Abstract
The electrical construction industry faces persistent challenges in coordination, productivity, safety, and workforce development, and artificial intelligence (AI) is widely proposed as a remedy. Evidence specific to the electrical trade, however, remains fragmented across studies that treat electrical work as one component of broader mechanical, electrical, and plumbing (MEP) systems. This review synthesizes 129 peer-reviewed studies published between 2008 and 2026, retrieved from Scopus and Web of Science and coded across six dimensions: AI technique, digital technology, construction phase, electrical work category, maturity level, and industry involvement. Building information modelling dominates, appearing in approximately half of the studies, while large language models and generative AI remain scarce. Research concentrates on preconstruction coordination and closeout documentation, with limited attention to execution-phase productivity, code compliance automation, and cost estimation. Publication volume more than tripled after 2021 while the field-tested share fell from above 75% to between 29% and 45%, and quantified benefits are dominated by algorithm performance rather than by cost, safety, or productivity outcomes. The review proposes a taxonomy organized by the electrical contractor value chain and sets out six priority research directions, including LLM-assisted code compliance and AI-driven cost estimation.