For public-sector actors and local authorities, these dimensions cannot be ignored and must be considered from the earliest exploratory phases.
Gains that are not yet clearly demonstrated
Contrary to some media narratives, quantum computing does not yet demonstrate clear and systematic superiority over classical approaches for realistic GIS problems. In many cases, well-optimized classical algorithms running on CPUs, GPUs, or clusters remain faster, more reliable, and easier to deploy.
The risk, therefore, is to invest significant resources for marginal — or even nonexistent — gains if use cases are not carefully selected.
Rare skills and demanding interdisciplinarity
Applying quantum computing in geomatics requires skills at the intersection of several domains:
- GIS and spatial modeling,
- algorithms and optimization,
- basic notions of quantum physics and computing.
Such profiles are still rare, and capacity building represents a significant investment. Without close collaboration between geomaticians, computer scientists, and researchers, projects risk remaining theoretical or disconnected from operational needs.
The risk of technological hype
Finally, like any emerging technology, quantum computing is exposed to hype effects. Marketing discourse can suggest imminent breakthroughs, while concrete uses remain under construction.
In the GIS domain, vigilance is essential:
it is crucial to distinguish genuine scientific advances from speculative announcements, and to favor approaches grounded in experimentation, transparency, and critical evaluation.
What timeline for GIS?
Short term: technology monitoring and acculturation
In the short term, quantum computing is not an operational tool for GIS. Its value lies primarily in technology monitoring, understanding key concepts, and identifying families of potentially relevant problems.
For geomatics professionals, this implies:
- becoming familiar with general principles of quantum computing,
- following applied research work,
- experimenting with simplified or pedagogical cases,
- integrating these reflections into a broader culture of advanced computing.
At this stage, the challenge is not performance, but gradual capacity building and the ability to engage with other disciplines.
Medium term: hybrid classical–quantum approaches
In the medium term, the most realistic perspectives involve hybrid architectures combining classical computing with specialized quantum modules. In this scenario, GIS remains the core system:
- spatial data management,
- data preparation and aggregation,
- visualization and interpretation of results.
Quantum computing would intervene in a targeted manner, for example to:
- solve a particularly complex optimization sub-problem,
- rapidly explore a large solution space,
- enrich probabilistic or multi-scenario models.
Such hybrid approaches align well with GIS needs, which often prioritize robust and interpretable solutions over theoretically optimal but opaque results.
Long term: new modes of spatial reasoning
In the longer term, if quantum technologies mature, they could more deeply influence how spatial problems are formulated. The goal would no longer be merely to accelerate existing computations, but to rethink certain models.
One can imagine:
- more global approaches to territorial optimization,
- better integration of uncertainty and multiple scenarios,
- decision-support tools capable of simultaneously exploring many spatial hypotheses.
These evolutions would require a shift in perspective: from searching for a single solution to the structured exploration of a space of possibilities.
What skills for tomorrow’s geomaticians?
The emergence of quantum computing also raises questions about skills. Without becoming quantum physicists, geomaticians may increasingly need to:
- formalize spatial problems more rigorously,
- reason in terms of global optimization,
- interact with advanced computing specialists,
- integrate external tools into complex GIS workflows.
These skills are already partly required today with high-performance computing, AI, and advanced modeling. Quantum computing therefore represents a continuity rather than a rupture.
An opportunity to be approached pragmatically
The quantum horizon for GIS is neither imminent nor guaranteed. It will depend on technological progress, the relevance of identified use cases, and the ability of geomatics actors to appropriate these tools critically.
Rather than waiting for a revolution, it is more relevant to view quantum computing as an opportunity for reflection on the current limits of GIS and on new ways of addressing increasingly complex spatial problems.
Conclusion
Quantum computing generates many expectations, sometimes excessive, but it also raises fundamental questions that strongly resonate with current challenges in geographic information systems. Faced with growing complexity in data, networks, and spatial models, it invites a reconsideration of certain structural limits of classical computing.
For GIS, the issue is neither an imminent revolution nor a replacement of existing tools. Quantum applications remain experimental, limited to highly simplified problems and research demonstrators. Classical algorithms, high-performance computing, and GPUs remain far more efficient and mature for most operational uses.
However, ignoring quantum computing would be a mistake. Exploratory work shows a genuine conceptual affinity between certain geographic problems — optimization, graphs, multi-criteria analysis, scenarios — and quantum paradigms. In the medium term, hybrid approaches could enrich GIS processing chains by providing new ways to explore complex solution spaces rather than producing single, definitive answers.
Beyond the technology itself, quantum computing already acts as a methodological catalyst. It forces more rigorous problem formulation, explicit constraints, critical examination of trade-offs, and acceptance of uncertainty as a central component of territorial decision support.
For geomaticians, researchers, educators, and practitioners, the appropriate stance is likely one of curiosity, critical thinking, and pragmatism. Quantum computing is not a miracle solution for GIS, but it represents a stimulating field of exploration at the intersection of geography, computing, and modeling — deserving observation, experimentation, and discussion, free from hype but rich in intellectual ambition.
