Please use this identifier to cite or link to this item:
|Title:||A fuzzy knowledge-based system for premium rating of workers' compensation insurance for building projects|
Occupational health and safety
|Citation:||Imriyas, K.,Pheng, L.S.,Teo, E.A.-L. (2007). A fuzzy knowledge-based system for premium rating of workers' compensation insurance for building projects. Construction Management and Economics 25 (11) : 1175-1193. ScholarBank@NUS Repository. https://doi.org/10.1080/01446190701398462|
|Abstract:||Occupational injuries and fatalities are rampant in construction. The significance of the workers' compensation insurance (WCI) is immeasurable in safeguarding the interests of construction workers and contractors. From the insurers' perspective, the commitment under this insurance is extremely broad; there are no exclusions and a maximum limit on their liability. Thus, insurers must accomplish rigorous risk and market assessments to decide optimal premiums for construction projects. The conventional experience rating approach of premium rating has been proven ineffective for construction applications. Based on the findings of a literature review and an interview questionnaire survey, a new WCI premium rating model was developed for building projects. A hybrid of interviews and past workers' compensation claims data analysis was adopted to develop the conceptual model of a fuzzy knowledge-based system (KBS) to automate the proposed model. It was then prototyped, and verified with Turing tests. The proposed model and its fuzzy KBS advocate real time structured assessments of project hazards, safety, market condition and insurers' internal factors for premium rating. They also establish an effective risk control strategy via a well-structured incentive system for contractors and clients. Their implementation in the general insurance industry can facilitate accident control in the construction industry, thereby minimizing insurers' financial risks.|
|Source Title:||Construction Management and Economics|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Jan 19, 2019
checked on Jan 12, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.