Workshop Ticket (Week 4; submitted Week 3)
Engineering appears to be a practical field which would be immune to many of the problems and layers of complexity suffered by political and social processes and endeavours. However, Bar-Yam points out that it is not possible to take the human element out of engineering as systems are engineered by and for individuals and organisations. The relationships between human actors and their technology in engineered systems provides a new layer of complexity to be unravelled.
Thus engineers need to acknowledge the complexity of the problems they face and minimise the complexity inherent in their solutions. Engineering problems are made complex when they involve layers of interdependent components of technology and human stakeholders, where their interaction encourages both adaptivity and ‘operator error’. Engineers must avoid making complex problems worse by simplifying their objectives to the minimum functionality required to solve their problems.
Bar-Yam suggests that one method of effectively engineering the introduction of innovation into complex systems is what he calls ‘enlightened evolutionary engineering’. This method entails allowing various new technologies and procedures to evolve organically at different stages of the system. These solutions will all compete for the right to provide services in the system by a process mimicking natural selection as the more efficient solutions will be used more often.
However, the process of enlightened evolutionary engineering seems somewhat unlikely to succeed in practice. Using the example of air traffic control, Bar-Yam proposes that evolutionary engineering could be achieved by a redundancy of staff which allows new ideas to be tested while maintaining a safety override. Though he claims that this would cost less than wasteful conventional engineering approaches, surely the people in management would baulk at doubling their staff as a huge upfront cost. Management types would also be unhappy at having to relinquish control of their systems to allow lower-order ‘evolutionary’ innovation.
I would like to explore the efficacy of the technique of evolutionary engineering in my tutorial. We could have two groups develop solutions to a problem using ‘conventional’ and ‘evolutionary’ engineering methods. The conventional group might have a strong hierarchy for accepting ideas while the evolutionary group would contain smaller teams which competed to find solutions. We could all come back together to evaluate which group created the most solutions, and which was the more efficient (minimised duplication).
Perhaps Bar-Yam’s idea of using biological systems to solve complex human problems can find other, more palatable iterations. I would like to explore ways to use biology to explain and understand complex human-technological systems. Scientists and engineers are increasingly accepting the value of biomimicry, or copying animal forms, in their design. For example, the way that birds fly is a useful puzzle for those who work on improving the aerodynamics of planes, trains and professional cyclists.
I would like to have the tutorial group read part of this website about biomimicry: http://www.asknature.org/. I would then have them compare how useful the Ask Nature ideas and Bar-Yam’s ‘evolutionary’ approach are in solving the problems before engineers. We would extend our focus to explore how biological systems can improve our own complex systems. A relevant and recent example is Spiegelhalter and Arch’s ‘Biomimicry and circular metabolism for the cities of the future’ in Brebbia et al’s The Sustainable City VI: Urban Regeneration and Sustainability, available on google books. Rather than the engineering of planes or materials, this case study demonstrates how biology can be used to improve a system as complex as a city. We could then explore other such applications of biology to our systems. For example, one advantage of using biological systems to improve our own complex systems is that the former are self-organising and hence quite resilient. Bar-Yam mentions the immune system and it would be interesting to see whether the functioning of antibodies and cell defences could be used as an analogy to improve the resilience of technological systems such as the internet when faced with viruses or hacking.
Thus engineers need to acknowledge the complexity of the problems they face and minimise the complexity inherent in their solutions. Engineering problems are made complex when they involve layers of interdependent components of technology and human stakeholders, where their interaction encourages both adaptivity and ‘operator error’. Engineers must avoid making complex problems worse by simplifying their objectives to the minimum functionality required to solve their problems.
Bar-Yam suggests that one method of effectively engineering the introduction of innovation into complex systems is what he calls ‘enlightened evolutionary engineering’. This method entails allowing various new technologies and procedures to evolve organically at different stages of the system. These solutions will all compete for the right to provide services in the system by a process mimicking natural selection as the more efficient solutions will be used more often.
However, the process of enlightened evolutionary engineering seems somewhat unlikely to succeed in practice. Using the example of air traffic control, Bar-Yam proposes that evolutionary engineering could be achieved by a redundancy of staff which allows new ideas to be tested while maintaining a safety override. Though he claims that this would cost less than wasteful conventional engineering approaches, surely the people in management would baulk at doubling their staff as a huge upfront cost. Management types would also be unhappy at having to relinquish control of their systems to allow lower-order ‘evolutionary’ innovation.
I would like to explore the efficacy of the technique of evolutionary engineering in my tutorial. We could have two groups develop solutions to a problem using ‘conventional’ and ‘evolutionary’ engineering methods. The conventional group might have a strong hierarchy for accepting ideas while the evolutionary group would contain smaller teams which competed to find solutions. We could all come back together to evaluate which group created the most solutions, and which was the more efficient (minimised duplication).
Perhaps Bar-Yam’s idea of using biological systems to solve complex human problems can find other, more palatable iterations. I would like to explore ways to use biology to explain and understand complex human-technological systems. Scientists and engineers are increasingly accepting the value of biomimicry, or copying animal forms, in their design. For example, the way that birds fly is a useful puzzle for those who work on improving the aerodynamics of planes, trains and professional cyclists.
I would like to have the tutorial group read part of this website about biomimicry: http://www.asknature.org/. I would then have them compare how useful the Ask Nature ideas and Bar-Yam’s ‘evolutionary’ approach are in solving the problems before engineers. We would extend our focus to explore how biological systems can improve our own complex systems. A relevant and recent example is Spiegelhalter and Arch’s ‘Biomimicry and circular metabolism for the cities of the future’ in Brebbia et al’s The Sustainable City VI: Urban Regeneration and Sustainability, available on google books. Rather than the engineering of planes or materials, this case study demonstrates how biology can be used to improve a system as complex as a city. We could then explore other such applications of biology to our systems. For example, one advantage of using biological systems to improve our own complex systems is that the former are self-organising and hence quite resilient. Bar-Yam mentions the immune system and it would be interesting to see whether the functioning of antibodies and cell defences could be used as an analogy to improve the resilience of technological systems such as the internet when faced with viruses or hacking.