Using Fuzzy Failure Mode Effect Analysis to Model Cave-In Accidents

Behavioral, managerial, and technical problems cause construction injury accidents. Among these accidents, cave-ins are some of the most frequently occurring accidents. Mitigating hazards that result in cave-ins are of great importance in managing and controlling excavation practices. This paper introduces a new conceptual risk mitigation tool that can be adopted to minimize cave-in hazards and be employed as a quality-control tool before and during excavations. It provides a list of hazards in ranked order from the most frequent to least frequent and most hazardous to least hazardous causes that managers should focus on and invest resources into. Precise risk evaluation for cave-ins using traditional methods of failure mode and effect analysis (FMEA) can be challenging, because the variables involved in cave-ins are unique, and historical data are often incomplete. In construction, estimation of risk is based on severity, frequency of occurrence, and likelihood of detection; management teams assess these factors prior to failure occurrences using subjective judgment expressed by linguistic values, such as severe and very likely. Such linguistic terms can be best modeled using the fuzzy set theory. The traditional FMEA method has been widely used to calculate a risk priority number (RPN), which is a weighted assessment used to prioritize risk items. The calculation of the RPN is based on the rating of three variables: (1) severity, (2) frequency of occurrence, and (3) likelihood of detection. The traditional RPN is the product of these variables. Implementation of the FMEA for construction projects must be modified, because calculation of the RPN in construction projects is based on managerial experience using experiential, subjective expressions. The RPN is substituted for fuzzy rating calculation using alpha cuts and fuzzy arithmetic, in order to perform the fuzzy operations of the variables and to introduce a workable model of a fuzzy risk priority number (FRPN). For assessment purposes, the FRPNs are ranked. A numerical example is provided to illustrate the potential application of the proposed fuzzy failure mode effect analysis (FFMEA) and the detailed computational process of the FRPN. An application of the FFMEA is presented for the assessment of construction trenching safety risk. The FFMEA method introduced is intended to be used for (1) anticipating potential cave-in accidents prior to their occurrence; (2) assessing risks associated with excavation work; (3) providing a list of likely cave-in accident scenarios associated with excavation work in advance; and (4) preparing solutions and taking corrective actions before cave-in accidents occur.