The Method of Employee Competencies Management Based on the Ontological Approach Anna Bakanova[0000−0002−8674−3686] and Aleksey Shikov[0000−0002−9942−0907] ITMO University, 49 Kronverksky Avenue, Saint Petersburg, Russia bakanova55anna@gmail.com shik-off@mail.ru Abstract. The company’s management is based on a set of manage- ment decisions in various fields. Effective competency management in the company allows you to make the most appropriate decisions on any issue related to the selection and appointment of the main project executives or a specific job assignment. Ontologies simplify this process and help objectively choose the right employee based on their actual knowledge, skills, abilities, and also allow you to take into account various additional conditions. This paper presents an original method developed by the au- thors for managing the competencies of employees to make managerial decisions when choosing the best performers for specific projects and job assignments. The process of collecting and managing competencies is considered as a key element in making managerial decisions, the ba- sic concepts regarding this topic are determined, an analytical review of studies is presented, and a method is proposed that can be implemented in the competency management model presented in authors’ previous studies and research. Keywords: Ontologies · Competencies · competency management · Knowledge base · Ontological approach · Management tasks. 1 Introduction Competence is one of the most important concepts for any modern organization or industrial enterprise. In response to the need to promote continuous improve- ment and flexibility in staffing in the 1980s, a competency model emerged. In- stead of comparing the activities of the employee with pre-defined work activities, this model made it possible to qualify a person on the basis of the competen- cies he actually possesses. Torkkeli and Tuominen (2002) define competencies as opportunities related to the subject’s ability to use his resources [?]. Tobias and Dietrich (2003) represent competency as a set of personal characteristics (knowledge, skills, abilities) that are relatively stable in different situations [?]. Competencies expand the idea of an employee as a key element of the company. The same knowledge or skill may be included in different competencies, given Copyright c 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). 2 Anna Bakanova and Aleksey Nikolaevich Shikov that competencies have a complex structure, appropriate tools for working and accounting for competencies are required, which should ensure their effective and transparent management. The elimination of errors in the selection and appointment of appropriate employees for specific positions or determining the personnel of the performers of specific projects of the enterprise must be en- sured. Managers should have a clear idea of the potential and capabilities of the selected employees to solve specific work tasks and projects. The ontological approach, accounting and competency management allows this to be done with a higher degree of objectivity. In this case, one doesn’t have to figure out why the appointed qualified employees did not cope with a particular job or project. And the answer generally is not complicated: the selected staff did not have the necessary competencies. In addition, the proposed approach allows, based on the analysis of missing competencies, to determine individual learning paths and develop the necessary knowledge, skills. Obviously, today the company management is closely related to accounting and competency management. It should cover all processes that are involved in the production, implementation and development of specific competencies, in- cluding their identification, identifying gaps, filling gaps with the help of training sources, coaching and other types of corporate education. Competency manage- ment is also one of the elements of personalized corporate training as it allows employees to identify gaps in knowledge and skills, as well as gain new compe- tencies. Competency management requires a holistic structured approach, namely, a model developed by specialists that meets all the requirements of a modern com- pany. Jorge Luis Victoria Barbosa, Marcos Ricardo Kich et al. (2015) in their work compare different competency management models, such as MCSWILK, HRCSystem, MCGA and MASEL, and also offer their DeCom model, which takes into account the location of employees and information about training events [2]. In all of these models, improvements are possible regarding informa- tion on the level of competencies and the conditions for their use. In general, the main functions performed by the competency management model are the formation of a competency map in a company and support for corporate educa- tion. Using the proposed method, the set of conditions is expanded, with which you can detail the requests for management decisions, it is also possible to mon- itor competencies, thereby controlling their availability and updating through corporate educational programs. 2 Materials and methods Using ontologies allows one to apply modern experience, accelerate the processes of information interaction, save time and other resources, uniquely fix all the basic concepts and definitions, and eliminate ambiguity and duplication of infor- mation. The application of the ontological approach in knowledge management systems provides great opportunities for the analysis of available information. The combination of ontologies and workflows within the company allows you to The Method of Emp. Comp. Man. Based on the Ont. Appr. 3 create an intelligent management system based on the knowledge of the com- pany. D.V. Alexandrov, N.N. Zhebrun (2007) in their work propose using an ontological approach to create a business process management system for an enterprise [1]. This kind of system allows you to use all the accumulated and formalized experience in managing the organization. An environment is being created that provides the creation of a network corporate information technol- ogy space [10]. This approach allows one to share common knowledge, reuse it together and create a single information field for managing both company’s in- ternal and external processes. Sarma Cakula and Maija Sedleniece (2013) show in their work the use of an ontological approach for knowledge management and the organization of personalized e-learning [3]. The availability of that kind of information allows one to control the competencies’ distribution in the company and restore their lack. The objective of this study is to use the most flexible on- tological model focused on corporate learning. In order to do this, the ontological model presented below was developed, which is presented in Fig. 1. Fig. 1. Ontological model of accounting competencies in the learning process Each employee has the competence, which is assigned the value of “Level”, based on testing and interviews conducted in the company, as well as condi- tions for work and training are determined. Each of the conditions is assigned a semantic value “Score of meaning”, which is used in further calculations and analytics. Many modern studies show the relevance of understanding the context not only in the learning process, but also in the performance of work tasks. In ad- dition to the set of competencies for each employee, it is possible to determine a set of conditions in which it can be applied. In their work, M. Harzallah and F.B. Vernadat (1999) report that the context in which competence is used is crucial and that the same competence can be used to achieve different goals and objectives [4]. The context of competence is related to the area in which it is ap- 4 Anna Bakanova and Aleksey Nikolaevich Shikov plied. It presents the conditions and limitations in which competency should be used. Also, the meaning of context is mentioned in the work of Sergio Miranda, Francesco Orciuoli, Vincenzo Loia, Demetrios Sampson (2017) [6]. The context for the application of competence is important in the competency management model they have presented. Conditions are also an essential component of the DeCom model presented by Jorge Luis Victoria Barbosa, Marcos Ricardo Kich et al.(2015) [2]. Context allows one to consider competency as a complex component and in- clude various parts. One competency can be invested in another and at the same time created as a result of a request consisting of several competencies and condi- tions. For example, knowledge of the English language may have the conditions for its use: in business negotiations, when studying technical documentation, when working with specific vocabulary. These conditions require additional clar- ification as to whether the employee is ready to take on additional responsibilities or change the scope of activities. If the combination of conditions and compe- tence becomes permanent, then it becomes the new competency necessary for the company, by which a template request is generated. In order to solve any management problem, one needs to correctly determine the input data and what is necessary to look for as a result. The human factor plays a great role in this case [6]. But information technology can help, if not completely, but partially. The conditions for work and the task that must be solved are determined by the company and the current circumstances. Company executives decide what kind of employee is needed: manager, programmer, food industry technologist, and then determine further search parameters for the right employee [8]. But it happens that the problem needs to be solved suddenly and then there is no time to hire a new employee, it is necessary to choose from those who currently work in the company and are competent in the question posed or are ready to perform certain tasks. 3 Employee competency management method Competency management includes creating a competency map, regulating the needs for competencies, identifying gaps and filling them with sources of train- ing, coaching and other corporate educational activities. Competencies relate to synergy, integration, and interaction between technology, people, organiza- tional systems, and culture. Therefore, information technology is a key resource in helping managers map and develop competencies in the context of the current and strategic objectives of the company. The first point on the way to solving this problem is its formulation and description. At this stage, it is important to determine the set of conditions and the necessary competencies for its solution and form a request. Next, one needs to identify employees who meet the specified requirements, and then finds out the conditions in which employees with the necessary competencies are ready to work. To do this, it is proposed to calculate the compatibility result for this situation. To calculate this result, we use an approach based on the theory of The Method of Emp. Comp. Man. Based on the Ont. Appr. 5 fuzzy sets. Having received the value for each of the selected employees, it is nec- essary to compare them with the reference value specified earlier. For example, if as a result of calculations we get the value 0.87, and the set threshold is 0.8, then the employee is suitable for solving the problem [6]. Using this algorithm, we can calculate the values for all employees for each competency and various additional conditions. The function that describes this action is represented as follows: n X P (c) = wi µi (ci ) (1) i=1 where µi – is the membership function. Thus, the solution to the management problem is modeled by defining a set of attributes of conditions and competencies. Using a set of concepts, you can set a task, for example, in this case we need a set of conditions and competencies: C - a set of concepts defined by the task, C = (c1 , c2 , c3 , c4 ) (2) where c1 , c2 , c3 , c4 are the selected concepts that fill the knowledge base of the ontological model of the knowledge and competency management system in the company [11]. n P For each concept, the weight value wi where 0 ≤ wi ≥ 1 and wi = 1. i=1 which establishes the strength of the connection between the use case and the corresponding ontology concept. It is necessary to make a condition and build the parameters in the necessary sequence from the most important to the least important: (w1 = w2 ) > w3 > w4 . The weight wi is the more greater, the more important the presence of this parameter as the main one for solving the problem. To calculate the weights, we use the method of constructing the rating scale - the pair conditions method. Each of the parameters is compared in pairs and, based on the signs = , < and > , the priority is set, and the value 1, 2 and 0, respectively. A matrix of pairwise comparisons is formed: Table 1. Matrix of paired comparisons w1 w2 w3 w4 Sum Weight w1 1 1 2 2 6 0,38 w2 1 1 2 2 6 0,38 w3 0 0 1 2 3 0,19 w4 0 0 0 1 1 0,06 Accordingly, the weights are distributed as follows: w1 = w2 = 0, 38; w3 = 0, 19; w4 = 0, 06. The weight of concepts can be determined in advance for 6 Anna Bakanova and Aleksey Nikolaevich Shikov specific situations and used as a template. It is possible to determine the weight of the concepts and expand the search capabilities It is possible to take both the conditions and the competencies themselves as concepts. The membership function µi for the competence is set by the values µi = 1, µi = 0.5 and µi = 0, corresponding to the level of competency, where 0 is low, 0.5 is medium, 1 is high. Each of the conditions is also determined by three values. Accordingly, a large number of parameters can be made and it is possible to select more accurately the right employee in the weight distribution results, given the many conditions and combining several competencies. After making the request, the function P (c) calculates the maximum possible value, that is, the value µmax i and the weight wi we need are selected. The value µmax i does not always indicate the level of competency, and therefore, cannot always be equal to one. Regarding conditions, it shows different semantic meanings. For example, for the condition “Time” it can be µmax i = 1 – full-time, µmax i = 0.5 – part-time, µmax i = 0 – hourly payment. In addition, the request may not necessarily have excellent competency and a fairly average level, and therefore µi = 0.5. The maximum possible value of the function P (c)max or the query is calcu- lated automatically after setting the necessary values of membership functions for each parameter. Further, when analyzing data for each employee, the func- tion P (c) calculates the value for each. After that, the value P (c) for each k employees is compared with the value P (c)max obtained earlier and the absolute value of the difference is calculated: lk = P (c)max − P (c)k (3) where −P (c)k ≤ lk ≥ P (c)max . As a result of calculating the values lk the following answers are defined: 1) the employee with the lowest value lk → 0 meets the requirements of the request. 2) for lk = P (c)max query execution is impossible, that is, a suitable option was not found. 3) for lk < 0 employee parameter values exceed the values specified in the request. As a result of such comparison with the reference maximum value specified during the compilation of the request, it is also possible to determine how differ- ent each of the indicators wi µi ∈ [0; 1] for ci parameter k employees, calculating the following absolute value of the difference: wi µmax i − wi µki = ski (4) where −wi µki ≤ ski ≥ wi µmax i For this comparison of individual parameters’ values the following answers can be defined: 1) for ski → 0 the employee fully meets the requirements of the request for this parameter; 2) for ski = wi µmax i employee value is not enough; 3) for ski = −wi µki employee value exceeded The Method of Emp. Comp. Man. Based on the Ont. Appr. 7 4 Results In the course of the research, the verification and evaluation of the developed method was organized with the fulfillment of various requests on 6 users. The task of the search is as follows: to find a specialist with knowledge of SQL. The following competencies were identified in the experiment (elements for analyzing the “Knowledge of SQL” competency): “Setting access rights”, “Knowl- edge of the basic functions in SQL statements”, “Triggers”, “Designing the database structure”, “Stored procedures”, “Normalization/denormalization.” Table 2. Matrix of paired experiment comparisons w1 w2 w3 w4 w5 w6 Sum Weight w1 1 1 1 1 1 1 6 0,17 w2 1 1 1 1 1 1 6 0,17 w3 1 1 1 1 1 1 6 0,17 w4 1 1 1 1 1 1 6 0,17 w5 1 1 1 1 1 1 6 0,17 w6 1 1 1 1 1 1 6 0,17 Total 36 1,00 Table 3. Matrix of paired experiment comparisons Setting access rights database structure in SQL statements Stored procedures Knowledge of the denormalization Normalization/ basic functions Designing the Triggers Name P (c)k lk Anton 1 1 1 1 0 1 5 1 Alexey 1 1 0 1 0 1 4 2 Igor 1 1 0,5 0,5 0 0 3 3 Anna 1 1 1 0,5 0 1 4,5 1,5 Vasiliy 1 1 1 0,5 0,5 0,5 4,5 1,5 Egor 1 0,5 0,5 0 0,5 0 2,5 3,5 All these competencies are part of the more complex competency ”Knowl- edge of SQL”, and therefore determine the level of its knowledge. Each of the competencies-elements is important, as for the solution of the problem, “Set- ting access rights” and “Knowledge of the basic functions in SQL statements” 8 Anna Bakanova and Aleksey Nikolaevich Shikov may be necessary. In our request, it is important that the employee possesses all the competencies-elements, then it is possible to assign him the more com- plex competence “Knowledge of SQL”, determine the level of her knowledge and formulate recommendations for training. Due to the fact that all competencies-elements are equally important, the matrix of paired comparisons for our request will look like on the Table 2. To have the “Knowledge of SQL” competency, you need a high level for all Pn competencies-elements, and therefore P (c) = wi µi (ci ) = 6 i=1 Table 3 presents the membership functions µi for competencies-elements cor- responding to the level of competency, where 0 - low, 0.5 - medium, 1 - high. As you can see, according to the results of the calculations, only user Anton meets the condition lk → 0 the rest of the participants can receive training recommendations. 5 Conclusions and recommendations The presented competency management method was tested at the enterprises of the SCAUT group of companies and allowed to more than halve the errors in the selection and appointment of personnel for specific projects. It is applicable to competency monitoring, statistics collection and corporate training support. After implementing this method, it became possible to update the employee com- petency map, since one knowledge can come in different competencies, moreover, it became possible to collect separate data about employees working in certain projects and the competencies involved in them. If, for example, the number of people with this competency is less than 3, then you can accordingly engage in training employees or their retraining. If you want to get a certain position or role in the company, the employee can also check himself for the presence of the necessary level of competencies and additional conditions. In agreement with the management of the company, an employee can apply for training to obtain the relevant competencies. Thus, a bench for key positions will be formed. The method allows one to compare a potential employee for various parameters that are included in the request of the employer. The results of the presented studies are largely correlated with the results of other scientists: Overall, the contribution of the study has been to prove the viability of the General Systems theory when applied to the relationship between HRM practices and employee competence. This is evidenced from pre- vious studies, such as GhebregiorgisKarsten (2007), Zingheim Schuster (2009) and Fey et al., (2000), which established positive and significant relationship between the various HRM practices and employee’s hard and soft skill orienta- tion of competence. The application of the theory illustrates the effectiveness of HRM practices in enhancing employee’s competence at the firm level. This is crucial as firms’ workforce could be designed and conditioned to support the firms’ strategy and to boost the firms’ performance [7]. For many companies, a prerequisite for the workflow is to solve business problems with minimal time and resources. To do this, it is crucial to control the The Method of Emp. Comp. Man. Based on the Ont. Appr. 9 availability of these resources, and in this case, competencies. Using a detailed comparison, it is less complicated to identify the lack of complex, but necessary competencies in the company. By means of this method, one can calculate the total number of employees with certain competencies and set a minimum value for them. In the future, conducting calculations anew as a result of the dismissal, training or retraining of employees, comparing with this value, it is possible to monitor the presence of competencies in the company, carry out effective management and maintain their required level. References 1. Aleksandrov, D. V.,Zhebrun,N. N.: Ontological approach to the creation of a knowledge-based enterprise business process management syste. Bulletin of Tam- bov state technical University 5, 26–31 (2007). 2. Barbosa, J. L. V., Kich, M. R., Barbosa, D. N. F., Klein, A. Z., Rigo, S. J.: De- Com: A model for context-aware competence management. Comput. Ind. (2015). https://doi.org/10.1016/j.compind.2015.03.012 3. Cakula, S., Salem, A. B. M.: E-learning developing using ontological engineering. WSEAS Trans. Inf. Sci. Appl. 10, 14–25 (2013). 4. Harzallah, M., Vernadat, F. B.: Human resources competency management in enterprise engineering. IFAC Proc. Vol.(1999). https://doi.org/10.1016/s1474- 6670(17)56028-0 5. Ley, T., Albert, D.: Identifying employee competencies in dynamic work domains: Methodological considerations and a case study. Journal of Universal Computer Science (2003). https://doi.org/10.3217/jucs-009-12-1500 6. Miranda, S., Orciuoli, F., Loia, V., Sampson, D.: An ontology- based model for competence management. Data Knowl. Eng. (2017). https://doi.org/10.1016/j.datak.2016.12.001 7. Mohammed, Y.A. Alsabbah, HazrilIzwar Ibrahim: HRM practices and employee competence: a general system perspective. International Journal of Business, Eco- nomics and Law, vol. 4, Issue 1 (2014). ISSN 2289-1552 8. Munir, K., Sheraz Anjum, M.: The use of ontologies for effective knowl- edge modelling and information retrieval. Appl. Comput. Informatics (2017). https://doi.org/10.1016/j.aci.2017.07.003 9. Nikitina, L., Shikov, P., Shikov, Y., Bakanova, A., Shikov, A.: Recommended sys- tem of personalized corporate e-learning based on ontologies. in MATEC Web of Conferences (2018). https://doi.org/10.1051/matecconf/201819305074 10. Okulov S. A., Chunaev A.V., Shikov A. N. Formation of corporate training man- agement system by means of information technologies. Achievements of modern science vol.1, 4, 170-174 (2017). 11. Pérez-Rodrı́guez, R., Anido-Rifón, L., Gómez-Carballa, M., Mouriño-Garcı́a, M.: Architecture of a concept-based information retrieval system for educational re- sources. Sci. Comput. Program. (2016). https://doi.org/10.1016/j.scico.2016.05.005 12. Torkkeli, M.,Tuominen, M.: The contribution of technology selection to core com- petencies. Int. J. Prod. Econ. (2002). https://doi.org/10.1016/S0925-5273(01)00227- 4