Title: Application of association rule algorithm to industrial safety data mining

Authors: Jirachai Buddhakulsomsiri; Warut Pannakkong; Suebsak Nanthavanij

Addresses: School of Manufacturing Systems and Mechanical Engineering, Sirindhorn International Institute of Technology, Thammasat University, KlongLuang, Pathum Thani, 12121, Thailand ' School of Manufacturing Systems and Mechanical Engineering, Sirindhorn International Institute of Technology, Thammasat University, KlongLuang, Pathum Thani, 12121, Thailand ' School of Management Technology, Sirindhorn International Institute of Technology, Thammasat University, 131 Moo 5, Tiwanont Road, Bangkadi, Muang, Pathum Thani, 12000, Thailand

Abstract: This article presents an association rule-generation algorithm for mining industrial safety data. Examples of accident data are information about injured workers (e.g., age, gender, work experience), date and time of accidents, and severity level of accidents. The proposed algorithm implements the elementary set concept to generate useful relationships between accidents and worker-related attributes and severity of accidents. The relationships are presented as IF-THEN association rules, where the IF statement(s) include a set of accident condition attributes and the THEN statement(s) include attributes that represent a decision outcome (i.e., accident severity). After the rules are generated, the algorithm applies a user-specified filter and a statistical significance test to identify important rules. The rules that pass the filter and the significance test are then reported in the solution. A case study of using the algorithm to mine real safety data obtained from a selected industry is presented, along with examples of reported rules and their interpretations.

Keywords: industrial safety data; association rules; data mining; accident data; workplace accidents; worker-related attributes; accident severity; rule generation; case study.

DOI: 10.1504/IJISE.2015.072728

International Journal of Industrial and Systems Engineering, 2015 Vol.21 No.4, pp.415 - 437

Received: 03 Sep 2013
Accepted: 23 Mar 2014

Published online: 28 Oct 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article