Developing ANFIS-FMEA model for assessment and prioritization of potential trouble factors in Iraqi building projects

Abstract

The “Failure Mode and Effects Analysis” (FMEA) is a widely recognized and practical method in reliability engineering. Its primary objective is to investigate, identify, and prioritize potential failure modes in a given system, design, operation, or service. However, it possesses several limitations. The present research employs a novel hybrid assessment model that combines the conventional FMEA with the “Adaptive Neuro-Fuzzy Inference System” (ANFIS) technique and “Fuzzy Analytic Hierarchy Process” (FAHP) method. The aim is to overcome the limitations of conventional FMEA and establish a system-based decision model that can serve as an efficient tool for evaluating and ranking potential trouble factors in construction projects, identifying suitable responses promptly and assisting the project management team in recovering troubled projects. The model proposed utilized the amalgamated impact (I T) of cost and time, both of which are gauged via fuzzy AHP, alongside the Probability of Occurrence (P) and the degree of detection and control for potential trouble factors (DCD), as the basis for decision-making criteria, This (decision) is achieved by computing a trouble priority number. The hybrid approach was utilized to assess trouble factors in troubled construction projects. The findings demonstrate that the FMEA-ANFIS-FAHP model can provide a better assessment of trouble factors in construction projects experiencing difficulties. The hybrid model is a viable alternative to the conventional FMEA approach, as it mitigates various limitations and weaknesses inherent in the latter. Additionally, the hybrid model circumvents the need for many experts and expedites the assessment process.

Keywords

• neural-fuzzy (anfis)
• fuzzy analytical hierarchy process (fahp)
• failure mode and effects analysis (fmea)
• troubled projects