Influenza remains a global health concern due to its potential to cause pandemics as a result of rapidly mutating influenza virus strains. Existing vaccines often struggle to keep up with these rapidly mutating flu viruses. Therefore, the development of a broad-spectrum peptide vaccine that can stimulate an optimal antibody response has emerged as an innovative approach to addressing the influenza threat. In this study, an immunoinformatic approach was employed to rapidly predict immunodominant epitopes from different antigens, aiming to develop an effective multiepitope influenza vaccine (MEV). The immunodominant B-cell linear epitopes of seasonal influenza strains hemagglutinin (HA) and neuraminidase (NA) were predicted using an antibody-peptide microarray, involving a human cohort including vaccinees and infected patients. On the other hand, bioinformatics tools were used to predict immunodominant cytotoxic T-cell (CTL) and helper T-cell (HTL) epitopes. Subsequently, these epitopes were evaluated by various immunoinformatic tools. Epitopes with high antigenicity, high immunogenicity, non-allergenicity, non-toxicity, as well as exemplary conservation were then connected in series with appropriate linkers and adjuvants to construct a broad-spectrum MEV. Moreover, the structural analysis revealed that the MEV candidates exhibited good stability, and the docking results demonstrated their strong affinity to Toll-like receptors 4 (TLR4). In addition, molecular dynamics simulation confirmed the stable interaction between TLR4 and MEVs. Three injections with MEVs showed a high level of B-cell and T-cell immune responses according to the immunological simulations in silico. Furthermore, in-silico cloning was performed, and the results indicated that the MEVs could be produced in considerable quantities in Escherichia coli (E. coli). Based on these findings, it is reasonable to create a broad-spectrum MEV against different subtypes of influenza A and B viruses in silico.