DOI QR코드

DOI QR Code

A Study on Data Mapping for Integrated Analysis of Railway Safety Data

철도 위험관리 데이터 연계 분석을 위한 기준 데이터 매핑 연구

  • Byun, Hyun-Jin (Safety Innovation Headquarters, Korea Railroad Corp.) ;
  • Lee, Yong-Sang (Department of Railroad Management, Woosong University)
  • Received : 2017.09.04
  • Accepted : 2017.09.29
  • Published : 2017.10.31

Abstract

The railway system is an interface industry that can be safely operated by organically operating the lines, vehicles, controls, etc. Various data are generated in the operation and maintenance activities of the railway system. These data are utilized in cooperation with safety and maintenance activities in each field, but amount of data is insufficient for data analysis of safety management due to relevant data being produced without any synchronous criteria such as time or space. In particular, reference data such as location and time of failure data for each field are set to different criteria according to the work characteristics in each field. So, it is not easy to analyze data integrally based on location and time. Therefore, mapping of reference data can be required for integrated analysis of data defined in different formats. By selecting data mapping tools and verifying the results of safety relevant data with the same criteria, the purpose of this paper is to enable integrated analysis of railway safety management data occurring in different fields based on location and time.

철도는 인터페이스 산업으로 철도차량 선로 관제 등이 유기적으로 맞물려야 움직일 수 있다. 이를 정상적으로 운영하기 위해서는 많은 유지보수 활동이 필요한데 이러한 활동과정에서 발생하는 데이터들은 각 분야별로 관리하고 있어 업무특성에 따라 위치, 시간 등 기준 데이터가 각각 상이하여 시 공간 기반의 통합 분석을 할 수 없는 실정이다. 각 분야별로 상이한 데이터를 통합 분석하기 위해서는 데이터마이닝 기법을 통해 유의미한 정보를 추출하여 분석하는데, 이러한 분석을 위해서는 데이터 유형의 상호관련성을 인지하기 위한 기준정보의 추출과 매핑이 반드시 필요하다. 본 연구는 서로 다른 분야에서 발생하는 안전과 관련된 데이터에 대한 매핑 툴 선정과 결과에 대하여 검증을 수행하고자 한다.

Keywords

References

  1. N. Attoh-Okine (2014) Big data challenges in railway engineering, IEEE International Coference on Big Data, Washington DC, pp. 7-9.
  2. H. Devid, K. Jason (2013) Big data and railroad analytics, Newsletter of the Railway Applications Section, p. 12.
  3. Y. Ko, T. Seo (2005) A Study on the naming Rules of metadata based on ontology, Journal of the Korean Society for Information Management, 22(4), pp. 97-109. https://doi.org/10.3743/KOSIM.2005.22.4.097
  4. T. Seo, D. T. Pham.(2007) Data modeling process to ensure semantic interoperability of data, Information Management Research, 37(1), pp. 59-73.
  5. H. Kim, Y. Jeong, H. Kang, D. Park (2016) Data mapping and development guideline for PHR system establishment using CMD, Journal of Korea Institute of Information Technology, pp. 133-141.
  6. C. Pluempitiwiriyawej, J. Hammer (2000), A classification scheme for semantic and schematic heterogeneities in XML data sources, Technical Report TR00-004, University of Florida, Gainesville, FL, pp. 36.
  7. J. Han, M. Kamber (2011) Data Mining: Concepts and Techniques (Third Edition), Burlington, MA: Morgan Kaufmann.
  8. U. Fayyad, G. Piatetsky-shapino, P. Smyth (1996) From data mining to knowledge discovery in databases, AI magazine, 17(3), pp. 37- 54.
  9. S.D. Nelson (2016) Informatics and interoperability: Speaking the same language, ISPOR 20th Aunnual International Meeting.
  10. Korea Data Agency (2016) Data Quality Management Guidelines ver 2.1.