The approach for estimation of probabilistic risks for the performance of system human resource management process considering information security requirements is proposed. The recommended models for risks prediction are described. The use of the proposed approach helps to identify "bottlenecks", reduce risks in system human resource management process, justify conditions and period, in which guarantees of risks retention within admissible limits are maintained, taking into account the requirements for system information security. The usability of the approach is illustrated by examples.
In general, the main output of the human resource management process are information and nonmaterial results. The information results of management include plan for managing system human resource and personnel selection plans, personnel database, employment contracts, plans and reports on the implementation of projects. In turn, the non-material results include directly qualified and motivated personnel assigned to the relevant positions, acquired skills, publicly available knowledge, staff satisfaction with work, the level of staff turnover that meets the needs of the enterprise in employees, an acceptable socio-psychological climate at the enterprise, the required level of safety, quality and efficiency of the system and the innovative potential of the enterprise (connected with human resource) etc.
In the life cycle of systems, both the reliable performance of the human resource management process itself and the system information security proper to this process should be ensured.
To predict proper risks the approach for modeling human resource management process is proposed below.
To predict the risks for a given prognostic time it is proposed to use the following quantitative probabilistic measures:
human( ) − the probability of failure in reliable perform human resource management process without consideration of system information security; sec( ) − the probability of violating system information security requirements; int ( ) − the integral probability of failure in reliable perform human resource management process considering system information security.
To calculate the risk measures, the entities under study can be considered as a system of simple or complex structure. Models and methods for risks prediction use data obtained "upon the occurrence of events", according to the identified prerequisites for the occurrence of events, and data collected and accumulated statistics and possible conditions for their implementation of the process.
A simple structure system for modeling is a system consisting of a single element or a set of elements logically combined for analysis as a single element. The analysis of a simple structure system is carried out according to the «Black box" principle, when the inputs and outputs are known, but the internal details of the system operation are unknown. A system of a complex structure for modeling is represented as a set of interacting elements, each of which is represented as a «Black box" operating under conditions of uncertainty.
The modeling is based on using concept of the probabilities of "success" and/or "unsuccess" (risk
of "failure" considering consequences) during the given prognostic time period. There are
recommended some «Black box” models for which probabilistic space (Ω, B, P) is created (see for
example [
Applicable models for predicting such different risks, including the ways for generating models for
complex system with parallel or serial structure in the part of system human resource management
process, see in [
From engineering point of view the modelled system may be presented as “Black box” an as
complex system composed from «Black box” elements. There may be two cases for estimating the
probability of failure in “successful” operation of the j-th composing element (j ≥ 1) during given
prognostic time: the case of observed repeatability and the case of assumed repeatability of random
events [
According to observed repeatability the inputs for the calculations of human ( ) and/or sec ( ) (denoted below as fai ( )) use statistical data. Failure to perform the necessary actions of the j-th composing system is a threat of possible damage. From the point of view of the composition of actions and/or the severity of possible damage, all varieties of the actions can be divided into K groups, K ≥ 1 (if necessary). Based on the statistical data, the probability of failure to perform the actions of the j-th composing system element for the k-th group for a given time (it also may be related to human ( ) or sec ( ) ) may be calculated by the formula act ( ) = failure ( )/ ( ), (1) (2) where failure ( ) , ( )- are accordingly, the number of cases of failures when performing the necessary actions of the j-th composing system element and the total number of necessary actions from the k-th group to be performed in a given time .
The probability fai ( ) of failure in “successful” operation of the j-th composing system element during a given prognostic period is proposed to be estimated for the option when only those cases are taken into account for which the actions were not performed properly (they are the real cause of the damage):
fai ( ) = 1 − ∑ =1 [1 − ( )] (α )⁄∑ =1 , where is the maximum time for the j-th composing system element operation, including all particular values for the entire set of actions from different groups, taking into account their overlaps;
taken into account:
− is the quantity of actions for the j-th composing system element from the k-th group taken into account for multiple performances of the actions.
For the k-th group the requirement to perform the actions using the indicator function (α ) is (α) = 1, if condition α is peformed, 0, if condition α isn′t peformed. performing the necessary actions from the k-th group.
The condition α used in the indicator function is formed by the analysis of different specific conditions, proper to the j-th composing system element operation (defined in terms of system quality, safety, effectiveness etc.). It allows to take into account the consequences associated with the failure to perform the necessary actions – see (1), (2). Condition α means a set of conditions for all process actions, subject to quality, safety, effectiveness etc. and time constraints within the given time for 4.2.
As modelled system (concerning a formalization of human resource management process) there are considered as «Black box” with virtual random events affecting system operation – for estimating ( ) and/or
( ) in modelled system, presented as one element.
In general case “successful” modelled system operation is connected with counteraction against various dangerous influences on system integrity - these may be counteractions against human failures or “human factors” events in actions on time line.
There are proposed the formalization for the general technology of counteraction against various dangerous influences on system integrity. The technology is based on periodical diagnostics of system integrity, that is carried out to detect danger sources penetration into a system or consequences of negative influences (see Figure 1). The lost system integrity can be detected only as a result of diagnostics, after which the system recovery is started. Dangerous influence on system is acted stepby step: at first a danger source penetrates into the system and then after its activation begins to influence. The system integrity can’t be lost before penetrated danger source is activated. A danger for “successful” operation is considered to be realized only after a danger source has influenced on the modelled system.
It is supposed that used diagnostic tools allow to provide necessary integrity recovery after
revealing danger sources penetration into modelled system or the consequences of influences. Using
the probabilistic models (described in details in [
σ - frequency of the occurrences of potential threats (or mean time between the moments of the occurrences of potential threats which equals to 1/frequency); β - mean activation time of threats; Tbetw - time between the end of diagnostics and the beginning of the next diagnostics; Tdiag - diagnostics time; Trecov - recovery time
T - given prognostic period. 4.3.
For a complex system estimation with parallel or serial structure existing models can be developed by usual methods of probability theory. For this purpose in analogy with reliability it is necessary to know a mean time between losses of integrity for each element. Let's consider the elementary structure from two independent series elements this means logic connection “AND” and for two parallel elements this means logic connection “OR”. Let’s probability distribution function (PDF) of time between losses of j-th element integrity is Вj(t) =Р (τj≤ t), and random values τ1, τ2 are independent, then:
1) time between losses of integrity for system combined from series connected independent
elements is equal to a minimum from two times τj: failure of 1st or 2nd elements (i.e. the system goes
into a state of lost integrity when either 1st, or 2nd element integrity will be lost). For this case the
PDF of time between losses of system integrity is defined as
В(t) = Р(min (τ1,τ2)≤t)=1-Р(min (τ1,τ2)>t)= 1-Р(τ1>t)Р(τ2 > t)= 1 – [1-В1(t)] [1- В2(t)].
(3)
2) time between losses of integrity for system combined from parallel connected independent
elements (hot reservation) is equal to a maximum from two times τj: failure of 1st or 2nd elements
(i.e. the system goes into a state of lost integrity when both 1st and 2nd element integrity will be lost).
For this case the PDF of time between losses of system integrity is defined as
В(t)=Р(max (τ1,τ2)≤t)=Р(τ1 ≤ t)Р(τ2 ≤t)=В1(t)В2(t).
(4)
Note. The same approach is developed also by Prof. E.Ventcel in 80th and by others researchers, see [
Thus, an adequacy of probabilistic models is reached by the consideration of real processes of control, monitoring, element recovery for complex structure. Applying recurrently expressions (3) – (4), it is possible to receive PDF of time between losses of integrity for any complex modelled system with series and/or parallel structure. 4.4.
The integral probability of failure in reliable perform human resource management process considering system information security int ( ) for the period T is proposed to be calculated by the formula: int ( ) = 1 − [1 − human( )] · [1 − sec( )]. (5)
Here the probabilistic measure human( ) is probability of failure in reliable perform human resource management process without consideration of system information security and sec( ) is probability of violating system information security requirements. They are estimated according to recommendations of section 3 and subsections 4.1-4.3 considering the possible damage.
Note. The condition of independence between the random time before failure in performing the human resource management process and the random time before violating system information security requirements is supposed.
Without deviation from the general understanding of the proposed approach, the examples are given with reference to the human resource management process in application to standard IEC 62508 “Guidance on human aspects of dependability”.
Let some enterprise implement a set of actions for human resource management. According to the recommendations of IEC 62508, devoted to the analysis of the influence of the human factor on the system dependability, the main actions of the enterprise should be: the formation of human resources; the use of human resources; the development of human resources; the evaluation of efficiency related to human resource management.
Without going into the details of the considered aspects, the structure of actions set for receiving results of human resource management process is presented by Figure 2. For example 1 the actions set of system human resource management process is considered as complex modelled system. The approach of 4.3 is applied (because the approaches of 4.1, 4.2 are more simple, for them many aspects of system human resource management process are not considered). The elements of the modelled system are: 1st element (subsystem) - the actions of the formation of human resources; 2nd subsystem (elements 2.1 and 2.2) - the actions to use of human resources; 3rd element - the actions to the development of human resources; 4th element - the actions to the evaluation of efficiency related to human resource management.
Subsystem 2 is designated in the modelled system as two duplicate elements of the system elements 21 and 22. Duplication in practice means that actions are performed by more than one performer, one of whom is a person, the functions of another performer can be performed either by another person (for example, a boss) and/or supported by a robot and/or some artificial intelligence system. From the point of view of elementary events, such interaction essentially means that actions will be performed by subsystem 2 if " OR " element 2.1 "OR" element 2.2 will be in the elementary state "The integrity of the element of the modeled system is retained".
By definition, the reliable performance of human resource management process in the modelled system is considered to be ensured during a given prognostic period, if during this period the "AND" actions of the process for the formation of human resources (according to element 1), "AND" for the use of human resources (according to element 2.1 "AND"/"OR" element 2.2), "AND" for the development of human resources (according to element 4), "AND" for the evaluation of efficiency (according to element 4) are reliably performed. The prognostic period itself for an individual element can be interpreted as referring to the stage of creation (for threats inherent in this stage), and to the stage of operation in the future (for potentially possible threats), modeling the acceptability of solutions and confirming guarantees that acceptable risks are not exceeded. 5.2.
The risk of violating the reliability of the process performance without taking into account the requirements for system information security is estimated for modelled structure of Figure 1. Many possible threats affecting the each of the structural elements of the modelled system have been identified. At the same time, not only health threats and the possibility of human errors are taken into account, but also hypothetical threats associated with the possible consequences of these errors at the stage of enterprise operation. The generated input data for modeling, which cover each of the composite elements, are presented in Table 1. β - mean activation time of threats
Elements 1 time in 5 years (because of lost qualifications or knowledge for solving problems) 2 times in a year (because of insufficient qualifications or knowledge to solve problems or due to health problems of the staff) The same as for element 2.1 1 time in 5 years (because of the violation of the necessary terms of professional training and advanced training) 1 time in a year (because of the violation of the necessary deadlines or the quality of the periodic evaluation of the effectiveness of the process performance) 3 months up to possible damage 2 months up to possible damage 2 months up to possible damage 6 months up to possible damage 6 months up to possible damage Tbetw - time between the end of diagnostics and the beginning of the next diagnostics Tdiag - diagnostics time Trecov - recovery time
For all elements
1 time in a week The analysis of the calculation results showed that in probabilistic terms, the risk of failure in reliable perform human resource management process without consideration of system information security for 2 years will be about 0.02 for the entire set of actions (see Figure 2). With an increase in the prognostic period from 1 year to 4 years (see Figure 3), the risk increases from 0.043 to 0.241. For an acceptable risk at the level of 0.05, a period of up to 14 months is justified, in which guarantees are maintained that the acceptable risk is not exceeded in the conditions of the example from Table 1.
Figure 3. Dependence human ( ) on the prognostic period lasting from 1 to 4 years
The" bottleneck", the characteristics of which it makes sense to analyze for risk reduction, is only subsystem 2 – this is a set of actions for the use of human resources related to functional support, estimation and control. The identification of this "bottleneck" forces an additional analysis to identify ways to reduce the risk. The simplest option is to combine efforts in the use of human resources. These efforts imply mutual assistance, including mutual control of activities, and from the point of view of modeling in the structure, instead of element 2.2 with characteristics identical to element 2.1, the use of element 2.2, for which the frequency of occurrence of sources of threats associated with ineffective functional support, evaluation and control of actions (σ) will not be 2 times a year (as in Table 1 for medium-qualified personnel), but 1 time every 2 years, i.e. 4 times less often. This is quite achievable due to the performance of functions by a more highly qualified human performer and/or a robot and/or with the support of some kind of artificial intelligence system. All other input for modeling are the same as shown in Table 1.
As a result of additional modeling, it was found that due to the measures taken, the risk of failure in reliable perform human resource management process without consideration of system information security was reduced to the level of 0.076 (i.e. by 34.2%) and an increase from 14 to 16 months of the period for which guarantees of non-excess of acceptable risks are retained (see Figure 4). In practice, it is these measures (combining the efforts of several persons in the parallel solution of one task with mutual control of the prepared solutions) that lead to success. The example shows only a quantitative estimation of the results of applying such measures.
Continuing Example 1, the prediction of the risk of violation of information security requirements is illustrated for a set of actions according to the recommendations of ISO/IEC 27002 (Section 8) in terms of ensuring the safety of personnel (see Figure 5). The actions set is considered as complex modelled system. Still the approach of 4.3 is applied (because the approaches of 4.1, 4.2 are more simple for modeling in the example).
The input for each of the 3 constituent elements are presented in Table 2.
for 3rd element 2 times in a year (these are threats of damage caused by previous mistakes or due to dissatisfaction of
dismissed personnel) 1 day (it is assumed that
due to masking, the sources of threats are not activated immediately, but with a certain delay of at least 1 day)
1 hour (this time is determined by
the regulations for monitoring assets related to staff)
Analysis of the calculation results showed that in probabilistic terms, the risk of violating the requirements for information security within two years will be about 0.130 for the entire set of actions, amounting to 0.014 for the 1st element, 0.041 for the 2nd element, 0.080 for the 3rd element ("bottleneck"). With an increase in the prognostic period from a year to 4 years, the risk increases from 0.067 to 0.243. For an acceptable risk at the level of 0.050, a period of up to 8 months is justified, in which guarantees are maintained that the acceptable risk is not exceeded in the selected set of actions characterized by the conditions of the example from Table 2.
A "bottleneck" has been identified – it is the preservation of the ability of a person who has stopped or changed his duties to use the information received (element 3). At the same time, the cause of the "bottleneck" is a violator who is able (according to the accepted information security model) to use this hypothetical vulnerability during a day - see Table 2, the value for β - mean activation time of threats up to violation of information security. 5.4.
In continuation of Examples 1 and 2, the integral probability ( ) of failure in reliable perform human resource management process considering system information security is calculated using the recommendations of section 4. Considering that ( = ) = 0.076 and ( = ) = 0.130, by formula (5)
( = ) = 1 ─ (1─0.076)·(1─0.130) ≈ 0.196.
For commensurate damages in resulting value of integral risk 0.196 the risk of violating system information security requirements (0.130) is 1.7 times higher than the risk of failure to reliable perform human resource management process without consideration of system information security. Comparing with the admissible level of 0.05, we can state that the calculated risks exceed the acceptable risk (in probability value). It means the rationale that the system decisions are not balanced and the improvement of human resource management process is needed. And the main goal is to reduce the risk of violating information security requirements.
Thus, the examples 1-3 demonstrated a usability of the approach.
The proposed approach allows to estimate probabilistic risks for the performance of system human resource management process considering information security requirements. It uses the measure for uncertainty conditions – the integral probability of failure in reliable perform human resource management process considering system information security. Considering system information security the approach application helps to identify "bottlenecks" and the ways to reduce risks in human resource management process, and justify conditions and period, in which guarantees of risks retention within admissible limits are maintained, taking into account the requirements for system information security. power plants. Probability, combinatorics and control. IntechOpen, 2020, pp. 191-220. URL: https://www.intechopen.com/books/probability-combinatorics-and-control [21] I. Goncharov, N. Goncharov, P. Parinov, S. Kochedykov, A. Dushkin Modelling the information-psychological impact in social networks. Probability, combinatorics and control. IntechOpen, 2020, pp. 293-308. URL: https://www.intechopen.com/books/probabilitycombinatorics-and-control