=Paper=
{{Paper
|id=Vol-2588/paper5
|storemode=property
|title=Integration Deterministic, Stochastic and Non-Stochastic Uncertainty Models in Conflict Situations
|pdfUrl=https://ceur-ws.org/Vol-2588/paper5.pdf
|volume=Vol-2588
|authors=Tetiana Shmelova
|dblpUrl=https://dblp.org/rec/conf/cmigin/Shmelova19
}}
==Integration Deterministic, Stochastic and Non-Stochastic Uncertainty Models in Conflict Situations==
https://ceur-ws.org/Vol-2588/paper5.pdf
Integration Deterministic, Stochastic and Non-Stochastic
Uncertainty Models in Conflict Situations
Tetiana Shmelova1[0000-0002-9737-6906],
1
National Aviation University, Komarova Ave, 1, 03058, Kyiv, Ukraine
shmelova@ukr.net
Abstract. The authors present an approach to conflict management in groups of
operators, which is to enhance the effectiveness of Collaborative Decision Mak-
ing (CDM) in an organizational setting. The decision making (DM) models
such as DM in Risk and Uncertainty, DM in Certainty offered. The authors
made an analysis of the International civil aviation organization (ICAO) docu-
ments on risk assessment. To determine the quantitative characteristics of risk
levels, models for DM by the operators of the aviation systems under risk and
uncertainty have been developed. The new methodology includes the process of
Integration Deterministic Stochastic and Non-Stochastic Uncertainty Models
for Network Planning models in Conflict Situations.
Keywords: Collaborative Decision Making, Decision Making in Risk and in
Uncertainty, Decision Making in Certainty, Conflict management, Network
Planning, Stochastic and Non-Stochastic Uncertainty Models
1 Introduction
Conflict management has advantages in different organization systems, including in
Aviation systems, where operators decision making in difficult situations. Properly
managed conflict can improve group results of decisions [1; 2]. The effectiveness of
aviation systems and the provision of flight safety still depend primarily upon the
reliability of aviation specialists and human decision making, individual and group
outcomes of decisions.
In aviation, significant attention is paid to Safety Management. Safety is the state
in which the risk is reduced to and maintained at or below, an acceptable level
through a continuing process of hazard identification and risk management [3; 4]. In
determining an acceptable level of safety, it is necessary to consider such factors as
the level of risk that applies the cost/benefits of improvements to the system, out-
comes after a decision by Conflict management methods, and public expectations on
the safety of the aviation industry.
To determine an acceptable level of risk and balance between decision perfor-
mance and safety requirements, it is necessary to have quantitative characteristics of
DM under risk conditions in conflict (critical) situations. Safety is a dynamic charac-
teristic of aviation with the help of which risk factors for flight safety should steadily
decrease. The adoption of efficiency indices of ensuring flight safety is frequently
Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attrib-
ution 4.0 International (CC BY 4.0) CMiGIN-2019: International Workshop on Conflict Management in
Global Information Networks.
influenced by internal and international standards and also by cultural features [5].
Aviation systems cannot be wholly free from dangerous factors and connected with
them risks, while, the elimination of aviation events and serious incidents continues to
be the final goal of human activity in the sphere of aviation safety. Neither human
activity nor systems created by it guarantee a total absence of operating errors and
their consequences [6].
International civil aviation organization constantly develops and improves proac-
tive, based on the evaluation of the risks. A modern approach, founded on the charac-
teristics (performance-based approach – PBA), based on the next three principles
monitoring [7; 8].
the main accent on desired/necessary results;
decision making, oriented on desired/necessary results;
using facts and data while DM.
Herein the principle “using facts and data while decision making” admits that tasks
shall comply with the widely known in Western management criteria SMART (spe-
cific, measurable, achievable, relevant and tіmebound) [3; 8].
The Global Air Traffic Management Operational Concept [3] assumes provision
collaborative decision making (СDМ) between all operational partners [9]. Imple-
mentation of the CDM requires the use of a modern information environment based
on the concepts of System Wide Information Management (SWIM) and Flight &
Flow Information for a Collaborative Environment (FF-ICE) [3; 9; 10].
Such a level of accuracy of tasks determination may be achieved only using the
new methodology includes the process of Integration Stochastic and Non-Stochastic
Uncertainty Models for Network Planning models in Conflict Situations.
The purposes of the work are:
analysis multi-DM in certainty using Network Planning Models for all op-
erators in the same organization;
building DM models - DM in Uncertainty and DM under Risk for maximi-
zation of effectiveness and minimization of risks in the Network Planning
Model for operators in the organization.
2 The integration Stochastic and Non-Stochastic Uncertainty
Models to Deterministic Model of Multi-Decision Making
2.1 Deterministic Models in Multi-Decision Making
The Decision making by operators in a conflict situation and for the analysis of the
actions of operators with the aid of the Network Planning methods gave a chance to
obtain:
1. Identify the technology of H-Os for the project.
2. Determination of a group of DM in problem/conflict situation for building a
multi DM graph.
3. Expert estimation of priority of DM in problem/conflict situation using Ex-
pert Judgment Method (EJM)
4. Decomposition of main technology (problem/conflict situation) on proce-
dures of each DM graph.
5. Flowchart of performance technology works (problem/conflict situation) by
procedures for each DM graph.
6. Determination of the times of operating procedures using the Expert Judg-
ment Method (according to experimental, to statistics data too).
7. Structural-timing table of operational procedures and time on the operating
procedures in main technology (problem/conflict situation).
8. Network graph of operating procedures main technology (problem/conflict
situation).
9. Analysis each part of main technology (problem/conflict situation) using as-
sessment by DM in Stochastic Uncertainty (DM in Risk) and Non-stochastic
Uncertainty (DM in Uncertainty) methods.
10. Integration of Deterministic and Stochastic models for CDM.
11. Determination of critical time for each DM in the project (problem/conflict
situation) and main DM.
12. Determination of the critical path for each DM of the project main DM
As known, the environmental conditions (natural, social, communication, etc.) de-
termine the reaction of operators, while the reaction of the latter, in its turn, changes
the environmental conditions themselves. The systemic analysis has been carried out
as well as the formalization of the factors which affect DM by operators in the Air
Navigation System (individual-psychological, psycho-physiological and social-
psychological) in the emergency [11; 12]. The impact of individual-psychological and
socio-psychological factors on the professional activities of operators during the con-
flict situation and development from normal to catastrophic has been studied. On the
basis of the reflexive theory of bipolar choice, the expected risks of DM have been
studied and the influence of the external environment, previous experience and inten-
tion of the operator have been identified [11].
It is very important to create highly intelligent joint DM systems for operators
those decision problems in the one team. In research are presented DM models for
operators (pilots of manned and unmanned aircraft, air traffic controller's , engineers,
flight dispatch, etc.) in emergencies in ANS [11]; the deterministic and stochastic
models of DM for different operators of ANS and collaborative DM; stochastic mod-
els type Markov Chains; Stochastic models type GERT’s (Graphical Evaluation and
Review Technique) network; Neural Network models; Fuzzy logic models; Reflexive
models of bipolar choice; models of diagnostics of emotional state deformation in the
professional activity of operators in the ANS; Graphical-Analytical Models of situa-
tion Development; Graphical-Analytical Models of DM by human-operator (H-O)
etc. [11].
In the recent documents, ICAO defined new approaches - the organization of Col-
laborative Decision Making (CDM) by all aviation operators using collaborative DM
models (CDMM) based on general information on the flight process and features of
the critical situation [9; 10].
In the process of analysis and synthesis of DM models in critical situations makes
sense to simplify complex models and solutions. So, for example, stochastic and non-
stochastic of uncertainty models, the Markov and GERT-models, reflexion models
integrate into deterministic models. The models for decision and predicting of the
critical situation using CDMM – technology presented in Table 1.
Table 1. The models for decision in critical situations using CDMM-technology
Models Describing of modelling emergency
Expert assessment of the complexity of the
Situations, for example flight stages.
Neural Network Model to determine potential
alternative of the critical situation completion.
Determination of weight coefficients of neural
network (probabilities for the model – DM in
risk) and effectiveness of critical situation
completion: {YG ;YGаеr;YGlf; W}.
Fuzzy logic to determine quantitative estimates
of potential loss - functions of estimation risk
R / outcomes U for next models of DM in Risk
and Uncertainty-{gr}
DM in Risk (stochastic of uncertainty model).
Stochastic models types’ tree, GERT’s network
for DM and critical situation developing. The
optimal solution is found by the criterion of an
expected value with the principle of risk - Adopt
DM in Uncertainty (non-stochastic of uncer-
tainty models). In DM matrix: alternative ac-
tions А = {А1, А2, … Аi ,…, Аm}, states of situa-
tion or factors λ = {λ1, λ2, … λj ,…, λn} and
outcomes uij
DM in certainty (Deterministic model) using
Network Planning method and DM in Risk /
Uncertanty for each branch. Determined mod-
els for an operators with deterministic proce-
dure - ti; ;Тcr;Тmid;Тmin;Тmax
Optimal decision for action in critical situation
for all operators in team.
However, for the formation (modeling) of DM, H-O has the property such as the
ability to apply different levels of DM complexity depending on the factors that influ-
ence the DM. For DM in a difficult situation (S) it is necessary to identify:
The class of situation (Q);
Level of Complexity (U);
Choosing the optimal actions (A*).
For example, Q = {qj} - the set of consequences of choosing the completion alter-
native; U = {uj} - vector of the characteristics of the consequences, the results of the
choice of the alternative of the completion; A = {ai} is the set of alternative solutions)
and choice the optimal actions (A*).
On Figure 1 scheme of process of simplifying a difficult situation presented. This
process is necessary to apply for complex systems and solution too. It is important to
create Expert system (ES) when analyzing the complexity, significance, and responsi-
bility of subsystems before synthase and synchrony of collaborative mathematical
models.
Fig. 1. The scheme of process of simplifying a difficult situation
The ES, one branch of artificial Intelligence (AI), is a computer system that simu-
lates the DM ability of a human-operators. The ICAO documents recommend devel-
oping Intelligent ESs in aviation to support of operators [10]. Knowledge - character-
istics of systems obtained as a result of practice and professional experience of ex-
perts. To build an ES, the following Algorithm of the building of Expert Systems is
used:
The Algorithm of the building of Expert Systems.
1. Building main components of ES: Users interface; Database; Base Knowledge.
2. System analysis of complex system. Decomposition of complex systems on sub-
systems:
1) Definition subsystems for expert estimation of their significance and description
of the characteristics of subsystems.
2) Definition of criteria estimation and description of criteria features.
3) Estimation of subsystems using EJM by criterion and obtaining weight coeffi-
cients of subsystem significance by criterion.
3. Aggregation subsystems in systems.
1) Additive aggregation of subsystems:
n
Wj F ,i 1,n, j 1,m
i 1
i ij
(5)
2) Multiplicative aggregation of subsystems:
n
W j' F , i 1,n, j 1,m ,
i 1
i
ij (6)
4. Results of significance of subsystems in ES.
To build an ES, it is necessary to determine the significance of the subsystems (pa-
rameters, characteristics, values, etc.) in the system, which is investigated with the
help of expert knowledge. The main method for building the Knowledge Base in the
Expert System is the EJM. To build an ES, the following Algorithm of Expert Judg-
ment Method is used:
Algorithm of Expert Judgment Method (EJM)
1) Questioners for experts, m – is a number of experts, m 30
Matrix of individual preferences - determine opinion of the experts and their sys-
tems of individual preferences, Ri – is a system of preferences of i-expert, i 1, m .
2) Matrix of group preferences Rij
m
R i
Rgrj i1
3) The experts’ group opinion Rgrj (sample average, arithmetical mean): m
4) The coordination of experts’ opinion:
a. Dispersion for each factors (procedures, phases of flight of the aircraft,…):
R Ri
m
2
grj
Dj i 1
m 1
b. Square average deviation (Squared deviations): j D j
c. Coefficient of the variation for each factors (procedure, phases of flight of the
aircraft, etc…):
j
j 100%
Rgrj
d. Kendal’s coefficient of concordance or to provide interrogation of the experts:
,
12S
W m
m ( n n) m T j
2 3
j 1
Rs
e. Rating correlation coefficient (Spearman's coefficient):
n
6 (xij yij )2
j 1
Rsi 1 .
n(n 2 1 )
5) Significance of the calculations.
S
χf χt ,
2 2
1 1 m
m(n 1 ) Tj
a. The significance of the calculations W, criterion - χ2: 2 12(n 1 ) j 1
2
χ χ
2
where f - factual value of variable; t - table value of variable.
b. The significance of the calculations Rs using Student's t - criterion:
n2
t rs t st .
critical
1 rs2
6) Effectiveness of solution (preferences, priority of solutions) - weight coefficient
C R 1
wj n j ; C j 1 n
wj:
C j j 1
7) Results of solutions.
For example, analysis and synthesis of DM models in critical situation for opera-
tors in team presented on Figure 2.
When analyzing a critical situation in a team, each operator determines his actions
to solve this problem. After building a structural-timing table of operational proce-
dures with time on the operating procedures (using EJM for obtaining solution times)
building Network graphs of operating procedures for all operators on Figure 2.
Fig. 2. The DM in certainty for 3 operators (H-Os)
The parallel process of simultaneous execution of technological operations in the
situation can be represented as a consolidated multi-channel network. For a consistent
optimization of such a network in order to achieve the cross-cutting efficacy of joint
decisions, it is advisable to use a multi-criteria approach: achieving a minimum time
for consolidation of critical situation operators' actions. Ways to optimize the network
graph for performing procedures by operators in the critical situation (by minimizing
time with maximum safety) are:
Aopt min Rm .
Risk of CDM by the operators in the critical situation (Fig. 3, situation S2):
𝑅𝑚 = 𝐹𝑚 (𝑡𝑚 ; {𝐴, α, 𝑝, 𝑢}) = 𝑡𝑚 (∑𝑛𝑘=1 𝑝𝑘 𝑢𝑘 + α𝑘 ),
where 𝑅𝑚 and (<; >)𝑅𝑚−1 .
For example, for decision tree in Fig. 3:
𝑅3 (𝐴41 ; 𝐴42 ) = 𝐴41 , because 𝐴41 < 𝐴42 ,
where 𝐴41 = 𝑡3 (𝑝411 𝑢411 + 𝑝412 𝑢412 ) + α41 ;
𝐴42 = 𝑡3 (𝑝421 𝑢421 + 𝑝422 𝑢422 ) + α42 ;
Analogically for 𝑅2 (𝐴31 ; 𝐴32 ) and 𝑅1 (𝐴11 ; 𝐴12 ):
𝑅2 (𝐴31 ; 𝐴32 ) = 𝐴32 , 𝐴31 > 𝐴32 ,
𝑅1 (𝐴11 ; 𝐴12 ) = 𝐴12, 𝐴11 > 𝐴12.
u211; p211
A21; α21
u111; p111 21
u212; p212
u112; p112
2 u221; p221
22
A11; α11 11 A22; α22
u222; p222
u121; p121
u311; p311
1
u122; p122
12 A31; α31
A12; α12 u123; p123 31 u312; p312
3
u321; p321 A41; α41 u411; p411
32 u322; p322 41 u412; p412
A32; α32
4 A42; α42 u421; p421
42
u422; p422
R1 = F1(t1; {A, α, p, u}) R2 = F2(t2; {A, α, p, u}) R3 = F3(t3; {A, α, p, u})
Fig.3 Decision tree for example DM in critical situation (CS), for part of situation S2: t – is a time of CDM
stage; А – is an alternative of decision; α – is a shift in the risk of developing CS according to stages on
decision tree; р – is a probability of adverse effects; u – is a damage due to negative solution.
In order to simulate CDM under conditions of a critical situation, the next steps are
a deep analysis of CS; intelligent data processing; identification of situation; formali-
zation of the situation using integrated models; decomposition of the complex situa-
tion into subclasses; synthesis of adapted deterministic models to determine certain
actions.
For decision-making in the presence of several decisions (part of situation S1), it is
advisable to apply the DM model in uncertainty. In conditions of non-stochastic un-
certainty, when the probability distribution that corresponds to the factors which in-
fluence the decision making (DM), either unknown or cannot be determined, the
methodological basis for CDM is a matrix of decisions in Uncertainty (Figure 4).
Table 2. The matrix of DM in Uncertanty
Factors influence DM in critical situation
λ1 λ2 … λj … λn
Alternative А1 U11 U12 … U1j … U1n
actions in А2 U21 U22 … U2j … U2n
critical situa- … … … … … … ….
tion Аi Ui1 Ui2 … Uij … Uin
… … … … … … …
Аm Um1 Um2 … Umj … Umn
Selecting the method (criteria for analyzing the decision problem) of decision making
under uncertainty such as Laplace criterion (for often decisions); Criterion of Wald
(for rare decisions); Savage criterion (for re-calculation of decisions); Hurwicz crite-
rion (for decisions with different risk using the coefficient of optimism-pessimism) is
carried out in accordance with the conditions of a problem situation.
The aggregated deterministic model with integrated stochastic models is shown in
Fig. 4.
Fig.4 The aggregated deterministic model with integrated stochastic model
After correction network graphs using Stochastic and Non-Stochastic Uncertainty
Models next step the determination of critical time and the critical path of the optimal
collective solution
3 Conclusion
The CDM an uninterrupted process of presenting information and individual DM by
various interacting participants, as well as providing synchronization of decisions
taken by participants and the exchange of information between them. It is important
to ensure the possibility of making a joint, integrated solution with partners at an ac-
ceptable level of efficiency. This is achieved by completeness and accuracy of availa-
ble information. Solutions planning should provide using DM different models such
as deterministic models; stochastic and non-stochastic of uncertainty models; the
Markov and GERT-models, reflexion models. After analysis of the situation needs
synthesis (aggregation) of stochastic models for the correction of deterministic model.
References
1. Kuhn, T.; Poole, M. S.: "Do conflict management styles affect group decision making?".
Human Communication Research. 26 (4): 558–590. (2000).
2. Alper, S.; Tjosvold, D.; Law, K. S.: "Conflict management, efficacy, and performance in
organizational teams". Personnel Psychology (2000).
3. Global Air Traffic Management Operational Concept: Doc. ICAO 9854 (1st ed.). Internation-
al Civil Aviation Organization, Montreal, Canada (2005).
4. Air Traffic Management: Doc. ICAO 4444-RAC/5011 (5th еd.). International Civil Aviation
Organization, Montreal, Canada (2007)
5. Cross-Cultural Factors in Aviation Safety: Human Factors Digest № 16. Сirc. ІСАО 302-
AN/175. Canada, Montreal2004).
6. Safety Management Manual (SMM) (3rd ed.). Doc. ICAO 9859-AN 474. Canada, Montreal:
(2013).
7. State of Global Aviation Safety. – Canada, Montreal. In ICAO, (2013).
8. International Civil Aviation Organization. (2013b). State of Global Aviation Safety. Canada,
Montreal: Author.
9. Manual on Collaborative Decision-Making (CDM), 2nd ed., Doc. 9971. Canada, Montreal:
ICAO Publ., 2014.
10. Manual on Flight and Flow Information for a Collaborative Environment (FF-ICE), 1st ed.,
Doc. 9965. Canada, Montreal: ICAO Publ., 2012.
11. Shmelova T., Sikirda Yu., Rizun N., Salem Abdel-Badeeh, Kovalyov Yu.: Socio-Technical
Decision Support in Air Navigation Systems: Emerging Research and Opportunities: Interna-
tional Publisher of Progressive Information Science and Technology Research, USA, Penn-
sylvania. (2017)
12. Mazin Al Hadidi, J. Samih Al-Azzeh, R. Odarchenko etc, ZigBee, Bluetooth and Wi-Fi Com-
plex Wireless Networks Performance Increasing, International Journal on Communications
Antenna and Propagation, Vol. 7, № 1, рр. 48-56, 2017.
13. Shmelova T., Rizun N :Strategic Imperatives and Core Competencies in the Era of Robotics
and Artificial Intelligence Chapter 9. Decision-Making Models of the Human-Operator as an
Element of the Socio-Technical Systems. International Publisher of Progressive Information
Science and Technology Research, USA, Pennsylvania. (2016).
14. Kasatkin М. Sikirda Yu., Shmelova T. Network Analysis of Collaborative Decision Making
by Air Navigation System's Human-Operators during Emergency Cases in Flight Proceedings
of the National Aviation University (2019).