Collaborative Decision Making in Emergencies by the Integration of Deterministic, Stochastic, and Non-Stochastic Models

Collaborative Decision Making in Emergencies by the Integration of Deterministic, Stochastic, and Non-Stochastic Models

DOI: 10.4018/978-1-7998-7210-8.ch010
(Individual Chapters)
No Current Special Offers


The author makes an analysis of the ICAO documents on risk assessment. To determine the quantitative characteristics of risk levels, models for decision-making (DM) by the operators (pilots, air traffic controllers, engineers) under risk and uncertainty have been developed. The new methodology includes the process of integration deterministic, stochastic, and non-stochastic uncertainty models. Application of artificial intelligence (AI) models for the organization of collaborative decision making (CDM) by all aviation operators using individual models based on general information on the flight. The CDM models involve an uninterrupted process of presenting information, ensuring the synchronization of decisions, and the exchange of information an acceptable level of efficiency and safety were obtained. Models of multi-stage DM in risk conditions are presented taking into account threats at the stages of development of the situation. In addition, the chapter presents some examples of DM in an emergency (light strike) by the author and students at National Aviation University, Ukraine.
Chapter Preview


Every year aircraft are becoming more reliable and secure; however, it is not possible to completely eliminate the human factor, which is one of the most common causes of accidents. One of the primary reasons for the accidents is an 80% deviation in the actions of aviation personnel in the organization and operations of flights; only 20% are due to failures of the aviation equipment (International Civil Aviation Organization (ICAO), 2004; Interstate Aviation Committee (IAC), 2016). The effective interaction of aircraft crew and the Air Traffic Control Operator is, therefore, a prerequisite for ensuring flight safety in standard conditions and in emergencies. Of particular importance is the coherence of communication between the crew and the controller in situations that arise in flight due to the influence of dangerous factors, i.e., emergencies (ICAO, 2007). Several common characteristics in emergencies are an acute shortage in time for Decision-Making (DM), incompleteness in communication, a lack of information, and the significant psychophysiological load on the aircraft crew.

The human factor remains the primary cause of aviation accidents (ICAO, 2004). People are the most flexible, adaptable, and important element in the aviation system, but they are also vulnerable to circumstantial impact on the quality of operations. At the initial stage in the development of aviation, many problems were caused by mechanics and emerging or evolving technologies, ecology, and ergonomic conditions (e.g., the design and configuration displays, controls, and cockpit equipment, and the layout of the cabin) in operational conditions such as exposure to noise, vibration, heat and cold, as well acceleration forces. The optimization of the human role in complex systems is related to all aspects of human activity, such as decision-making processes and obtaining knowledge; the communication opportunities, psychological, and the psycho-emotional states of the operators, as well as the interaction between operators; the technical opportunities of maintaining communications and new software; the preparation of flight plans and maps; and more recently improving development in the application of Artificial Intelligence (AI) (ICAO, 2018, 2019; International Air Transport Association (IATA), 2019).

The effects of the human factor on safety performance has continued to evolve through the development and testing of models that can be tested and applied in practice. Since the 1970s, ICAO has pursued a systematic approach to aviation safety using conceptual safety models such as the SHEL (S-Software, H-Hardware, E-Environment, L-Liveware) model. New components have been developed and added, such as SHELL (L-L), SHELL-T (L-L- Team), SHCHELL 's models (C- Culture), Reason’s model of latent conditions; the Threat and Error Management (TEM) model, and other models of human factors (ICAO, 1998, 2002, 2004, 2013a, 2013b, 2014, 2017, 2018). At the center of the SHEL model is a human individual—the most critical and most flexible component in the system—to which other components of the system must be carefully matched if stress and eventual breakdown in the system are to be avoided.

In order to achieve this matching, an understanding of the characteristics of this central component (i.e., the human-operator (H-O)) is essential. People are subject to considerable variations in decision making and behavior in the complex systems and suffer many limitations, most of which are now predictable, at least in general terms (Campbell, Bagshaw, 2002; ICAO, 2017). ICAO addresses the application of innovative technologies in the continuing development of aviation systems, such as the inclusion of AI methods, with modern information technology (ICAO, 2018; 2019; IATA, 2018).

Key Terms in this Chapter

Collaborative Decision Making (CDM): Collaborative Decision Making by operators in an Air Navigation System.

Expert Systems (ES): A computer system that emulates the decision-making ability of a human expert. Expert systems are designed to solve complex problems by reasoning through bodies of knowledge represented primarily as If–Then rules rather than through conventional procedural code.

Air Navigation System: A complex of organizations, personnel, infrastructure, technical equipment, procedures, rules and information that is used to provide of airspace users of safe, regular and efficient air navigation service.

Expected Operating Conditions: The operating conditions in which pilot’s actions are prescribed by the flight manual.

AI (Artificial Intelligence): The simulation of human intelligence processes by modeling, computer systems, and machines.

Unexpected Operating Conditions: Operating conditions in which the pilot’s actions are not prescribed by the flight manual.

Markov Network: A graphical model in which a set of random variables possess the Markov property described by an undirected graph. The Markov network differs from the other graphic model, i.e., the Bayesian network, in the representation of the dependencies between random variables. It can express some of the dependencies that the Bayesian network cannot express (e.g., cyclic dependencies) while there are others that it cannot express.

GERT Network (Graphical Evaluation and Review Technique): An alternative probabilistic method of network planning that is applicable when these actions can start only after the completion of a prior action including cycles and loops.

Complete Chapter List

Search this Book: