Designing of dynamic labor inspection system for construction industry

摘要

Labor inspection services in many countries are not able to carry out their roles and functions. They are often understaffed, underequipped, undertrained and underpaid. In addition, workplaces vary enormously, especially in the construction industry. A labor inspectorate must build its policies and deploy its resources in accordance with the variations. Ranking workplaces from worst to best is perhaps the most straightforward way for managing labor inspectorates. This study employed 10-fold cross-validation in analyzing 715 fatal occupational injuries in the Taiwan construction industry. Accident reports during the period 1999–2008 were extracted from case reports of the Northern Region Inspection Office of the Council of Labor Affairs of Taiwan. Based on association rule mining, this article then involves 2 databases, including an occupational accidents database and a workplaces database, to present a dynamic labor inspection system for the construction industry. Besides the factors contributing to accidents, the factors about working groups, workplaces and working conditions have been considered in this article. Finally, the proposed system is applied to 2 cases to illustrate how it can effectively indicate potential hazardous working conditions and high-risk workplaces in practice. The system not only provides a support for labor inspection planning, but also makes inspection strategies more effective and flexible in changeable conditions surrounding the workplace.