# Classification of Economic Regions with Regards to Selected Factors Characterizing the Construction Industry

^{*}

## Abstract

**:**

## 1. Introduction

- proposing our own universal methodology for the classification of economic regions that are characterized by different factors
- applying the developed methodology for the classification of Polish regions with regards to selected factors that characterize the construction industry within the aspect of occupational safety and economic development.

## 2. Literature Review—Application of Data Classification Methods in Scientific Research

## 3. Proposed Research Methodology

## 4. Application of the Proposed Methodology Using the Classification of Polish Voivodeships as an Example

## 5. Conclusions

- The proposed methodology for classifying objects is of universal character and can be used to group countries, regions of a given country and other territorial units, as well as economic facilities, e.g., enterprises. The grouping can be conducted with regards to various classification criteria, including criteria related to occupational safety in the construction industry. The necessary condition for an appropriate calculation is a substantive identification of factors that characterize regions or economic objects within the analyzed aspect, and also the obtaining of reliable numerical data on them.
- The basis of the proposed methodology is a method of multidimensional analysis of statistical data, namely cluster analysis. Specific solutions include data standardization, a measure of similarity in the form of Euclidean distance, grouping objects using the hierarchical method, and binding objects using the Ward method.
- The developed methodology was used to classify Polish voivodeships with regards to factors that characterize the rate of economic development in the construction industry and the level of occupational safety. The conducted calculations and analysis of the results allowed the following conclusions to be formulated:
- The qualitative and quantitative structure of statistical data, which was the basis for the classification of voivodeships, allowed four distinct clusters consisting from two to five voivodeships to be separated. Voivodeships included in a cluster are characterized with a similar level of occupational safety in the construction industry.
- The isolated clusters are characterized by different levels of similarity, which is confirmed by the values of the merging distance measure for individual clusters. Cluster ranking with regards to the similarity of the voivodeships that form clusters is as follows:
- cluster IV consists of the voivodeships Lubuskie, Podlaskie, Swietokrzyskie, Opolskie and Warminsko-Mazurskie,
- cluster III consists of the voivodeships Kujawsko-Pomorskie, Lodzkie, Lubelskie, Podkarpackie and Zachodniopomorskie,
- cluster I consists of the voivodeships Dolnoslaskie, Pomorskie, Malopolskie and Wielkopolskie,
- cluster II consists of the voivodeships Mazowieckie and Slaskie.

- The very big similarity between voivodeships located in clusters III and IV means that voivodeships included in these clusters are characterized by a similar level of construction and assembly production value, occupational safety, the number of people employed in the construction industry, and the number of people living in the voivodeship.
- The Mazowieckie and Slaskie are atypical voivodeships. They are the most different when compared with the others. Although they form one cluster, the distance at which the merging between them occurs is relatively large when compared to the merging distance in the other clusters.

- The proposed methodology can be applied in both the area of scientific research and engineering practice. The results of tests and analyses obtained using this methodology can be the basis for classifying and comparing objects and determining their rankings. The correct classification of objects (which are described by many factors) into groups can be important in determining the characteristics of a given community, making an assessment, or looking for dependencies that apply to this community. The practical aspect of the proposed methodology is connected to the possibility of formulating conclusions, which could be important at a higher management level.
- In the research conducted by the authors, information about voivodeships belonging to the same cluster will be used to possess statistical data from these voivodeships, which in turn will be used for the construction of multifactorial linear regression models for predicting indicators describing the level of occupational safety in the construction industry in the group of voivodeships. Mathematical models, which were developed in this way, will be more accurate when compared to the general model that was built for the whole of Poland. The creation of separate mathematical models for individual voivodeships is impossible, due to the insufficient amount of reliable statistical data that can be used to construct them.

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Central Statistical Office in Katowice. Indicators of the Sustainable Development of Poland 2015; Central Statistical Office in Katowice, Silesian Center for Regional Research: Katowice, Poland, 2015; ISBN 978-83-89641-54-0.
- Eurostat. Eurostat Regional Yearbook 2017 Edition; Publications office of the European Union: Luxembourg, 2017; ISBN 978-92-79-71616-4. [Google Scholar]
- International Labour Organization. Safety and Health at Work: A Vision for Sustainable Prevention. In Proceedings of the 20th World Congress on Safety and Health at Work 2014, Global Forum for Prevention, Frankfurt, Germany, 24–27 August 2014. [Google Scholar]
- Hoła, B.; Szóstak, M. Analysis of the state of the accident rate in the construction industry in European Union countries. Arch. Civil Eng.
**2015**, 61, 19–34. [Google Scholar] [CrossRef] - Central Statistical Office. Statistical Yearbook 2008–2016; Department of Statistical Publishing: Warsaw, Poland, 2009–2017.
- Stanisz, A. A Simple Statistic Course with the Use of STATISTICA PL on the Basis of Medical Examples, Volume 3: Multivariate Analysis; StatSoft: Cracow, Poland, 2007; ISBN 978-83-88724-19-0. [Google Scholar]
- Wierzchoń, S.; Kłopotek, M. Algorithms of Cluster Analysis; Publishing House of WNT: Warsaw, Poland, 2015; ISBN 978-83-7926-261-8. [Google Scholar]
- Tryon, R.C. Cluster Analysis: Correlation Profile and Orthometric (Factor) Analysis for the Isolation of Unities in Mind and Personality; Edwards Brothers: Ann Arbor, MI, USA, 1939. [Google Scholar]
- Hartigan, J.A. Clustering Algorithms; Wiley: New York, NY, USA, 1975. [Google Scholar]
- Tkaczynski, A. Segmentation Using Two-Step Cluster Analysis. In Segmentation in Social Marketing; Springer: Singapore, 2017; pp. 109–125. [Google Scholar]
- Markowska, M.; Sokołowski, A.; Strahl, D.; Sobolewski, M. Dynamic classification of regions of the European Union at the NUTS 2 level with regards to sensitivity to the economic crisis in the area of the labour market. HSS
**2015**, 22, 37–50. [Google Scholar] - Napoli, G.; Giuffrida, S.; Valenti, A. Forms and functions of the real estate market of Palermo (Italy). Science and knowledge in the cluster analysis approach. Green Energy Technol.
**2017**, 191–202. [Google Scholar] [CrossRef] - Wang, W.; Yang, J.; Gong, X. The Regional Real Estate Investment Environment Research Based on Prime Component Analysis: The Case of Shandong. In ICCREM 2017: Real Estate and Urbanization—Proceedings of the International Conference on Construction and Real Estate Management; ASCE: Reston, VA, US, 2017; pp. 217–224. [Google Scholar]
- Gabrielli, L.; Giuffrida, S.; Trovato, M.R. Gaps and overlaps of urban housing sub-market: Hard clustering and fuzzy clustering approaches. Green Energy Technol.
**2017**, 203–219. [Google Scholar] [CrossRef] - Wang, N.; Li, H.-M.; Tan, X.; Zhong, X.-R. Development evaluation of real estate industry in China’s major cities based on factor and cluster analysis. Xi’an Jianzhu Keji Daxue Xuebao. J. Xi’an Univ. Archit. Technol.
**2010**, 42, 590–594. [Google Scholar] - Migała-Warchoł, A.; Sobolewski, M. Evaluation of voivodeships diversification in Poland according to transport infrastructure indications. Quant. Methods Econ.
**2013**, 14, 89–98. [Google Scholar] - Zalewska, E. Application of cluster analysis and methods of linear ordering in the assessment of Polish higher education. Sci. Works Wroclaw Univ. Econ.
**2017**, 469, 234–242. [Google Scholar] - Chiang, Y.-H.; Wong, F.K.-W.; Liang, S. Fatal Construction Accidents in Hong Kong. J. Constr. Eng. Manag.
**2018**, 144. [Google Scholar] [CrossRef] - Wong, F.K.; Chan, A.P.; Wong, A.K.; Hon, C.K.; Choi, T.N. Accidents of Electrical and Mechanical Works for Public Sector Projects in Hong Kong. Int. J. Environ. Res. Public Health
**2018**, 15, 485. [Google Scholar] [CrossRef] [PubMed] - Drozd, W. Identification and profiling the pattern of construction accidents with the cluster analysis. Mater. Bud.
**2017**, 101–102. [Google Scholar] [CrossRef] - Hoła, B.; Nowobilski, T.; Szer, I.; Szer, J. Identification of factors affecting the accident rate in the construction industry. Procedia Eng.
**2017**, 11. [Google Scholar] [CrossRef] - Jarocka, M. Selection of the normalization formula in the comparative analysis of multi-feature objects. Econ. Manag.
**2015**, 113–126. [Google Scholar] [CrossRef] - Ward, J.H., Jr. Hierarchical Grouping to Optimize an Objective Function. J. Am. Stat. Assoc.
**1963**, 58, 236–244. [Google Scholar] [CrossRef] - Grabiński, T. Methods of Taxononometry; Cracow University of Economics: Cracow, Poland, 2007. [Google Scholar]
- Mojena, R. Hierarchical grouping methods and stopping rulet: An evaluation. Comput. J.
**1977**, 20, 359–363. [Google Scholar] [CrossRef]

**Figure 2.**Dendrogram—a graph showing the connection of individual voivodeships in the subsequent calculation steps.

$\mathit{v}$ | Economic Indicators | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|

Value of Construction and Assembly Production ${\mathit{I}}_{1,\mathit{v},\mathit{y}}$ | Number of People Employed in the Construction Industry ${\mathit{I}}_{2,\mathit{v},\mathit{y}}$ | Population of a Given Region ${\mathit{I}}_{3,\mathit{v},\mathit{y}}$ | Number of People Injured in Occupational Accidents ${\mathit{I}}_{4,\mathit{v},\mathit{y}}$ | |||||||||

2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | |

1 | 14,568.1 | … | 13,721.2 | 68,874 | … | 65,039 | 2,877,059 | … | 2,903,710 | 892 | … | 413 |

2 | 5288.9 | … | 6216.6 | 43,952 | … | 42,918 | 2,067,918 | … | 2,083,927 | 607 | … | 269 |

… | … | … | … | … | … | … | … | … | … | … | … | … |

16 | 7325.1 | … | 9104.0 | 39,288 | … | 35,483 | 1,692,957 | … | 1,708,174 | 362 | … | 144 |

$\mathit{v}$ | Parameters | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|

Value of Construction and Assembly Production ${\mathit{P}}_{1,\mathit{v},\mathit{y}}$ | Number of People Employed in the Construction Industry ${\mathit{P}}_{2,\mathit{v},\mathit{y}}$ | Population of a Given Region ${\mathit{P}}_{3,\mathit{v},\mathit{y}}$ | Number of People Injured in Occupational Accidents ${\mathit{P}}_{4,\mathit{v},\mathit{y}}$ | |||||||||

2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | 2008 $\mathit{y}=1$ | … | 2016 $\mathit{y}=9$ | |

1 | 0.734 | … | 0.394 | 0.481 | … | 0.291 | 0.397 | … | 0.395 | 0.534 | … | 0.288 |

2 | −0.537 | … | −0.504 | −0.264 | … | −0.323 | −0.254 | … | −0.251 | −0.110 | … | −0.294 |

… | … | … | … | … | … | … | … | … | … | … | … | … |

16 | −0.258 | … | −0.158 | −0.403 | … | −0.530 | −0.555 | … | −0.546 | −0.664 | … | −0.798 |

**Table 3.**The obtained groups of voivodeships, which are characterized by a similar speed of construction industry development and a similar level of occupational safety.

Cluster | Voivodeships | The Distance at Which the Voivodeships Merged |
---|---|---|

I | Dolnoslaskie, Pomorskie, Malopolskie, Wielkopolskie | 5.38 |

II | Mazowieckie, Slaskie | 6.25 |

III | Kujawsko-Pomorskie, Lodzkie, Lubelskie, Podkarpackie, Zachodniopomorskie | 2.26 |

IV | Lubuskie, Podlaskie, Swietokrzyskie, Opolskie, Warminsko-Mazurskie | 1.73 |

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Hoła, B.; Nowobilski, T.
Classification of Economic Regions with Regards to Selected Factors Characterizing the Construction Industry. *Sustainability* **2018**, *10*, 1637.
https://doi.org/10.3390/su10051637

**AMA Style**

Hoła B, Nowobilski T.
Classification of Economic Regions with Regards to Selected Factors Characterizing the Construction Industry. *Sustainability*. 2018; 10(5):1637.
https://doi.org/10.3390/su10051637

**Chicago/Turabian Style**

Hoła, Bożena, and Tomasz Nowobilski.
2018. "Classification of Economic Regions with Regards to Selected Factors Characterizing the Construction Industry" *Sustainability* 10, no. 5: 1637.
https://doi.org/10.3390/su10051637