Modelling problems in a regional labour market in Poland with MARS and CMARS — supported by optimisation

The study mainly aims to learn about job candidates’ characteristics that influence the declared levels of their general skills (G). The set of characteristics examined included common skills (C), professional skills (P) and other economic and social characteristics of candidates. In this context, it is essential to observe the required knowledge and core skills on the market and to make this knowledge available to decision-makers from public institutions, entrepreneurs and educational institutions. The main research objective is also related to verifying the method used to analyse large multi-dimensional databases of competencies that can be used to improve vocational education and its fit to the labour market. An additional objective of the study is also to find the relationship between the characteristics of the offers (the time of visibility and publication of offers, the distance of the job from the place of residence) and the grouping of competencies in individual professions and occupational groups. It is linked to another problem: identifying factors that influence the design and availability of employee competencies. MARS and CMARS methods were used as neutral computing methodologies to determine the relationship between the response variable (G) and the input variables. Their use allows for commenting on how fulfilment can be determined between the requirements of employers and decision-makers in government. The duration of visibility and publication of the offer might be informative for job candidates and, thus, significantly influence the development of competences in the region. The innovative use of advanced statistical methods to achieve the goal in the area of competence management allows for high precision of the results, reducing risks in making management decisions.