Risk Academy, led by the Nexus Consortium · Systems and Nexus Studies · Department of Systemic Risk and Complexity
Networks, agent-based and coupled models, digital twins and their limits
જટિલ પ્રણાલીઓ અને પ્રણાલી મોડેલિંગ
Complex systems and systems modelling is the shared methodological core of nexus and systemic-risk work.
Map complex systems and systems modelling on a real placeOpen RAS-401
દૃષ્ટાંતરૂપ: લોરેન્ઝ આકર્ષક પર ટ્રેસર, એક પ્રમાણભૂત જોડાયેલી બિનરેખીય પ્રણાલી
01વ્યાપ
What the discipline covers
The subtopics its courses and labs teach.
- વિદ્યાશાખા
- પ્રણાલી અને Nexus અભ્યાસ ફેકલ્ટી
- અભ્યાસક્રમો
- 1 course open
- સ્તરો
- L2 Intermediate
- પ્રયોગશાળાઓ
- Network lab, Map lab
- Complexity concepts
- Network analysis
- Agent-based models
- Coupled systems models
- Digital twins and their limits
- Model intercomparison
- Reproducible modelling
- Emergence and feedback
Courses in development extend the scope to further subjects: Systems thinking foundations; Agent-based and coupled systems modelling; System dynamics and participatory modelling; Digital twins for risk.
02નકશા
Build a twin of any place
Search any place or open the world; the layers and generators follow complex systems and systems modelling on the planet map, the globe and in 3D.
- Terrain 0 to 1,700 m
- River 2,956 km
- 14 dams
- 6 power plants, 88 MW
- Tree cover 37.8 percent of the box
- Cropland 39.9 percent of the box
- Built-up 0.9 percent of the box
- Permanent water 4.7 percent of the box
- River discharge, GloFAS via Open-Meteo: read live when the 3D twin opens
- Vegetation index, MODIS Terra NDVI 8-day: read live when the 3D twin opens
- Surface water extent, OPERA DSWx (HLS): the flood pulse: read live when the 3D twin opens
- Land cover classes, MODIS IGBP (annual): read live when the 3D twin opens
Drag to move, right drag or Shift with a drag to turn and tilt, Ctrl with the wheel to zoom. Keyboard: arrows move, Shift with arrows turn and tilt, + and – zoom, 0 resets, V steps through the views, Escape clears.
Flat plan of the same data: the lake and rivers, the dams, the plants by capacity, the cropland and tree cover cells and the numbers below. The 3D twin opens on devices that can draw it, once its data and code have been checked; discharge, vegetation, surface water and the annual land cover load in the twin on request, each with its data date.
The boundaries, names and designations shown do not imply any judgement on the legal status of any territory, or any endorsement or acceptance of such boundaries. Countries: Natural Earth (public domain).
03The subject
The subject, methods and measures and debates and bodies
The discipline
The field and its scope
For modellers, analysts and reviewers who build or judge models of coupled systems and need to state what a model assumes, what it can show and how its uncertainty is reported.
Complex systems and systems modelling is the shared methodological core of nexus and systemic-risk work. It studies systems whose behaviour emerges from the interaction of many parts: feedback loops, delays, heterogeneity, adaptation and network structure. Its tools include system dynamics, network analysis, agent-based and coupled models, exploratory modelling under deep uncertainty, model intercomparison and the reproducibility practices that let others rerun and challenge a result. Its users are the analysts who build models of basins, grids, markets, epidemics and cities, and the decision makers who must know how far to trust them.
The field is organised by purpose rather than by domain. A model built to explain a mechanism is judged differently from one built to forecast, to explore policy options or to communicate. Verification asks whether the code does what the equations say; validation asks whether the model reproduces observed behaviour for the right reasons; and confirmation, for open systems, is always partial. Protocols for describing models, archives of community datasets, and the growing use of digital twins and machine-learned emulators have made the field more reproducible and more exposed to over-claiming at the same time.
Measures
Quantities and units
| Quantity | Unit or scale | What to know |
|---|---|---|
| Degree centrality, betweenness and eigenvector centrality | counts or normalised scores (0 to 1) per node | Rank the importance of a node by its links, by the paths that cross it or by the importance of its neighbours. Rankings change with the choice of measure and with missing edges; a node's rank is reported with its sensitivity to edge sampling. |
| Spectral radius of the adjacency matrix | dimensionless (largest eigenvalue) | Sets the epidemic or cascade threshold on a network: spreading with transmissibility above its reciprocal does not die out. It is sensitive to a few high-degree nodes and to unobserved links, so it is given with a range from network resampling. |
| Percolation threshold | fraction of nodes or edges removed (0 to 1) | The removal fraction at which the largest connected component collapses. Random and targeted removal give different thresholds; the gap between them is the measure of a network's exposure to deliberate or correlated failure. |
| Sensitivity index (Sobol first-order and total) | fraction of output variance (0 to 1) | The share of a model output's variance attributable to each input alone and with its interactions. Requires thousands of runs and a stated input distribution; the indices are only as meaningful as those distributions. |
| Ensemble spread and skill | standard deviation in the output unit; skill scores (0 to 1) | In intercomparison, the spread across models is a lower bound on structural uncertainty, not a confidence interval, because models share code, data and assumptions and their errors are correlated. |
પદ્ધતિઓ
Methods and their limits
Each method is listed with what it assumes and where it fails.
| પદ્ધતિ | Assumes | Fails when |
|---|---|---|
| System dynamics (after Forrester and Sterman) | A system can be represented as stocks, flows and feedback loops with aggregate variables; behaviour follows from the loop structure and delays rather than from individual heterogeneity. | When heterogeneity, spatial structure or network topology drive the outcome, when parameters are set by judgement without calibration, and when a loop diagram is presented as evidence of causation rather than as a hypothesis. |
| Agent-based modelling | Macroscopic patterns can be generated from rules followed by heterogeneous interacting agents, and the rules can be stated, documented (for example with the ODD protocol) and varied. | Rules that are plausible but unvalidated produce any pattern desired; results are hard to calibrate to data, computationally costly to explore and frequently irreproducible when the code and seeds are not archived. |
| Network analysis and percolation | The relevant structure is a graph whose nodes and edges are observed with known completeness, and that the process of interest spreads along the edges. | When edges are inferred, weighted by guesswork or change over time; when flows have capacity and direction the undirected graph ignores; when the multilayer coupling between networks matters more than any single layer. |
| Causal discovery on time series (for example PCMCI) | Causal links can be recovered from lagged conditional independence in observed series under causal sufficiency and stationarity. | Unobserved confounders, contemporaneous effects, non-stationarity and short records produce spurious or missing links; the output is a hypothesis for testing, not a validated causal graph. |
| Exploratory modelling and model intercomparison | When probabilities cannot be assigned to structural choices, running many model variants over a wide input space and comparing results across independent models bounds what is plausible. | Independence between models is often weaker than it looks because they share lineages and data; the ensemble can agree for shared wrong reasons, and the scenario space itself may exclude the outcomes that matter. |
| Digital twins | A continuously updated model linked to live data from a physical system can reproduce its state and test interventions. | For social and ecological systems the twin cannot be closed: human responses, unmodelled processes and data gaps mean it is an interpretable model, not a replica, and its outputs need the same validation discipline as any other model. |
Debates
What the field debates
- Explanation or prediction: what is a complex-systems model for?
- One school, following Epstein (2008), holds that models earn their keep by explaining mechanisms, exploring possibilities and disciplining intuition, and that demanding forecasts from them misreads their purpose. The forecasting school replies that policy needs numbers with skill scores, and points to ensemble weather and epidemic forecasting as proof that complex-systems models can be validated out of sample. Both agree that a model's purpose must be declared before its evaluation criteria can be chosen.
- Do machine-learned emulators and twins replace mechanistic models or depend on them?
- Evidence from weather forecasting shows that machine-learned ensembles can match or beat physics-based models on many scores at a fraction of the computing cost. Yet they are trained on reanalyses produced by physics-based data assimilation, and their behaviour outside the training distribution is untested. The debate turns on whether that dependence is permanent and on whether learned models can be trusted for the rare, unprecedented states that risk analysis cares about most.
- How much reproducibility should be required?
- Advocates of full reproducibility argue that code, data, seeds and environment should be archived for every published model, citing the share of agent-based results that could not be rerun from their descriptions. Others argue that complex models are legitimately exploratory, that replication of a mechanism matters more than replication of a run, and that mandatory archiving favours well-resourced groups. The ODD protocol and model archives are a middle ground that both sides accept.
Bodies and data
Bodies that set practice, and the data they publish
Bodies and standards are named so learners can find and apply their published work; the Academy teaches their use and links to the publisher, and no affiliation is implied.
| Body | Role in the field |
|---|---|
| Santa Fe Institute | The research institute that shaped complexity science as a field; its working papers and summer programmes are the common reference for emergence, networks and adaptation across disciplines. |
| Complexity Science Hub Vienna | Applies network and data science to supply chains, finance, health and infrastructure, including firm-level supply-network reconstruction used in systemic-risk assessment. |
| International Institute for Applied Systems Analysis (IIASA) | Builds and hosts integrated assessment and scenario models and the shared scenario databases on which many coupled-system studies rest. |
| Potsdam Institute for Climate Impact Research (PIK) | Develops coupled Earth-system, economic and network models and the causal-discovery methods used on climate and socio-economic time series. |
| System Dynamics Society | The professional society of the system dynamics method; its journal and conference set the standards for model documentation and testing in that tradition. |
| Dataset | Steward |
|---|---|
| Netzschleuder network catalogue | Tiago Peixoto, Inverse Problems and Complex Networks group |
| Stanford Large Network Dataset Collection (SNAP) | Stanford University |
| CMIP6 model output | World Climate Research Programme, served by ESGF |
| SSP scenario database | IIASA |
04પદ્ધતિ
From a degree distribution to a decision: how much of a network survives when nodes are removed
A network’s mean degree sets the fraction of nodes it can lose before it falls apart; the same network survives random failure far better than a targeted attack on its hubs, and a model of that survival is only useful once its sensitivity and its spread across runs are stated.
Sources: Newman, The structure and function of complex networks (2003); Albert, Jeong and Barabasi, Error and attack tolerance of complex networks (2000); Grimm et al., The ODD protocol for describing agent-based and other simulation models: a second update (2020); Saltelli et al., Five ways to ensure that models serve society: a manifesto (2020).
05At a glance
At a glance, with 5 related topics
- Build a twin of any placeTwin builderWorked example: the Tonle Sap and the Lower Mekong at Phnom Penh · 5 generators · 13 layers · 2 instruments
3D ગ્રાફ
પ્રણાલી અને Nexus અભ્યાસ ફેકલ્ટી
211 nodes1,557 relations9 views
06પ્રયોગશાળાઓ
How the subject is practised in labs
| Lab | What you do in it | Courses using it |
|---|---|---|
| Network lab | Build dependency graphs, find critical nodes, trace cascades | 2 |
| Map lab | Read hazard, exposure and boundaries on real maps | 1 |
07કન્સોલ
How Complex systems and systems modelling is taught, as a process you can run
Press Play to run this process from this page’s own data.
જટિલ પ્રણાલીઓ અને પ્રણાલી મોડેલિંગ
How Complex systems and systems modelling is taught
5 subjects in scope, 3 key terms, 1 course open, 1 lab, in the Systemic Risk and Complexity department of Systems and Nexus Studies: from the scope to the method and its terms, the courses that teach it, the labs on the twin, the assessment and the record.
-
01 Scope
- 01
Complexity concepts
Start · the discipline
Complexity concepts
- 02
Network analysis
Start · the discipline
Network analysis
- 03
Agent-based models
Start · the discipline
Agent-based models
- 04
Coupled systems models
Start · the discipline
Coupled systems models
- 05
Digital twins and their limits
Start · the discipline
Digital twins and their limits
- 01
-
02 Method and terms
- 06
From a degree distribution to a decision
Model · the learner
A network's mean degree sets the fraction of nodes it can lose before it falls apart.
- 07
Emergence
Record · the discipline
System behaviour that arises from interactions and is not visible in the parts alone.
- 09
Feedback loop
Record · the discipline
A chain of effects that returns to influence its starting point, amplifying or damping change.
- 11
Network centrality
Record · the discipline
Measures of how important a node is in a network, such as degree or betweenness.
- 06
-
03 Courses
- 08
RAS-401 જટિલતા અને નેટવર્ક
Step · the learner
Complexity and Networks: Seeing systemic risk through feedback, interdependence and network structure
- 08
-
04 Labs
- 10
નેટવર્ક
Exercise · the learner, on the faculty twin
The method runs on the faculty twin and the admitted map layers.
- 10
-
05 Assess
- 12
Module checks
Gate · the course
Each module ends with a short check against its outcomes.
- 13
Assessed work
Decision · the learner
The assessed work that closes a course: a lab record, a case memo or an assignment, reviewed against the rubric.
- 12
-
06 Record
- 14
પૂર્ણતા રેકોર્ડ
Outcome · the learner
The record stacks into the discipline achievement.
- 15
Discipline achievement
Record · the Academy
The discipline achievement follows the discipline rule, read from the records.
- 14
Complex systems and systems modelling · Systemic Risk and Complexity An outline of how the work is designed to run; it does not report the status of any decision or record.
08શિક્ષણ
Learning, practice and connections
Achievements
Achievements in this discipline
Course achievements
One for each course with this as its primary discipline, earned when you pass the course's assessed activities. One can be earned now.
Discipline: Complex systems and systems modelling
Complete 3 assessed courses in complex systems and systems modelling, at least 2 with it as their primary discipline and at least 1 at Intermediate (L2) or above, totalling at least 45 measured learning hours. In development.
Department: Systemic Risk and Complexity
Earn two discipline achievements in the Department of Systemic Risk and Complexity. In development.
Programme: Nexus-Based Risk Management flagship pathway
Complete every core course of the Nexus-Based Risk Management flagship pathway under assessed completion, ending with its capstone. In development.
An achievement is a record of the learning completed and assessed here. How recognition works.
માર્ગો
Pathways that teach it
- Nexus-Based Risk Management flagship pathwayAcademy flagship2 of its 20 courses are in this discipline.
- Climate risk officerRole pathwayOne of its 10 courses is in this discipline.
- Data and GeoAI specialistRole pathwayOne of its 12 courses is in this discipline.
- Disaster risk reduction practitionerRole pathwayOne of its 12 courses is in this discipline.
- Enterprise risk managerRole pathwayOne of its 12 courses is in this discipline.
Where this is practised
Where the subject is practised
- The Global Centre for Risk and Innovation (GCRI Canada)Systems Science and ModellingAgent-based and complexity models, Coupled systems modelling, Digital twins: method and limits, Model intercomparison, Networks and cascades, Optimisation under uncertainty, Reproducible modelling, Uncertainty and validation
- FoundryModelling and simulation
- પદ્ધતિઓ અને મોડેલિંગ
Field platforms of the ecosystem where the subject is studied and applied.
Practised in
The two Foundry systems where it is practised
- FoundryNXS-GRX systemCoupled system dynamics
- NXS-LAB systemSimulation experiments
Foundry system pages describe what each system and its rooms are designed to do; the courses teach the methods these systems apply.
Across the ecosystem
Across the ecosystem
The science, platforms and standards this discipline draws on and contributes to across the ecosystem.
Foundations in the STEM Commons
- The Global Centre for Risk and Innovation (GCRI Canada)Complex systems and agent-based modellingComplex systems and agent-based modelling study systems whose behaviour emerges from interactions among many components.
- Network science and contagionNetwork science and contagion study how the structure of connections among entities shapes the spread of failures, information and disease.
- Systems engineeringSystems engineering is an interdisciplinary approach to designing and managing complex systems over their life cycle.
Where the field is worked on across the network
- The Global Centre for Risk and Innovation (GCRI Canada)FoundryThe platform's work on this field.
Standards for the field
- Nexus StandardsSystemic RisksThe subsite that keeps the standards this discipline teaches.
- Digital, Cyber and AIThe subsite that keeps the standards this discipline teaches.
Pages of the ecosystem, cited so learners can follow the field into research, platforms, standards and national practice.
09ભાગ લો
Take part in Systems and Nexus Studies
તમે કોણ છો ત્યાંથી શરૂ કરો.
Routes
માર્ગ
રાષ્ટ્રીય સંસ્થાઓ
Ministries, statistics offices and agencies: a national desk and chapter, and a seat in the working groups. National chapters
માર્ગ
Sectors and professions
Operators, regulators, banks, insurers and professional bodies whose work touches systems and nexus studies. For organisations
માર્ગ
શિક્ષણ
Courses and practice, with a record of every course completed. Free to study. Open learning
માર્ગ
ધોરણો
Draft, review and adopt the methods and data standards with the working groups. Working groups
માર્ગ
Communities and research
Universities, civil society and individual experts: projects, data and review. Write to the Academy
Open calls
| Open call | Seeks | Closes |
|---|---|---|
| Founding call for experts to the working groups CALL-FOUNDING-EXPERTS | expert, editor, chair | 2026-12-15 |
| Open nomination for the three independent expert seats of the Global Standards Council CALL-GSC-IND | પરિષદ-સભ્ય | 2026-12-01 |
| Nominations of interim regional members (applies only if the transitional provision is decided) CALL-GSC-REG | પરિષદ-સભ્ય | 2026-12-15 |
| Review of the wave-1 risk core outlines (C1, C2, C3, C4, C8, C10, C11) CALL-REVIEW-WAVE1 | contributor | 2027-01-31 |
| જોખમ વિજ્ઞાનનાં મુખ્ય ધોરણો માટે નિષ્ણાતો: અનિશ્ચિતતા અને પુરાવા ગ્રેડિંગ, માપક્રમો, સંરચિત નિષ્ણાત નિર્ણય, સહનશીલતા, ચરમસીમાઓ અને સંયુક્ત ઘટનાઓ, આપત્તિજનક અને અસ્તિત્વગત જોખમ, નીતિશાસ્ત્ર અને સમાનતા, એકમો અને માપન (NRS-C17 થી NRS-C24) CALL-CORE-SCIENCE | expert, editor, chair | 2027-01-31 |
| Independent reviewers for the families of the hazard atlas, three from at least two regions for each family (NHS2-R19), priority families first CALL-ATLAS-REVIEW | નિષ્ણાત | 2027-03-31 |
| વચગાળાની અગ્રણી સલામતી સમીક્ષા પેનલના સભ્યો અને રેડ-ટીમ સમીક્ષકો (જૈવ સલામતી અને જૈવ સુરક્ષા, AI સલામતી અને સુરક્ષા, મહત્વપૂર્ણ માળખાગત સુરક્ષા); અરજીઓ ગુપ્ત રીતે સંભાળાય છે CALL-FRONTIER-SAFETY-REVIEW | નિષ્ણાત | 2026-11-30 |
| આવશ્યકતા રજિસ્ટર, NHS-2 મશીન પ્રતિનિધિત્વ, અનુરૂપતા પરીક્ષણ સમૂહો અને અનુરૂપતા મૂલ્યાંકન યોજના માટે ઓન્ટોલોજી, માપવિજ્ઞાન, સોફ્ટવેર ચકાસણી અને અનુરૂપતા મૂલ્યાંકનના નિષ્ણાતો CALL-CONFORMANCE-ONTOLOGY | expert, editor | 2027-01-31 |
Calls past their closing date are hidden automatically. Every call, proposal and route: open areas and calls.
Cells and CERTs
Gather the cell
Bring together three to seven people with complementary skills: data, field practice, policy and communication.
Register it
Write to the team with the cell's name, members, area and region. You receive a reference and a contact. Write to the Academy
Choose a charter
Pick one focus: research and evidence, data and maps, standards review, or community preparedness.
Learn together
Follow shared study paths at Risk Academy, with practice and a record of every course completed. Hazards and resilience
Lead and report
Hold regular sessions. Publish what the cell learns through Reports. Bring findings to the working groups. Working groups
Specialise as a Community Emergency Response Team (CERT)
A CERT prepares its neighbourhood for the hazards it faces. It builds a local hazard profile from this site's map. It trains in household and street preparedness. It learns first aid and light search through recognised local providers, and it runs drills. A CERT works alongside the local emergency services and follows their direction. It never replaces them and is not an emergency service.
સૂચકાંક
What a national team or a local community can do on The Global Risks Index (NXS-GRX) today.
NXS-GRX
Your country's profile
Open risk profiles for 195 countries, with sources and licences on every value. Open the Index
NXS-GRX
Correct or reply
Correct a value or reply for your country. The change is logged with its effect. Open the Index
NXS-GRX
Offer data
Offer a dataset. It enters through the licence gate, with your credit. Open the Index
NXS-GRX
Use the open data
Layers, profiles and tiles through the open API. Open the Index
10References
Core references and where to begin
Start with Epstein, then Sterman, then Oreskes.
- Forrester, J. W., Counterintuitive behaviour of social systems (1971). DOI
- Sterman, J. D., All models are wrong: reflections on becoming a systems scientist (2002). DOI
- Epstein, J. M., Why model? (2008). jasss.org
- Bonabeau, E., Agent-based modelling: methods and techniques for simulating human systems (2002). DOI
- Grimm, V. et al., The ODD protocol for describing agent-based and other simulation models: a second update (2020). DOI
- Newman, M. E. J., The structure and function of complex networks (2003). DOI
- Albert, R., Jeong, H. and Barabási, A.-L., Error and attack tolerance of complex networks (2000). DOI
- Kivelä, M. et al., Multilayer networks (2014). DOI
- Runge, J., Nowack, P., Kretschmer, M., Flaxman, S. and Sejdinovic, D., Detecting and quantifying causal associations in large nonlinear time series datasets (2019). DOI
- Oreskes, N., Shrader-Frechette, K. and Belitz, K., Verification, validation, and confirmation of numerical models in the earth sciences (1994). DOI
- Bankes, S., Exploratory modelling for policy analysis (1993). DOI
- Saltelli, A. et al., Five ways to ensure that models serve society: a manifesto (2020). DOI
11પ્રશ્નો
પ્રશ્નો, જવાબો અને મુખ્ય શબ્દો
શું અહીં અભ્યાસ મફત છે?
હા. Nexus Consortium ની આગેવાની હેઠળની Risk Academy નો દરેક અભ્યાસક્રમ અને શિક્ષણ સામગ્રી અભ્યાસ માટે મફત છે, સ્વ-ગતિએ અથવા સમૂહ સાથે. સામાન્ય ભાગીદારી માટે કોઈ ફી નથી; ચૂકવણીવાળું કાર્ય, ઇન્ટર્નશિપ અને ફેલોશિપ અલગ વ્યવસ્થાઓ છે.
અભ્યાસક્રમ પૂર્ણ કરવાથી મને શું મળે છે?
A completion record: the course, its version, the date and what was assessed, private by default and shared by you. It stacks into discipline, department and programme achievements. A record states what was studied and assessed and does not certify competence; it is not a licence, a professional title or an employment status.
ડેટા અને સ્થળો ક્યાંથી આવે છે?
Every faculty teaches on a twin of a real place built from open data; each layer carries its provider, licence and date, and every result on the Academy states its source. Maps and figures are indicators for teaching, never an assessment of a place.
How do I take part beyond the courses in complex systems and systems modelling?
Form a competence cell of three to seven people, study a shared pathway, take a supervised assignment in the National Working Groups, or bring a cohort from your institution. For organisations.
મુખ્ય શબ્દો
- Emergence
- System behaviour that arises from interactions and is not visible in the parts alone.
- Feedback loop
- A chain of effects that returns to influence its starting point, amplifying or damping change.
- Agent-based model
- A simulation of individual actors and their interactions.
- Digital twin
- A model of a physical system kept in step with data from it, with stated limits.
અહેવાલોયોગદાન આપો
આ પૃષ્ઠ પરથી પુરાવાનું યોગદાન આપો.
Reports is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
Risk Academy
- EmailGateway GW-ACADEMY Send it from this page Evidence, a correction, a case, a method, a review, a translation or profile input. It arrives tagged with this page, so the editor who owns it receives it.
- ZenodoOpen deposit Deposit it and submit to the community Publish your own record on Zenodo with your ORCID iD, then submit it to the Reports community there. Curators review it; the DOI stays yours.
- ReportsRisk Academy See every way to contribute The contribution types, how each is reviewed, what credit you receive, the licence and what is never accepted.
Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.
જટિલ પ્રણાલીઓ અને પ્રણાલી મોડેલિંગ
Start with an open course
The catalogue lists each course with its outcomes, labs, hours and level.
Open RAS-401Back to Department of Systemic Risk and Complexity