Research lead: Professor David Fisk, Dr Panagiotis Angeloudis, Professor Washington Ochieng
What is the problem?
In a conventional approach to building a system, the whole design is tested only when construction is complete. This approach works for simple projects but becomes unreliable as the system complexity increases. The tried and tested systems engineering solution breaks the design task down into layers of ever smaller blocks until they correspond to identifiable components. The design is then realised by assembling the layers of blocks, testing realised performance against the design at each level of assembly. However, as the complexity of the design increases further, even this approach incurs problems. Controlling growing complexity is the issue.
How does our research address this?
Complexity has no intrinsic value, but it is an inevitable consequence of a large multi-function product. Although complexity is a common experience, research has brought some much-needed clarity into what it might mean. Although each engineering discipline may feel it has its own complexity problems, research can draw on a wide field to make more general insights. Disciplines such as ecology, computer science and statistical physics have provided insights into how engineering complexity can be better understood. It has also been possible to look at the large data sets that are now available for real world complex system dynamics.
What have we achieved so far?
In computer theory, the Kolmogorov complexity of an entity is the shortest full description that retains all its detail. So, whereas the description of a straight tunnel might simply be the same few meters repeated, the description of a new hospital of similar cost would have much less to simplify. In an argument originally made in ecology, it is the size of the complex core of the compressed description that hints at whether the system could become unstable. The larger the complex core the more chance it has to find an unstable pathway. This argument has helped understanding of how the system dynamics will change as more system components are added. For example, analysis of data from the world’s metro systems showed that they tended to avoid adding to stations with high levels of line connectivity for practical reasons, but as a consequence their flows would be more stable and more robust against attack. In contrast adding more flights to a hub risks a global system that could be unstable to interruptions. Now digital complexity adds a new dimension and research challenge.
What are the future directions for research and industry?
Data techniques such as building information models and digital twins promise to improve infrastructure design and assembly for complex projects. However digital technology presents a whole new world of complexity, because a program has an infinite number of possibilities. The only way to be sure that a program will have the properties required is to run it. As a consequence, the classical systems engineering approach of breaking down a system into subsystems is not guaranteed to provide assurance that the final stages will work when brought back together. The response has been to adopt an approach that is sometimes called ‘agile’. The most essential required outputs are programmed first. The next most essential are then added and tinkered with until they too work. In some engineering applications this approach may not be applicable, and the software simply has to be tested exhaustively. This can take time, and for that reason leaving writing software until the last stage of a project may be a disaster. So how should projects be run that benefit from the flexibility of data-driven systems without crashing into complexity failure?
What is the long term vision for impact?
A society that has the hope to remain sustainable needs to master how it will contain the threat of increasing complexity. The ability to collect, share and analyse data should give much better performance from future systems. This should include their resilience to shocks. However, without research on how this can be delivered, it must only be a matter of time before complexity overwhelms us all.
Related researchers: Dr Panagiotis Angeloudis, Dr Michel-Alexandre Cardin, Professor David Fisk, Professor Washington Ochieng
More information
- Complexity and Resilience at our 10 year celebration event
Related publications
- Whyte, J., Mijic, A., Myers, R. J., Angeloudis, P., Cardin, M. A., Stettler, M. E. J. & Ochieng, W. (Forthcoming). A Research Agenda on Systems Approaches to Infrastructure. Civil Engineering and Environmental Systems. DOI: https://doi.org/10.1080/10286608.2020.1827396
- Goldbeck, N., Angeloudis, P. & Ochieng, W.Y. (2019) Resilience assessment for interdependent urban infrastructure systems using dynamic network flow models. Reliability Engineering & System Safety. 188, 62–79. DOI: https://doi.org/10.1016/j.ress.2019.03.007.
- Babovic, F., Mijic, A. & Madani, K. (2018) Decision making under deep uncertainty for adapting urban drainage systems to change. Urban Water Journal 15 (6), 552–60. DOI: https://doi.org/10.1080/1573062X.2018.1529803
- Chatzimichailidou, M. M., & Whyte, J. (2018) Part III Complex Systems Engineering Annotated Bibliography. KTN Report. v1.2.
- Martinetti, A., Chatzimichailidou, M.M., Maida, L. & van Dongen, L. (2018) Safety I–II, resilience and antifragility engineering: a debate explained through an accident occurring on a mobile elevating work platform. International Journal of Occupational Safety and Ergonomics 25 (1), 66–75. DOI: 10.1080/10803548.2018.1444724
- Fisk, D. (2017) Deep thought. Physics World. 30 (8), 18. DOI: https://doi.org/10.1088/2058-7058/30/8/31
- Ossa-Moreno, J., Smith, K.M. & Mijic, A. (2017) Economic analysis of wider benefits to facilitate SuDS uptake in London, UK. Sustainable Cities and Society. 28, 411–9. DOI: 10.1016/j.scs.2016.10.002
- Santos, P.L.C.T., Monteiro, P.A.A., Studic, M. & Majumdar, A. (2017) A methodology used for the development of an Air Traffic Management functional system architecture. Reliability Engineering & System Safety. 165, 445–57. DOI: 10.1016/j.ress.2017.05.022
- Studic, M., Majumdar, A., Schuster, W. & Ochieng, W.Y. (2017) A systemic modelling of ground handling services using the functional resonance analysis method. Transportation Research Part C: Emerging Technologies. 74, 245–60. DOI: 10.1016/j.trc.2016.11.004
- Anvari, B., Angeloudis, P. & Ochieng, W.Y. (2016) A multi-objective GA-based optimisation for holistic Manufacturing, transportation and Assembly of precast construction. Automation in Construction. 71, 226–41. DOI: https://www.sciencedirect.com/science/article/pii/S0926580516301558
- Chatzimichailidou, M.M., & Dokas, I.M. (2016) Risk SOAP: On the Relationship Between Systems Safety and the Risk SA Provision Capability. IEEE Systems Journal. 12 (2), 1–10. DOI: 10.1109/JSYST.2016.2614953
- Fisk, D. (2016) 'Feedback Exergy exegesis', Physics World, 29(2), 20. https://doi.org/10.1088/2058-7058/29/2/26
- Tobaruela, G., Schuster, W., Majumdar, A. & Ochieng, W.Y. (2015) Framework to Assess an Area Control Centre's Operating Cost-efficiency: A Case Study. Journal of Navigation. 68 (6), 1088–1104. DOI: 10.1017/S0373463315000302
- Wilke, S., Majumdar, A. & Ochieng, W.Y. (2014) Airport surface operations: A holistic framework for operations modeling and risk management. Safety Science. 63, 18–33. DOI: https://doi.org/10.1016/j.ssci.2013.10.015
- Angeloudis, P., & Fisk, D. (2006) Large subway systems as complex networks. Physica A: Statistical Mechanics and its Applications. 367, 553–8. DOI: https://doi.org/10.1016/j.physa.2005.11.007
- Fisk, D., & Kerherve, J. (2006) Complexity as a cause of unsustainability. Ecological Complexity. 3 (4), 336–43. DOI: https://doi.org/10.1016/j.ecocom.2007.02.007
- Fisk, D. (2004) Engineering complexity. Interdisciplinary Science Reviews. 29 (2), 151–61. DOI: https://doi.org/10.1016/j.ecocom.2007.02.007