AI for Flood Detection, Decision Patterns & more …
Authored by John Fitch (PPI Course Presenter and Principal Consultant)
Introduction
This article is the follow-on to the Introduction to Decision Patterns piece that was published in PPI SyEN edition 107, December 2021. The first article shared the author’s 30+ year journey concerning the discovery, use and refinement of decision patterns, the conceptual basis (definitions, information model, methodology and elements) behind this construct, and the variety of use cases in which decision patterns have been applied (to the author’s direct knowledge). The article also provided simplified examples of decision patterns for Enterprise Strategy, System/Product Design, Process Capability Design, Service Design and Curriculum/Courseware Design.
The original paper identified eight unique aspects of a decision-centric information architecture that is applicable when engineering a solution to any type of problem. These constructs include:
This paper will elaborate on each of these information constructs, explain how they fill some “missing link(s)” in a typical digital thread and by doing so enable new and valuable system engineering capabilities. In addition to the potential benefits offered by use of these capabilities, the author will highlight current challenges to the effective use of decision patterns and experience-based tips to increase the likelihood of first-time success in applying them on a project.
Readers are encouraged to review the first article prior to diving deeper into this subject matter. In particular, note the somewhat unique definition of a decision as an element that decomposes the problem domain:
Decision = a fundamental question/issue that demands an answer/solution
Things to consider
Ultimately all faults are decision faults. Any failure in an engineered system or system of systems can be traced (by repeatedly asking “What went wrong?”) to one or a combination of:
Because decisions and the rationale behind them are lightly captured in most projects and strategic initiatives, it is difficult to build the cause-effect, e.g., fishbone diagram back through the web of missteps to answer questions such as:
If your organization has difficulty in answering such questions with confidence, you may want hit the Pause button on your current MBSE or Digital Thread initiative in order to rethink the role that decisions play in your scheme.
Ultimately the author believes that such failures are best mitigated by a combination of process steps (methods) that populate a lean, but comprehensive decision-centric information model, all jumpstarted by using decision patterns tailored to problem types/domains. Software tools are also helpful in enabling effective decision management at scale by:
In an innovative leap forward, engineers at Rice University, Houston, TX have developed an advanced AI-based system capable of providing real-time sensing of flooded roads. This groundbreaking technology is designed to enhance public safety by delivering timely and accurate information on road conditions during floods. The system leverages artificial intelligence and a network of sensors to detect and report flooded areas, allowing drivers and emergency services to navigate or avoid hazardous routes more effectively.
The Technology Behind the System
The core of this new system is a sophisticated AI algorithm that processes data collected from a network of sensors placed in vulnerable locations. These sensors, which can be easily installed on existing infrastructure such as streetlights and traffic signals, continuously monitor water levels on roadways. The AI system analyzes the data and when water levels exceed safe thresholds and triggers real-time alerts that can be sent to drivers through mobile apps, GPS systems, and other digital platforms.
One of the key advantages of this system is its ability to provide real-time updates. Traditional methods of monitoring flooded roads rely heavily on reports from drivers or emergency responders, which can be delayed or inaccurate. In contrast, Rice University’s AI system offers instantaneous detection, allowing for quicker response times and more informed decision-making by drivers and authorities.
领英推荐
Collaboration and Impact
This development is the result of a collaborative effort between Rice University’s engineering department and local government agencies in Houston, a city that has experienced significant flooding in recent years. The system was tested in a pilot program where it successfully identified flooded areas during heavy rainfall events, providing valuable data that helped to mitigate risks and direct traffic away from dangerous conditions.
The potential applications of this technology extend beyond flood detection. With further development, similar AI systems could be adapted to monitor other types of natural disasters or environmental hazards, such as landslides or wildfires, offering a versatile tool for public safety and disaster management ... Read more (3-minute read)
Answered by Robert Halligan (PPI Managing Director and Course Presenter)
Cognitive Systems Engineering: Cognitive Systems Engineering is a professional discipline that uses systematic methods of cognitive analysis and cognitive design to ensure that cognitive work is both efficient and robust. The aim is to amplify and extend the human capability to know, perceive, decide, plan, act, and collaborate by integrating system functions with the cognitive processes they need to support. Note that this is not about integrating humans into systems or humans with systems. The focus is on the cognitive work with a particular emphasis on how we might employ technological functionality to support that work.
Cognitive Systems Engineering deals with socio-technical systems in which the work is information intensive. The socio in socio-technical refers to the social processes of communication, cooperation and competition.
Human Systems Integration: Within Systems Engineering, Cognitive Systems Engineering can be located in the specialty area of Human Systems Integration. The INCOSE Systems Engineering Handbook (V 3.1, August 2007, Appendix M) defines Human Systems Integration as the interdisciplinary technical and management processes for integrating human considerations within and across all system elements. This definition and the term itself implies that we wish to integrate humans with systems or possibly, that we wish to integrate humans into systems.
Both the term, Human Systems Integration, and the INCOSE definition can be taken to imply that the human should be subservient to technology and many in the engineering professions assume this perspective at least implicitly. Less egregiously, some accord equal status to humans and technology. Although many Human Factors professionals and Cognitive Systems Engineers also express this latter view, it is the humans in the system who make the decisions. Technology must be subservient to that demand. Cognitive Systems Engineering should be directed at ensuring robust and effective coordination between the humans in the system and also ensuring that human functionality is supported and enhanced by technical functionality. The implication of this perspective is that we do not want to integrate humans with technology but rather to integrate humans with humans and humans with work and to use technological capabilities to facilitate that.
Cognitive Systems Engineering can be deployed to good effect in any information-intensive work domain. Health care, military command-and-control and industrial power generation are just three work domains that can benefit from the systematic analysis and design of cognitive work. The focus is on helping workers think more effectively by design of support technologies, work processes or training. In that regard, Cognitive Systems Engineering seems most relevant to the Human Systems Integration domains of Manpower, Training, Human Factors Engineering and Safety as defined in the INCOSE Systems Engineering Handbook (V 3.1, August 2007, Appendix M).
That is not to belittle the remaining domains of Personnel, Environment, Occupational Health, Habitability and Survivability, or even to claim that Cognitive Systems Engineering could not contribute in those areas, but rather to point out that the current work in the Cognitive Systems Engineering does not currently offer much that is useful for those remaining domains.
Human factors engineering: Human factors engineering is a professional discipline that uses formal methods of analysis and design to ensure that work is both efficient and robust. Note the similarity of this definition to the one for Cognitive Systems Engineering; the only difference being that all references to cognition have been removed. Human factors engineering is a broader discipline that takes account of physical as well as cognitive work. Alternatively, it could be said that Cognitive Systems Engineering is a sub-discipline of Human Factors Engineering. The terms Ergonomics and Engineering Psychology are sometimes used instead of Human Factors Engineering. Some think of Ergonomics as a discipline that focuses primarily on physical work but that view is not universal. There is no useful distinction between Engineering Psychology and Human Factors Engineering.
I do stuff.
5 个月Now we just need to apply more systems engineering principles to develop more Toyota Land Cruisers, so floods are no longer a problem...??