To provide theoretical reference for owners to identify unbalanced bids, this paper aims to construct an identification method based on grey relational and fuzzy set theory. Firstly, to measure the closeness degree between bidding unit price from engineering's estimated price, grey relational analysis theory is used to express the relationship between them. Secondly, a combined weight method determining all line items is calculated through integrating analytic hierarchy model and maximizing deviation method. Thirdly, based on fuzzy set theory, the membership degree and the fuzzy relation matrix are constructed, and then a fuzzy comprehensive identification method is established to identify unbalanced bidding. Fourthly, on the basis of fuzzy comprehensive identification method, the scoring set and total score vector are designed, and the rank of unbalanced bids is obtained by total score vector. Finally, a practical construction project bidding is stated to illustrate the effectiveness and practicability of the proposed method. Key words: project management, unbalanced bid, identification, grey-fuzzy evaluation method. Afin de fournir une reference theorique aux proprietaries pour determiner les soumissions inegales, le present document vise a construire une methode d'identification basee sur la theorie grise rationnelle et la theorie des ensembles flous. Premierement, dans le but de mesurer le degre de proximite entre le prix unitaire de la soumission et le prix estimatif de l'ingenierie, on utilise la theorie grise de l'analyse rationnelle pour exprimer la relation entre les deux. Deuxiemement, une methode de ponderation combinee pour determiner tous les postes est calculee en integrant un modele de hierarchie analytique et la methode d'ecart superieur. Troisiemement, sur la base de la theorie des ensembles flous, le degre d'appartenance et la matrice des relations floues sont construits, puis une methode d'identification globale floue est etablie pour determiner les soumissions inegales. Quatriemement, sur la base de la methode d'identification complete floue, l'ensemble des cotes et le vecteur des cotes globales sont concus, et le rang des soumissions inegales est obtenu par le vecteur des cotes globales. Enfin, une soumission pratique pour un projet de construction est presentee pour illustrer l'efficacite et la faisabilite de la methode proposee. Mots-cles: gestion de projet, soumission inegale, identification, methode devaluation grise-floue., Introduction Competitive bidding is one of the most common approaches for owners to select contractors in the construction industry (Takano et al. 2018). In the unit price contract, the measurement [...]